Annotation of html5/spec/syntax.html, revision 1.246
1.2 mike 1: <!DOCTYPE HTML>
1.12 mike 2: <html lang="en-US-x-Hixie"><head><title>9 The HTML syntax — HTML 5</title><style type="text/css">
1.1 mike 3: pre { margin-left: 2em; white-space: pre-wrap; }
4: h2 { margin: 3em 0 1em 0; }
5: h3 { margin: 2.5em 0 1em 0; }
6: h4 { margin: 2.5em 0 0.75em 0; }
7: h5, h6 { margin: 2.5em 0 1em; }
8: h1 + h2, h1 + h2 + h2 { margin: 0.75em 0 0.75em; }
9: h2 + h3, h3 + h4, h4 + h5, h5 + h6 { margin-top: 0.5em; }
10: p { margin: 1em 0; }
11: hr:not(.top) { display: block; background: none; border: none; padding: 0; margin: 2em 0; height: auto; }
12: dl, dd { margin-top: 0; margin-bottom: 0; }
13: dt { margin-top: 0.75em; margin-bottom: 0.25em; clear: left; }
14: dt + dt { margin-top: 0; }
15: dd dt { margin-top: 0.25em; margin-bottom: 0; }
16: dd p { margin-top: 0; }
17: dd dl + p { margin-top: 1em; }
18: dd table + p { margin-top: 1em; }
19: p + * > li, dd li { margin: 1em 0; }
20: dt, dfn { font-weight: bold; font-style: normal; }
21: dt dfn { font-style: italic; }
22: pre, code { font-size: inherit; font-family: monospace; font-variant: normal; }
23: pre strong { color: black; font: inherit; font-weight: bold; background: yellow; }
24: pre em { font-weight: bolder; font-style: normal; }
25: @media screen { code { color: orangered; } code :link, code :visited { color: inherit; } }
26: var sub { vertical-align: bottom; font-size: smaller; position: relative; top: 0.1em; }
27: table { border-collapse: collapse; border-style: hidden hidden none hidden; }
28: table thead { border-bottom: solid; }
29: table tbody th:first-child { border-left: solid; }
30: table td, table th { border-left: solid; border-right: solid; border-bottom: solid thin; vertical-align: top; padding: 0.2em; }
31: blockquote { margin: 0 0 0 2em; border: 0; padding: 0; font-style: italic; }
32:
33: .bad, .bad *:not(.XXX) { color: gray; border-color: gray; background: transparent; }
34: .matrix, .matrix td { border: none; text-align: right; }
35: .matrix { margin-left: 2em; }
36: .dice-example { border-collapse: collapse; border-style: hidden solid solid hidden; border-width: thin; margin-left: 3em; }
37: .dice-example caption { width: 30em; font-size: smaller; font-style: italic; padding: 0.75em 0; text-align: left; }
38: .dice-example td, .dice-example th { border: solid thin; width: 1.35em; height: 1.05em; text-align: center; padding: 0; }
1.2 mike 39: .applies th > * { display: block; }
1.1 mike 40: .applies thead code { display: block; }
41: .applies td { text-align: center; }
42: .applies .yes { background: yellow; }
43:
44: .toc dfn, h1 dfn, h2 dfn, h3 dfn, h4 dfn, h5 dfn, h6 dfn { font: inherit; }
45: img.extra { float: right; }
46: pre.idl { border: solid thin; background: #EEEEEE; color: black; padding: 0.5em 1em; }
47: pre.idl :link, pre.idl :visited { color: inherit; background: transparent; }
48: pre.css { border: solid thin; background: #FFFFEE; color: black; padding: 0.5em 1em; }
49: pre.css:first-line { color: #AAAA50; }
50: dl.domintro { color: green; margin: 2em 0 2em 2em; padding: 0.5em 1em; border: none; background: #EEFFEE; }
51: hr + dl.domintro, div.impl + dl.domintro { margin-top: 2.5em; margin-bottom: 1.5em; }
52: dl.domintro dt, dl.domintro dt * { color: black; text-decoration: none; }
53: dl.domintro dd { margin: 0.5em 0 1em 2em; padding: 0; }
54: dl.domintro dd p { margin: 0.5em 0; }
55: dl.switch { padding-left: 2em; }
56: dl.switch > dt { text-indent: -1.5em; }
57: dl.switch > dt:before { content: '\21AA'; padding: 0 0.5em 0 0; display: inline-block; width: 1em; text-align: right; line-height: 0.5em; }
58: dl.triple { padding: 0 0 0 1em; }
59: dl.triple dt, dl.triple dd { margin: 0; display: inline }
60: dl.triple dt:after { content: ':'; }
61: dl.triple dd:after { content: '\A'; white-space: pre; }
62: .diff-old { text-decoration: line-through; color: silver; background: transparent; }
63: .diff-chg, .diff-new { text-decoration: underline; color: green; background: transparent; }
64: a .diff-new { border-bottom: 1px blue solid; }
65:
66: h2 { page-break-before: always; }
67: h1, h2, h3, h4, h5, h6 { page-break-after: avoid; }
68: h1 + h2, hr + h2.no-toc { page-break-before: auto; }
69:
70: p > span:not([title=""]):not([class="XXX"]):not([class="impl"]), li > span:not([title=""]):not([class="XXX"]):not([class="impl"]) { border-bottom: solid #9999CC; }
71:
72: div.head { margin: 0 0 1em; padding: 1em 0 0 0; }
73: div.head p { margin: 0; }
74: div.head h1 { margin: 0; }
75: div.head .logo { float: right; margin: 0 1em; }
76: div.head .logo img { border: none } /* remove border from top image */
77: div.head dl { margin: 1em 0; }
78: p.copyright { font-size: x-small; font-style: oblique; margin: 0; }
79:
80: body > .toc > li { margin-top: 1em; margin-bottom: 1em; }
81: body > .toc.brief > li { margin-top: 0.35em; margin-bottom: 0.35em; }
82: body > .toc > li > * { margin-bottom: 0.5em; }
83: body > .toc > li > * > li > * { margin-bottom: 0.25em; }
84: .toc, .toc li { list-style: none; }
85:
86: .brief { margin-top: 1em; margin-bottom: 1em; line-height: 1.1; }
87: .brief li { margin: 0; padding: 0; }
88: .brief li p { margin: 0; padding: 0; }
89:
90: .category-list { margin-top: -0.75em; margin-bottom: 1em; line-height: 1.5; }
91: .category-list::before { content: '\21D2\A0'; font-size: 1.2em; font-weight: 900; }
92: .category-list li { display: inline; }
93: .category-list li:not(:last-child)::after { content: ', '; }
94: .category-list li > span, .category-list li > a { text-transform: lowercase; }
95: .category-list li * { text-transform: none; } /* don't affect <code> nested in <a> */
96:
97: .XXX { color: #E50000; background: white; border: solid red; padding: 0.5em; margin: 1em 0; }
98: .XXX > :first-child { margin-top: 0; }
99: p .XXX { line-height: 3em; }
1.159 mike 100: .annotation { border: solid thin black; background: #0C479D; color: white; position: relative; margin: 8px 0 20px 0; }
101: .annotation:before { position: absolute; left: 0; top: 0; width: 100%; height: 100%; margin: 6px -6px -6px 6px; background: #333333; z-index: -1; content: ''; }
102: .annotation :link, .annotation :visited { color: inherit; }
103: .annotation :link:hover, .annotation :visited:hover { background: transparent; }
104: .annotation span { border: none ! important; }
1.1 mike 105: .note { color: green; background: transparent; font-family: sans-serif; }
106: .warning { color: red; background: transparent; }
107: .note, .warning { font-weight: bolder; font-style: italic; }
108: p.note, div.note { padding: 0.5em 2em; }
109: span.note { padding: 0 2em; }
110: .note p:first-child, .warning p:first-child { margin-top: 0; }
111: .note p:last-child, .warning p:last-child { margin-bottom: 0; }
112: .warning:before { font-style: normal; }
113: p.note:before { content: 'Note: '; }
114: p.warning:before { content: '\26A0 Warning! '; }
115:
116: .bookkeeping:before { display: block; content: 'Bookkeeping details'; font-weight: bolder; font-style: italic; }
117: .bookkeeping { font-size: 0.8em; margin: 2em 0; }
118: .bookkeeping p { margin: 0.5em 2em; display: list-item; list-style: square; }
119:
120: h4 { position: relative; z-index: 3; }
121: h4 + .element, h4 + div + .element { margin-top: -2.5em; padding-top: 2em; }
122: .element {
123: background: #EEEEFF;
124: color: black;
125: margin: 0 0 1em 0.15em;
126: padding: 0 1em 0.25em 0.75em;
127: border-left: solid #9999FF 0.25em;
128: position: relative;
129: z-index: 1;
130: }
131: .element:before {
132: position: absolute;
133: z-index: 2;
134: top: 0;
135: left: -1.15em;
136: height: 2em;
137: width: 0.9em;
138: background: #EEEEFF;
139: content: ' ';
140: border-style: none none solid solid;
141: border-color: #9999FF;
142: border-width: 0.25em;
143: }
144:
1.149 mike 145: .example { display: block; color: #222222; background: #FCFCFC; border-left: double; margin-left: 2em; padding-left: 1em; }
146: td > .example:only-child { margin: 0 0 0 0.1em; }
1.1 mike 147:
148: .tall-and-narrow {
149: font-size: 0.6em;
150: column-width: 25em;
151: column-gap: 1em;
152: -moz-column-width: 25em;
153: -moz-column-gap: 1em;
154: -webkit-column-width: 25em;
155: -webkit-column-gap: 1em;
156: }
157:
158: ul.domTree, ul.domTree ul { padding: 0 0 0 1em; margin: 0; }
159: ul.domTree li { padding: 0; margin: 0; list-style: none; position: relative; }
160: ul.domTree li li { list-style: none; }
161: ul.domTree li:first-child::before { position: absolute; top: 0; height: 0.6em; left: -0.75em; width: 0.5em; border-style: none none solid solid; content: ''; border-width: 0.1em; }
162: ul.domTree li:not(:last-child)::after { position: absolute; top: 0; bottom: -0.6em; left: -0.75em; width: 0.5em; border-style: none none solid solid; content: ''; border-width: 0.1em; }
163: ul.domTree span { font-style: italic; font-family: serif; }
164: ul.domTree .t1 code { color: purple; font-weight: bold; }
165: ul.domTree .t2 { font-style: normal; font-family: monospace; }
166: ul.domTree .t2 .name { color: black; font-weight: bold; }
167: ul.domTree .t2 .value { color: blue; font-weight: normal; }
168: ul.domTree .t3 code, .domTree .t4 code, .domTree .t5 code { color: gray; }
169: ul.domTree .t7 code, .domTree .t8 code { color: green; }
170: ul.domTree .t10 code { color: teal; }
171:
1.235 mike 172: </style><link href="data:text/css," rel="stylesheet" title="Complete specification"><link href="data:text/css,.impl%20%7B%20display:%20none;%20%7D" rel="alternate stylesheet" title="Author documentation only"><link href="data:text/css,.impl%20%7B%20background:%20%23FFEEEE;%20%7D" rel="alternate stylesheet" title="Highlight implementation requirements"><link href="http://www.w3.org/StyleSheets/TR/W3C-ED" rel="stylesheet" type="text/css"><!-- ZZZ ED vs WD -->
1.2 mike 173: <script src="link-fixup.js"></script>
174: <link href="comms.html" title="8 Communication" rel="prev">
175: <link href="spec.html#contents" title="Table of contents" rel="index">
1.1 mike 176: <link href="named-character-references.html" title="9.6 Named character references" rel="next">
1.16 mike 177: </head><body onload="fixBrokenLink(); fixBrokenLink()"><div class="head">
1.1 mike 178: <p><a href="http://www.w3.org/"><img alt="W3C" height="48" src="http://www.w3.org/Icons/w3c_home" width="72"></a></p>
1.183 mike 179: <h1>HTML5</h1>
1.1 mike 180: <h2 class="no-num no-toc" id="a-vocabulary-and-associated-apis-for-html-and-xhtml">A vocabulary and associated APIs for HTML and XHTML</h2>
1.123 mike 181: </div><nav>
1.1 mike 182: <a href="comms.html">← 8 Communication</a> –
1.2 mike 183: <a href="spec.html#contents">Table of contents</a> –
1.1 mike 184: <a href="named-character-references.html">9.6 Named character references →</a>
1.246 ! mike 185: </nav><p>This is revision 1.2955.</p>
1.11 mike 186:
1.1 mike 187:
1.123 mike 188: <h2 id="syntax"><span class="secno">9 </span><dfn>The HTML syntax</dfn></h2><p class="note">This section only describes the rules for
1.83 mike 189: <code><a href="iana-considerations.html#text-html">text/html</a></code> resources. Rules for XML resources are
1.1 mike 190: discussed in the section below entitled "<a href="the-xhtml-syntax.html#the-xhtml-syntax">The XHTML
1.159 mike 191: syntax</a>".</p><h3 id="writing"><span class="secno">9.1 </span>Writing HTML documents</h3><p class="XXX annotation"><b>Status: </b><i>Working draft</i></p><div class="impl">
1.1 mike 192:
193: <p><i>This section only applies to documents, authoring tools, and
194: markup generators. In particular, it does not apply to conformance
195: checkers; conformance checkers must use the requirements given in
196: the next section ("parsing HTML documents").</i></p>
197:
1.123 mike 198: </div><p>Documents must consist of the following parts, in the given
199: order:</p><ol><li>Optionally, a single U+FEFF BYTE ORDER MARK (BOM) character.</li>
1.1 mike 200:
201: <li>Any number of <a href="#syntax-comments" title="syntax-comments">comments</a> and
202: <a href="infrastructure.html#space-character" title="space character">space characters</a>.</li>
203:
204: <li>A <a href="#syntax-doctype" title="syntax-doctype">DOCTYPE</a>.
205:
206: </li><li>Any number of <a href="#syntax-comments" title="syntax-comments">comments</a> and
207: <a href="infrastructure.html#space-character" title="space character">space characters</a>.</li>
208:
1.159 mike 209: <li>The root element, in the form of an <code><a href="semantics.html#the-html-element-0">html</a></code> <a href="#syntax-elements" title="syntax-elements">element</a>.</li>
1.1 mike 210:
211: <li>Any number of <a href="#syntax-comments" title="syntax-comments">comments</a> and
212: <a href="infrastructure.html#space-character" title="space character">space characters</a>.</li>
213:
214: </ol><p>The various types of content mentioned above are described in the
1.123 mike 215: next few sections.</p><p>In addition, there are some restrictions on how <a href="semantics.html#character-encoding-declaration" title="character encoding declaration">character encoding
1.1 mike 216: declarations</a> are to be serialized, as discussed in the
1.123 mike 217: section on that topic.</p><div class="note">
1.1 mike 218:
1.159 mike 219: <p>Space characters before the root <code><a href="semantics.html#the-html-element-0">html</a></code> element, and
220: space characters at the start of the <code><a href="semantics.html#the-html-element-0">html</a></code> element and
221: before the <code><a href="semantics.html#the-head-element-0">head</a></code> element, will be dropped when the
1.1 mike 222: document is parsed; space characters <em>after</em> the root
1.159 mike 223: <code><a href="semantics.html#the-html-element-0">html</a></code> element will be parsed as if they were at the end
224: of the <code><a href="semantics.html#the-body-element-0">body</a></code> element. Thus, space characters around the
1.1 mike 225: root element do not round-trip.</p>
226:
227: <p>It is suggested that newlines be inserted after the DOCTYPE,
228: after any comments that are before the root element, after the
1.159 mike 229: <code><a href="semantics.html#the-html-element-0">html</a></code> element's start tag (if it is not <a href="#syntax-tag-omission" title="syntax-tag-omission">omitted</a>), and after any comments
230: that are inside the <code><a href="semantics.html#the-html-element-0">html</a></code> element but before the
231: <code><a href="semantics.html#the-head-element-0">head</a></code> element.</p>
1.1 mike 232:
1.123 mike 233: </div><p>Many strings in the HTML syntax (e.g. the names of elements and
1.1 mike 234: their attributes) are case-insensitive, but only for characters in
235: the ranges U+0041 .. U+005A (LATIN CAPITAL LETTER A to LATIN CAPITAL
236: LETTER Z) and U+0061 .. U+007A (LATIN SMALL LETTER A to LATIN SMALL
237: LETTER Z). For convenience, in this section this is just referred to
1.168 mike 238: as "case-insensitive".</p><h4 id="the-doctype"><span class="secno">9.1.1 </span>The DOCTYPE</h4><p class="XXX annotation"><span><a href="http://www.w3.org/html/wg/tracker/issues/4">ISSUE-4</a> (html-versioning) blocks progress to Last Call</span></p><p>A <dfn id="syntax-doctype" title="syntax-doctype">DOCTYPE</dfn> is a mostly useless,
1.123 mike 239: but required, header.</p><p class="note">DOCTYPEs are required for legacy reasons. When
1.1 mike 240: omitted, browsers tend to use a different rendering mode that is
241: incompatible with some specifications. Including the DOCTYPE in a
242: document ensures that the browser makes a best-effort attempt at
1.123 mike 243: following the relevant specifications.</p><p>A DOCTYPE must consist of the following characters, in this
244: order:</p><ol class="brief"><li>A U+003C LESS-THAN SIGN (<code><</code>) character.</li>
1.1 mike 245: <li>A U+0021 EXCLAMATION MARK (<code>!</code>) character.</li>
246: <li>A string that is an <a href="infrastructure.html#ascii-case-insensitive">ASCII case-insensitive</a> match for the string "<code title="">DOCTYPE</code>".</li>
247: <li>One or more <a href="infrastructure.html#space-character" title="space character">space characters</a>.</li>
248: <li>A string that is an <a href="infrastructure.html#ascii-case-insensitive">ASCII case-insensitive</a> match for the string "<code title="">HTML</code>".</li>
249: <li>Optionally, a <a href="#doctype-legacy-string">DOCTYPE legacy string</a> (defined below).</li>
250: <li>Zero or more <a href="infrastructure.html#space-character" title="space character">space characters</a>.</li>
251: <li>A U+003E GREATER-THAN SIGN (<code>></code>) character.</li>
252: </ol><p class="note">In other words, <code><!DOCTYPE HTML></code>,
1.123 mike 253: case-insensitively.</p><p>For the purposes of HTML generators that cannot output HTML
1.1 mike 254: markup with the short DOCTYPE "<code title=""><!DOCTYPE
255: HTML></code>", a <dfn id="doctype-legacy-string">DOCTYPE legacy string</dfn> may be inserted
256: into the DOCTYPE (in the position defined above). This string must
1.123 mike 257: consist of:</p><ol class="brief"><li>One or more <a href="infrastructure.html#space-character" title="space character">space characters</a>.</li>
1.1 mike 258: <li>A string that is an <a href="infrastructure.html#ascii-case-insensitive">ASCII case-insensitive</a> match for the string "<code title="">SYSTEM</code>".</li>
259: <li>One or more <a href="infrastructure.html#space-character" title="space character">space characters</a>.</li>
260: <li>A U+0022 QUOTATION MARK or U+0027 APOSTROPHE character (the <i>quote mark</i>).</li>
261: <li>The literal string "<code title="">about:legacy-compat</code>".</li>
262: <li>A matching U+0022 QUOTATION MARK or U+0027 APOSTROPHE character (i.e. the same character as in the earlier step marked <i>quote mark</i>).</li>
263: </ol><p class="note">In other words, <code><!DOCTYPE HTML SYSTEM
264: "about:legacy-compat"></code> or <code><!DOCTYPE HTML SYSTEM
265: 'about:legacy-compat'></code>, case-insensitively except for the bit
1.123 mike 266: in quotes.</p><p>The <a href="#doctype-legacy-string">DOCTYPE legacy string</a> should not be used unless
1.1 mike 267: the document is generated from a system that cannot output the
1.123 mike 268: shorter string.</p><h4 id="elements-0"><span class="secno">9.1.2 </span>Elements</h4><p>There are five different kinds of <dfn id="syntax-elements" title="syntax-elements">elements</dfn>: void elements, raw text
1.1 mike 269: elements, RCDATA elements, foreign elements, and normal
1.123 mike 270: elements.</p><dl><dt><dfn id="void-elements">Void elements</dfn></dt>
1.1 mike 271:
272: <dd><code><a href="the-canvas-element.html#the-area-element">area</a></code>, <code><a href="semantics.html#the-base-element">base</a></code>, <code><a href="semantics.html#the-br-element">br</a></code>,
1.159 mike 273: <code><a href="tabular-data.html#the-col-element">col</a></code>, <code><a href="interactive-elements.html#the-command">command</a></code>, <code><a href="text-level-semantics.html#the-embed-element">embed</a></code>,
274: <code><a href="semantics.html#the-hr-element">hr</a></code>, <code><a href="text-level-semantics.html#the-img-element">img</a></code>, <code><a href="forms.html#the-input-element">input</a></code>,
1.1 mike 275: <code><a href="forms.html#the-keygen-element">keygen</a></code>, <code><a href="semantics.html#the-link-element">link</a></code>, <code><a href="semantics.html#meta">meta</a></code>,
1.159 mike 276: <code><a href="text-level-semantics.html#the-param-element">param</a></code>, <code><a href="video.html#the-source-element">source</a></code></dd>
1.1 mike 277:
1.91 mike 278: <dt>Raw text elements</dt>
1.1 mike 279:
280: <dd><code><a href="semantics.html#script">script</a></code>, <code><a href="semantics.html#the-style-element">style</a></code></dd> <!-- iframe and
1.91 mike 281: noscript don't count as raw text for syntax purposes -->
1.1 mike 282:
283: <dt>RCDATA elements</dt>
284:
1.159 mike 285: <dd><code><a href="forms.html#the-textarea-element">textarea</a></code>, <code><a href="semantics.html#the-title-element-0">title</a></code></dd>
1.1 mike 286:
287: <dt>Foreign elements</dt>
288:
289: <dd>Elements from the <a href="#mathml-namespace">MathML namespace</a>
290: and the <a href="#svg-namespace">SVG namespace</a>.</dd>
291:
292: <dt>Normal elements</dt>
293:
294: <dd>All other allowed <a href="infrastructure.html#html-elements">HTML elements</a> are normal
295: elements.</dd>
296:
297: </dl><p><dfn id="syntax-tags" title="syntax-tags">Tags</dfn> are used to delimit the start
1.91 mike 298: and end of elements in the markup. Raw text, RCDATA, and normal
1.1 mike 299: elements have a <a href="#syntax-start-tag" title="syntax-start-tag">start tag</a> to
300: indicate where they begin, and an <a href="#syntax-end-tag" title="syntax-end-tag">end
301: tag</a> to indicate where they end. The start and end tags of
302: certain normal elements can be <a href="#syntax-tag-omission" title="syntax-tag-omission">omitted</a>, as described
303: later. Those that cannot be omitted must not be omitted. <a href="#void-elements">Void
304: elements</a> only have a start tag; end tags must not be
305: specified for void elements. Foreign elements must either have a
306: start tag and an end tag, or a start tag that is marked as
1.123 mike 307: self-closing, in which case they must not have an end tag.</p><p>The contents of the element must be placed between just after the
1.1 mike 308: start tag (which <a href="#syntax-tag-omission" title="syntax-tag-omission">might be implied,
309: in certain cases</a>) and just before the end tag (which again,
310: <a href="#syntax-tag-omission" title="syntax-tag-omission">might be implied in certain
311: cases</a>). The exact allowed contents of each individual element
312: depends on the content model of that element, as described earlier
313: in this specification. Elements must not contain content that their
314: content model disallows. In addition to the restrictions placed on
315: the contents by those content models, however, the five types of
1.123 mike 316: elements have additional <em>syntactic</em> requirements.</p><p><a href="#void-elements">Void elements</a> can't have any contents (since there's
1.1 mike 317: no end tag, no content can be put between the start tag and the end
1.123 mike 318: tag).</p><p>Raw text elements can have <a href="#syntax-text" title="syntax-text">text</a>,
1.1 mike 319: though it has <a href="#cdata-rcdata-restrictions">restrictions</a>
1.123 mike 320: described below.</p><p>RCDATA elements can have <a href="#syntax-text" title="syntax-text">text</a>
1.1 mike 321: and <a href="#syntax-charref" title="syntax-charref">character references</a>, but
322: the text must not contain an <a href="#syntax-ambiguous-ampersand" title="syntax-ambiguous-ampersand">ambiguous ampersand</a>.
323: There are also <a href="#cdata-rcdata-restrictions">further
1.123 mike 324: restrictions</a> described below.</p><p>Foreign elements whose start tag is marked as self-closing can't
1.1 mike 325: have any contents (since, again, as there's no end tag, no content
326: can be put between the start tag and the end tag). Foreign elements
327: whose start tag is <em>not</em> marked as self-closing can have
328: <a href="#syntax-text" title="syntax-text">text</a>, <a href="#syntax-charref" title="syntax-charref">character references</a>, <a href="#syntax-cdata" title="syntax-cdata">CDATA sections</a>, other <a href="#syntax-elements" title="syntax-elements">elements</a>, and <a href="#syntax-comments" title="syntax-comments">comments</a>, but the text must not
329: contain the character U+003C LESS-THAN SIGN (<code><</code>) or
330: an <a href="#syntax-ambiguous-ampersand" title="syntax-ambiguous-ampersand">ambiguous
1.123 mike 331: ampersand</a>.</p><p>Normal elements can have <a href="#syntax-text" title="syntax-text">text</a>,
1.1 mike 332: <a href="#syntax-charref" title="syntax-charref">character references</a>, other
333: <a href="#syntax-elements" title="syntax-elements">elements</a>, and <a href="#syntax-comments" title="syntax-comments">comments</a>, but the text must not
334: contain the character U+003C LESS-THAN SIGN (<code><</code>) or
335: an <a href="#syntax-ambiguous-ampersand" title="syntax-ambiguous-ampersand">ambiguous
336: ampersand</a>. Some normal elements also have <a href="#element-restrictions">yet more restrictions</a> on what
337: content they are allowed to hold, beyond the restrictions imposed by
338: the content model and those described in this paragraph. Those
1.123 mike 339: restrictions are described below.</p><p>Tags contain a <dfn id="syntax-tag-name" title="syntax-tag-name">tag name</dfn>,
1.1 mike 340: giving the element's name. HTML elements all have names that only
341: use characters in the range U+0030 DIGIT ZERO .. U+0039 DIGIT NINE,
1.202 mike 342: U+0061 LATIN SMALL LETTER A .. U+007A LATIN SMALL LETTER Z, and
343: U+0041 LATIN CAPITAL LETTER A .. U+005A LATIN CAPITAL LETTER Z. In
344: the HTML syntax, tag names may be written with any mix of lower- and
345: uppercase letters that, when converted to all-lowercase, matches the
346: element's tag name; tag names are case-insensitive.</p><h5 id="start-tags"><span class="secno">9.1.2.1 </span>Start tags</h5><p><dfn id="syntax-start-tag" title="syntax-start-tag">Start tags</dfn> must have the
1.123 mike 347: following format:</p><ol><li>The first character of a start tag must be a U+003C LESS-THAN
1.1 mike 348: SIGN (<code><</code>).</li>
349:
350: <li>The next few characters of a start tag must be the element's
351: <a href="#syntax-tag-name" title="syntax-tag-name">tag name</a>.</li>
352:
353: <li>If there are to be any attributes in the next step, there must
354: first be one or more <a href="infrastructure.html#space-character" title="space character">space
355: characters</a>.</li>
356:
357: <li>Then, the start tag may have a number of attributes, the <a href="#syntax-attributes" title="syntax-attributes">syntax for which</a> is described
358: below. Attributes may be separated from each other by one or more
359: <a href="infrastructure.html#space-character" title="space character">space characters</a>.</li>
360:
361: <li>After the attributes, there may be one or more <a href="infrastructure.html#space-character" title="space character">space characters</a>. (Some attributes
362: are required to be followed by a space. See the <a href="#syntax-attributes" title="syntax-attributes">attributes section</a> below.)</li>
363:
364: <li>Then, if the element is one of the <a href="#void-elements">void elements</a>,
365: or if the element is a foreign element, then there may be a single
366: U+002F SOLIDUS (<code>/</code>) character. This character has no
367: effect on void elements, but on foreign elements it marks the start
368: tag as self-closing.</li>
369:
370: <li>Finally, start tags must be closed by a U+003E GREATER-THAN
371: SIGN (<code>></code>) character.</li>
372:
1.123 mike 373: </ol><h5 id="end-tags"><span class="secno">9.1.2.2 </span>End tags</h5><p><dfn id="syntax-end-tag" title="syntax-end-tag">End tags</dfn> must have the
374: following format:</p><ol><li>The first character of an end tag must be a U+003C LESS-THAN
1.1 mike 375: SIGN (<code><</code>).</li>
376:
377: <li>The second character of an end tag must be a U+002F SOLIDUS
378: (<code>/</code>).</li>
379:
380: <li>The next few characters of an end tag must be the element's
381: <a href="#syntax-tag-name" title="syntax-tag-name">tag name</a>.</li>
382:
383: <li>After the tag name, there may be one or more <a href="infrastructure.html#space-character" title="space
384: character">space characters</a>.</li>
385:
386: <li>Finally, end tags must be closed by a U+003E GREATER-THAN
387: SIGN (<code>></code>) character.</li>
388:
1.123 mike 389: </ol><h5 id="attributes"><span class="secno">9.1.2.3 </span>Attributes</h5><p><dfn id="syntax-attributes" title="syntax-attributes">Attributes</dfn> for an element
390: are expressed inside the element's start tag.</p><p>Attributes have a name and a value. <dfn id="syntax-attribute-name" title="syntax-attribute-name">Attribute names</dfn> must consist of
1.1 mike 391: one or more characters other than the <a href="infrastructure.html#space-character" title="space
392: character">space characters</a>, U+0000 NULL, U+0022 QUOTATION
393: MARK ("), U+0027 APOSTROPHE ('), U+003E GREATER-THAN SIGN
394: (>), U+002F SOLIDUS (/), and U+003D EQUALS SIGN (=) characters,
395: the control characters, and any characters that are not defined by
396: Unicode. In the HTML syntax, attribute names may be written with any
397: mix of lower- and uppercase letters that are an <a href="infrastructure.html#ascii-case-insensitive">ASCII
1.123 mike 398: case-insensitive</a> match for the attribute's name.</p><p><dfn id="syntax-attribute-value" title="syntax-attribute-value">Attribute values</dfn> are a
1.1 mike 399: mixture of <a href="#syntax-text" title="syntax-text">text</a> and <a href="#syntax-charref" title="syntax-charref">character references</a>, except with the
1.123 mike 400: additional restriction that the text cannot contain an <a href="#syntax-ambiguous-ampersand" title="syntax-ambiguous-ampersand">ambiguous ampersand</a>.</p><p>Attributes can be specified in four different ways:</p><dl><dt>Empty attribute syntax</dt>
1.1 mike 401:
402: <dd>
403:
404: <p>Just the <a href="#syntax-attribute-name" title="syntax-attribute-name">attribute
405: name</a>.</p>
406:
407: <div class="example">
408:
409: <p>In the following example, the <code title="attr-fe-disabled"><a href="forms.html#attr-fe-disabled">disabled</a></code> attribute is given with
410: the empty attribute syntax:</p>
411:
412: <pre><input <em>disabled</em>></pre>
413:
414: </div>
415:
416: <p>If an attribute using the empty attribute syntax is to be
417: followed by another attribute, then there must be a <a href="infrastructure.html#space-character">space
418: character</a> separating the two.</p>
419:
420: </dd>
421:
1.2 mike 422: <dt id="unquoted">Unquoted attribute value syntax</dt>
1.1 mike 423:
424: <dd>
425:
426: <p>The <a href="#syntax-attribute-name" title="syntax-attribute-name">attribute name</a>,
427: followed by zero or more <a href="infrastructure.html#space-character" title="space character">space
428: characters</a>, followed by a single U+003D EQUALS SIGN
429: character, followed by zero or more <a href="infrastructure.html#space-character" title="space
430: character">space characters</a>, followed by the <a href="#syntax-attribute-value" title="syntax-attribute-value">attribute value</a>, which, in
431: addition to the requirements given above for attribute values,
432: must not contain any literal <a href="infrastructure.html#space-character" title="space character">space
433: characters</a>, any U+0022 QUOTATION MARK (<code>"</code>)
434: characters, U+0027 APOSTROPHE (<code>'</code>) characters,
435: U+003D EQUALS SIGN (<code>=</code>) characters, U+003C LESS-THAN
436: SIGN (<code><</code>) characters, or U+003E GREATER-THAN SIGN
437: (<code>></code>) characters, and must not be the empty
438: string.</p>
439:
440: <div class="example">
441:
442: <p>In the following example, the <code title="attr-input-value"><a href="forms.html#attr-input-value">value</a></code> attribute is given
443: with the unquoted attribute value syntax:</p>
444:
445: <pre><input <em>value=yes</em>></pre>
446:
447: </div>
448:
449: <p>If an attribute using the unquoted attribute syntax is to be
1.101 mike 450: followed by another attribute or by the optional U+002F SOLIDUS
451: (<code>/</code>) character allowed in step 6 of the <a href="#syntax-start-tag" title="syntax-start-tag">start tag</a> syntax above, then there
1.1 mike 452: must be a <a href="infrastructure.html#space-character">space character</a> separating the two.</p>
453:
454: </dd>
455:
456: <dt>Single-quoted attribute value syntax</dt>
457:
458: <dd>
459:
460: <p>The <a href="#syntax-attribute-name" title="syntax-attribute-name">attribute name</a>,
461: followed by zero or more <a href="infrastructure.html#space-character" title="space character">space
462: characters</a>, followed by a single U+003D EQUALS SIGN
463: character, followed by zero or more <a href="infrastructure.html#space-character" title="space
464: character">space characters</a>, followed by a single U+0027
465: APOSTROPHE (<code>'</code>) character, followed by the <a href="#syntax-attribute-value" title="syntax-attribute-value">attribute value</a>, which, in
466: addition to the requirements given above for attribute values,
467: must not contain any literal U+0027 APOSTROPHE (<code>'</code>)
468: characters, and finally followed by a second single U+0027
469: APOSTROPHE (<code>'</code>) character.</p>
470:
471: <div class="example">
472:
473: <p>In the following example, the <code title="attr-input-type"><a href="forms.html#attr-input-type">type</a></code> attribute is given with the
474: single-quoted attribute value syntax:</p>
475:
476: <pre><input <em>type='checkbox'</em>></pre>
477:
478: </div>
479:
480: <p>If an attribute using the single-quoted attribute syntax is to
481: be followed by another attribute, then there must be a <a href="infrastructure.html#space-character">space
482: character</a> separating the two.</p>
483:
484: </dd>
485:
486: <dt>Double-quoted attribute value syntax</dt>
487:
488: <dd>
489:
490: <p>The <a href="#syntax-attribute-name" title="syntax-attribute-name">attribute name</a>,
491: followed by zero or more <a href="infrastructure.html#space-character" title="space character">space
492: characters</a>, followed by a single U+003D EQUALS SIGN
493: character, followed by zero or more <a href="infrastructure.html#space-character" title="space
494: character">space characters</a>, followed by a single U+0022
495: QUOTATION MARK (<code>"</code>) character, followed by the <a href="#syntax-attribute-value" title="syntax-attribute-value">attribute value</a>, which, in
496: addition to the requirements given above for attribute values,
497: must not contain any literal U+0022 QUOTATION MARK
498: (<code>"</code>) characters, and finally followed by a second
499: single U+0022 QUOTATION MARK (<code>"</code>) character.</p>
500:
501: <div class="example">
502:
503: <p>In the following example, the <code title="attr-fe-name"><a href="forms.html#attr-fe-name">name</a></code> attribute is given with the
504: double-quoted attribute value syntax:</p>
505:
506: <pre><input <em>name="be evil"</em>></pre>
507:
508: </div>
509:
510: <p>If an attribute using the double-quoted attribute syntax is to
511: be followed by another attribute, then there must be a <a href="infrastructure.html#space-character">space
512: character</a> separating the two.</p>
513:
514: </dd>
515:
516: </dl><p>There must never be two or more attributes on the same start tag
517: whose names are an <a href="infrastructure.html#ascii-case-insensitive">ASCII case-insensitive</a> match for
1.123 mike 518: each other.</p><h5 id="optional-tags"><span class="secno">9.1.2.4 </span>Optional tags</h5><p>Certain tags can be <dfn id="syntax-tag-omission" title="syntax-tag-omission">omitted</dfn>.</p><p class="note">Omitting an element's <a href="#syntax-start-tag" title="syntax-start-tag">start tag</a> does not mean the element
1.1 mike 519: is not present; it is implied, but it is still there. An HTML
1.159 mike 520: document always has a root <code><a href="semantics.html#the-html-element-0">html</a></code> element, even if the
1.1 mike 521: string <code title=""><html></code> doesn't appear anywhere in
1.159 mike 522: the markup.</p><!-- <html> --><p>An <code><a href="semantics.html#the-html-element-0">html</a></code> element's <a href="#syntax-start-tag" title="syntax-start-tag">start tag</a> may be omitted if the
523: first thing inside the <code><a href="semantics.html#the-html-element-0">html</a></code> element is not a <a href="#syntax-comments" title="syntax-comments">comment</a>.</p><!-- </html> --><p>An <code><a href="semantics.html#the-html-element-0">html</a></code> element's <a href="#syntax-end-tag" title="syntax-end-tag">end
524: tag</a> may be omitted if the <code><a href="semantics.html#the-html-element-0">html</a></code> element is not
525: immediately followed by a <a href="#syntax-comments" title="syntax-comments">comment</a>.</p><!-- <head> --><p>A <code><a href="semantics.html#the-head-element-0">head</a></code> element's <a href="#syntax-start-tag" title="syntax-start-tag">start tag</a> may be omitted <!--if the
1.1 mike 526: element is empty, or--> if the first thing inside the
1.159 mike 527: <code><a href="semantics.html#the-head-element-0">head</a></code> element is an element.</p><!-- </head> --><p>A <code><a href="semantics.html#the-head-element-0">head</a></code> element's <a href="#syntax-end-tag" title="syntax-end-tag">end
528: tag</a> may be omitted if the <code><a href="semantics.html#the-head-element-0">head</a></code> element is not
529: immediately followed by a <a href="infrastructure.html#space-character">space character</a> or a <a href="#syntax-comments" title="syntax-comments">comment</a>.</p><!-- <body> --><p>A <code><a href="semantics.html#the-body-element-0">body</a></code> element's <a href="#syntax-start-tag" title="syntax-start-tag">start tag</a> may be omitted if the
530: element is empty, or if the first thing inside the <code><a href="semantics.html#the-body-element-0">body</a></code>
1.1 mike 531: element is not a <a href="infrastructure.html#space-character">space character</a> or a <a href="#syntax-comments" title="syntax-comments">comment</a>, except if the first thing
1.159 mike 532: inside the <code><a href="semantics.html#the-body-element-0">body</a></code> element is a <code><a href="semantics.html#script">script</a></code> or
1.1 mike 533: <code><a href="semantics.html#the-style-element">style</a></code> element. <!-- Note that even if the </head> end
534: tag is present, the parser makes <style> and <script> elements
535: between </head> and <body> end up in the <head> instead of implying
1.159 mike 536: the <body> --></p><!-- </body> --><p>A <code><a href="semantics.html#the-body-element-0">body</a></code> element's <a href="#syntax-end-tag" title="syntax-end-tag">end
537: tag</a> may be omitted if the <code><a href="semantics.html#the-body-element-0">body</a></code> element is not
1.123 mike 538: immediately followed by a <a href="#syntax-comments" title="syntax-comments">comment</a>.</p><!-- </li> --><p>A <code><a href="semantics.html#the-li-element">li</a></code> element's <a href="#syntax-end-tag" title="syntax-end-tag">end
1.1 mike 539: tag</a> may be omitted if the <code><a href="semantics.html#the-li-element">li</a></code> element is
540: immediately followed by another <code><a href="semantics.html#the-li-element">li</a></code> element or if there
1.123 mike 541: is no more content in the parent element.</p><!-- </dt> --><p>A <code><a href="semantics.html#the-dt-element">dt</a></code> element's <a href="#syntax-end-tag" title="syntax-end-tag">end
1.1 mike 542: tag</a> may be omitted if the <code><a href="semantics.html#the-dt-element">dt</a></code> element is
543: immediately followed by another <code><a href="semantics.html#the-dt-element">dt</a></code> element or a
1.123 mike 544: <code><a href="semantics.html#the-dd-element">dd</a></code> element.</p><!-- </dd> --><p>A <code><a href="semantics.html#the-dd-element">dd</a></code> element's <a href="#syntax-end-tag" title="syntax-end-tag">end
1.1 mike 545: tag</a> may be omitted if the <code><a href="semantics.html#the-dd-element">dd</a></code> element is
546: immediately followed by another <code><a href="semantics.html#the-dd-element">dd</a></code> element or a
547: <code><a href="semantics.html#the-dt-element">dt</a></code> element, or if there is no more content in the
1.123 mike 548: parent element.</p><!-- </p> --><p>A <code><a href="semantics.html#the-p-element">p</a></code> element's <a href="#syntax-end-tag" title="syntax-end-tag">end
1.1 mike 549: tag</a> may be omitted if the <code><a href="semantics.html#the-p-element">p</a></code> element is
550: immediately followed by an <code><a href="semantics.html#the-address-element">address</a></code>,
551: <code><a href="semantics.html#the-article-element">article</a></code>, <code><a href="semantics.html#the-aside-element">aside</a></code>, <code><a href="semantics.html#the-blockquote-element">blockquote</a></code>,
1.86 mike 552: <!--v2DATAGRID <code>datagrid</code>,--> <code><a href="semantics.html#the-dialog-element">dialog</a></code>,
553: <code><a href="obsolete.html#dir">dir</a></code>, <code><a href="interactive-elements.html#the-div-element">div</a></code>, <code><a href="semantics.html#the-dl-element">dl</a></code>,
554: <code><a href="forms.html#the-fieldset-element">fieldset</a></code>, <code><a href="semantics.html#the-footer-element">footer</a></code>, <code><a href="forms.html#the-form-element">form</a></code>,
555: <code><a href="semantics.html#the-h1-h2-h3-h4-h5-and-h6-elements">h1</a></code>, <code><a href="semantics.html#the-h1-h2-h3-h4-h5-and-h6-elements">h2</a></code>, <code><a href="semantics.html#the-h1-h2-h3-h4-h5-and-h6-elements">h3</a></code>, <code><a href="semantics.html#the-h1-h2-h3-h4-h5-and-h6-elements">h4</a></code>,
556: <code><a href="semantics.html#the-h1-h2-h3-h4-h5-and-h6-elements">h5</a></code>, <code><a href="semantics.html#the-h1-h2-h3-h4-h5-and-h6-elements">h6</a></code>, <code><a href="semantics.html#the-header-element">header</a></code>,
557: <code><a href="semantics.html#the-hgroup-element">hgroup</a></code>, <code><a href="semantics.html#the-hr-element">hr</a></code>, <code><a href="interactive-elements.html#menus">menu</a></code>,
558: <code><a href="semantics.html#the-nav-element">nav</a></code>, <code><a href="semantics.html#the-ol-element">ol</a></code>, <code><a href="semantics.html#the-p-element">p</a></code>, <code><a href="semantics.html#the-pre-element">pre</a></code>,
1.1 mike 559: <code><a href="semantics.html#the-section-element">section</a></code>, <code><a href="tabular-data.html#the-table-element">table</a></code>, or <code><a href="semantics.html#the-ul-element">ul</a></code>,
560: element, or if there is no more content in the parent element and
1.123 mike 561: the parent element is not an <code><a href="text-level-semantics.html#the-a-element">a</a></code> element.</p><!-- </rt> --><p>An <code><a href="text-level-semantics.html#the-rt-element">rt</a></code> element's <a href="#syntax-end-tag" title="syntax-end-tag">end
1.1 mike 562: tag</a> may be omitted if the <code><a href="text-level-semantics.html#the-rt-element">rt</a></code> element is
563: immediately followed by an <code><a href="text-level-semantics.html#the-rt-element">rt</a></code> or <code><a href="text-level-semantics.html#the-rp-element">rp</a></code>
1.123 mike 564: element, or if there is no more content in the parent element.</p><!-- </rp> --><p>An <code><a href="text-level-semantics.html#the-rp-element">rp</a></code> element's <a href="#syntax-end-tag" title="syntax-end-tag">end
1.1 mike 565: tag</a> may be omitted if the <code><a href="text-level-semantics.html#the-rp-element">rp</a></code> element is
566: immediately followed by an <code><a href="text-level-semantics.html#the-rt-element">rt</a></code> or <code><a href="text-level-semantics.html#the-rp-element">rp</a></code>
1.123 mike 567: element, or if there is no more content in the parent element.</p><!-- </optgroup> (the text assumes <optgroup> can only be inside a
1.1 mike 568: <select>; commented out text below can handle the non-<select> case
1.123 mike 569: if we ever allow it) --><p>An <code><a href="forms.html#the-optgroup-element">optgroup</a></code> element's <a href="#syntax-end-tag" title="syntax-end-tag">end tag</a> may be omitted if the
1.1 mike 570: <code><a href="forms.html#the-optgroup-element">optgroup</a></code> element <!--has an ancestor
571: <code>select</code> element and--> is immediately followed by
572: another <code><a href="forms.html#the-optgroup-element">optgroup</a></code> element, or if <!--all of the elements
573: that are ancestors of the <code>optgroup</code> element, up to and
574: including the first ancestor element that is not an
575: <code>optgroup</code> element, have no more content--> there is no
1.123 mike 576: more content in the parent element.</p><!-- so e.g. the max number of </optgroup>s are omitted here:
1.1 mike 577: <select><optgroup></select>
578: <p id=x><optgroup></optgroup>x</p>
579: <p id=x><optgroup><optgroup></optgroup></optgroup>x</p>
580: <p><optgroup id=x><optgroup></optgroup>x</p>
581: <p><optgroup><optgroup id=x>x</p>
1.123 mike 582: --><!-- </option> --><p>An <code><a href="forms.html#the-option-element">option</a></code> element's <a href="#syntax-end-tag" title="syntax-end-tag">end
1.1 mike 583: tag</a> may be omitted if the <code><a href="forms.html#the-option-element">option</a></code> element is
584: immediately followed by another <code><a href="forms.html#the-option-element">option</a></code> element, or if
585: it is immediately followed by an <code><a href="forms.html#the-optgroup-element">optgroup</a></code> element, or
1.123 mike 586: if there is no more content in the parent element.</p><!-- <colgroup> --><p>A <code><a href="tabular-data.html#the-colgroup-element">colgroup</a></code> element's <a href="#syntax-start-tag" title="syntax-start-tag">start tag</a> may be omitted if the
1.1 mike 587: first thing inside the <code><a href="tabular-data.html#the-colgroup-element">colgroup</a></code> element is a
588: <code><a href="tabular-data.html#the-col-element">col</a></code> element, and if the element is not immediately
589: preceded by another <code><a href="tabular-data.html#the-colgroup-element">colgroup</a></code> element whose <a href="#syntax-end-tag" title="syntax-end-tag">end tag</a> has been omitted. (It can't be
1.123 mike 590: omitted if the element is empty.)</p><!-- </colgroup> --><p>A <code><a href="tabular-data.html#the-colgroup-element">colgroup</a></code> element's <a href="#syntax-end-tag" title="syntax-end-tag">end tag</a> may be omitted if the
1.1 mike 591: <code><a href="tabular-data.html#the-colgroup-element">colgroup</a></code> element is not immediately followed by a
1.123 mike 592: <a href="infrastructure.html#space-character">space character</a> or a <a href="#syntax-comments" title="syntax-comments">comment</a>.</p><!-- </thead> --><p>A <code><a href="tabular-data.html#the-thead-element">thead</a></code> element's <a href="#syntax-end-tag" title="syntax-end-tag">end
1.1 mike 593: tag</a> may be omitted if the <code><a href="tabular-data.html#the-thead-element">thead</a></code> element is
594: immediately followed by a <code><a href="tabular-data.html#the-tbody-element">tbody</a></code> or <code><a href="tabular-data.html#the-tfoot-element">tfoot</a></code>
1.123 mike 595: element.</p><!-- <tbody> --><p>A <code><a href="tabular-data.html#the-tbody-element">tbody</a></code> element's <a href="#syntax-start-tag" title="syntax-start-tag">start tag</a> may be omitted if the
1.1 mike 596: first thing inside the <code><a href="tabular-data.html#the-tbody-element">tbody</a></code> element is a
597: <code><a href="tabular-data.html#the-tr-element">tr</a></code> element, and if the element is not immediately
598: preceded by a <code><a href="tabular-data.html#the-tbody-element">tbody</a></code>, <code><a href="tabular-data.html#the-thead-element">thead</a></code>, or
599: <code><a href="tabular-data.html#the-tfoot-element">tfoot</a></code> element whose <a href="#syntax-end-tag" title="syntax-end-tag">end
600: tag</a> has been omitted. (It can't be omitted if the element is
1.123 mike 601: empty.)</p><!-- </tbody> --><p>A <code><a href="tabular-data.html#the-tbody-element">tbody</a></code> element's <a href="#syntax-end-tag" title="syntax-end-tag">end
1.1 mike 602: tag</a> may be omitted if the <code><a href="tabular-data.html#the-tbody-element">tbody</a></code> element is
603: immediately followed by a <code><a href="tabular-data.html#the-tbody-element">tbody</a></code> or <code><a href="tabular-data.html#the-tfoot-element">tfoot</a></code>
1.123 mike 604: element, or if there is no more content in the parent element.</p><!-- </tfoot> --><p>A <code><a href="tabular-data.html#the-tfoot-element">tfoot</a></code> element's <a href="#syntax-end-tag" title="syntax-end-tag">end
1.1 mike 605: tag</a> may be omitted if the <code><a href="tabular-data.html#the-tfoot-element">tfoot</a></code> element is
606: immediately followed by a <code><a href="tabular-data.html#the-tbody-element">tbody</a></code> element, or if there is
1.123 mike 607: no more content in the parent element.</p><!-- </tr> --><p>A <code><a href="tabular-data.html#the-tr-element">tr</a></code> element's <a href="#syntax-end-tag" title="syntax-end-tag">end
1.1 mike 608: tag</a> may be omitted if the <code><a href="tabular-data.html#the-tr-element">tr</a></code> element is
609: immediately followed by another <code><a href="tabular-data.html#the-tr-element">tr</a></code> element, or if there
1.123 mike 610: is no more content in the parent element.</p><!-- </td> --><p>A <code><a href="tabular-data.html#the-td-element">td</a></code> element's <a href="#syntax-end-tag" title="syntax-end-tag">end
1.1 mike 611: tag</a> may be omitted if the <code><a href="tabular-data.html#the-td-element">td</a></code> element is
612: immediately followed by a <code><a href="tabular-data.html#the-td-element">td</a></code> or <code><a href="tabular-data.html#the-th-element">th</a></code>
1.123 mike 613: element, or if there is no more content in the parent element.</p><!-- </th> --><p>A <code><a href="tabular-data.html#the-th-element">th</a></code> element's <a href="#syntax-end-tag" title="syntax-end-tag">end
1.1 mike 614: tag</a> may be omitted if the <code><a href="tabular-data.html#the-th-element">th</a></code> element is
615: immediately followed by a <code><a href="tabular-data.html#the-td-element">td</a></code> or <code><a href="tabular-data.html#the-th-element">th</a></code>
1.123 mike 616: element, or if there is no more content in the parent element.</p><p><strong>However</strong>, a <a href="#syntax-start-tag" title="syntax-start-tag">start
617: tag</a> must never be omitted if it has any attributes.</p><h5 id="element-restrictions"><span class="secno">9.1.2.5 </span>Restrictions on content models</h5><p>For historical reasons, certain elements have extra restrictions
618: beyond even the restrictions given by their content model.</p><p>A <code><a href="tabular-data.html#the-table-element">table</a></code> element must not contain <code><a href="tabular-data.html#the-tr-element">tr</a></code>
1.1 mike 619: elements, even though these elements are technically allowed inside
620: <code><a href="tabular-data.html#the-table-element">table</a></code> elements according to the content models
621: described in this specification. (If a <code><a href="tabular-data.html#the-tr-element">tr</a></code> element is
622: put inside a <code><a href="tabular-data.html#the-table-element">table</a></code> in the markup, it will in fact imply
1.123 mike 623: a <code><a href="tabular-data.html#the-tbody-element">tbody</a></code> start tag before it.)</p><p>A single U+000A LINE FEED (LF) character may be placed
1.1 mike 624: immediately after the <a href="#syntax-start-tag" title="syntax-start-tag">start
625: tag</a> of <code><a href="semantics.html#the-pre-element">pre</a></code> and <code><a href="forms.html#the-textarea-element">textarea</a></code>
626: elements. This does not affect the processing of the element. The
627: otherwise optional U+000A LINE FEED (LF) character <em>must</em> be
628: included if the element's contents start with that character
629: (because otherwise the leading newline in the contents would be
1.123 mike 630: treated like the optional newline, and ignored).</p><div class="example">
1.1 mike 631: <p>The following two <code><a href="semantics.html#the-pre-element">pre</a></code> blocks are equivalent:</p>
632: <pre><pre>Hello</pre></pre>
633: <pre><pre><br>Hello</pre></pre>
1.123 mike 634: </div><h5 id="cdata-rcdata-restrictions"><span class="secno">9.1.2.6 </span>Restrictions on the contents of raw text and RCDATA elements</h5><p>The text in raw text and RCDATA elements must not contain any
1.1 mike 635: occurrences of the string "<code title=""></</code>" (U+003C
636: LESS-THAN SIGN, U+002F SOLIDUS) followed by characters that
637: case-insensitively match the tag name of the element followed by one
638: of U+0009 CHARACTER TABULATION, U+000A LINE FEED (LF), U+000C FORM
639: FEED (FF), U+0020 SPACE, U+003E GREATER-THAN SIGN (>), or U+002F
1.123 mike 640: SOLIDUS (/), unless that string is part of an <a href="#syntax-escape" title="syntax-escape">escaping text span</a>.</p><p>An <dfn id="syntax-escape" title="syntax-escape">escaping text span</dfn> is a span
1.1 mike 641: of <a href="#syntax-text" title="syntax-text">text</a> that starts with an <a href="#syntax-escape-start" title="syntax-escape-start">escaping text span start</a> that is
642: not itself in an <a href="#syntax-escape" title="syntax-escape">escaping text
643: span</a>, and ends at the next <a href="#syntax-escape-end" title="syntax-escape-end">escaping text span end</a>. There
644: cannot be any <a href="#syntax-charref" title="syntax-charref">character
645: references</a> inside an <a href="#syntax-escape" title="syntax-escape">escaping
646: text span</a> — sequences of characters that would look
647: like <a href="#syntax-charref" title="syntax-charref">character references</a> do not
1.123 mike 648: have special meaning.</p><p>An <dfn id="syntax-escape-start" title="syntax-escape-start">escaping text span
1.1 mike 649: start</dfn> is a part of <a href="#syntax-text" title="syntax-text">text</a> that
650: consists of the four character sequence "<code title=""><!--</code>" (U+003C LESS-THAN SIGN, U+0021 EXCLAMATION
1.123 mike 651: MARK, U+002D HYPHEN-MINUS, U+002D HYPHEN-MINUS).</p><p>An <dfn id="syntax-escape-end" title="syntax-escape-end">escaping text span end</dfn> is
1.1 mike 652: a part of <a href="#syntax-text" title="syntax-text">text</a> that consists of the
653: three character sequence "<code title="">--></code>" (U+002D
1.134 mike 654: HYPHEN-MINUS, U+002D HYPHEN-MINUS, U+003E GREATER-THAN SIGN).</p><p>An <a href="#syntax-escape-start" title="syntax-escape-start">escaping text span
1.1 mike 655: start</a> may share its U+002D HYPHEN-MINUS characters with its
656: corresponding <a href="#syntax-escape-end" title="syntax-escape-end">escaping text span
1.123 mike 657: end</a>.</p><p>The text in raw text elements and RCDATA elements must not have
1.91 mike 658: an <a href="#syntax-escape-start" title="syntax-escape-start">escaping text span start</a>
659: that is not followed by an <a href="#syntax-escape-end" title="syntax-escape-end">escaping
1.123 mike 660: text span end</a>.</p><h4 id="text-0"><span class="secno">9.1.3 </span>Text</h4><p><dfn id="syntax-text" title="syntax-text">Text</dfn> is allowed inside elements,
1.1 mike 661: attributes, and comments. Text must consist of Unicode characters.
662: Text must not contain U+0000 characters. Text must not contain
663: permanently undefined Unicode characters. Text must not contain
664: control characters other than <a href="infrastructure.html#space-character" title="space character">space
665: characters</a>. Extra constraints are placed on what is and what
666: is not allowed in text based on where the text is to be put, as
1.123 mike 667: described in the other sections.</p><h5 id="newlines"><span class="secno">9.1.3.1 </span>Newlines</h5><p><dfn id="syntax-newlines" title="syntax-newlines">Newlines</dfn> in HTML may be
1.1 mike 668: represented either as U+000D CARRIAGE RETURN (CR) characters, U+000A
669: LINE FEED (LF) characters, or pairs of U+000D CARRIAGE RETURN (CR),
1.123 mike 670: U+000A LINE FEED (LF) characters in that order.</p><h4 id="character-references"><span class="secno">9.1.4 </span>Character references</h4><p>In certain cases described in other sections, <a href="#syntax-text" title="syntax-text">text</a> may be mixed with <dfn id="syntax-charref" title="syntax-charref">character references</dfn>. These can be used
1.1 mike 671: to escape characters that couldn't otherwise legally be included in
1.123 mike 672: <a href="#syntax-text" title="syntax-text">text</a>.</p><p>Character references must start with a U+0026 AMPERSAND
1.1 mike 673: (<code>&</code>). Following this, there are three possible kinds
1.123 mike 674: of character references:</p><dl><dt>Named character references</dt>
1.1 mike 675:
676: <dd>The ampersand must be followed by one of the names given in the
677: <a href="named-character-references.html#named-character-references">named character references</a> section, using the same
678: case. The name must be one that is terminated by a U+003B SEMICOLON
679: (<code title="">;</code>) character.</dd>
680:
681:
682: <dt>Decimal numeric character reference</dt>
683:
684: <dd>The ampersand must be followed by a U+0023 NUMBER SIGN
685: (<code>#</code>) character, followed by one or more digits in the
686: range U+0030 DIGIT ZERO .. U+0039 DIGIT NINE, representing a
687: base-ten integer that itself is a Unicode code point that is
688: not U+0000, U+000D, in the range U+0080 .. U+009F, or in the range
689: 0xD800 .. 0xDFFF (surrogates). The digits must then be followed by
690: a U+003B SEMICOLON character (<code title="">;</code>).</dd>
691:
692:
693: <dt>Hexadecimal numeric character reference</dt>
694:
695: <dd>The ampersand must be followed by a U+0023 NUMBER SIGN
696: (<code>#</code>) character, which must be followed by either a
697: U+0078 LATIN SMALL LETTER X or a U+0058 LATIN CAPITAL LETTER X
698: character, which must then be followed by one or more digits in the
699: range U+0030 DIGIT ZERO .. U+0039 DIGIT NINE, U+0061 LATIN SMALL
700: LETTER A .. U+0066 LATIN SMALL LETTER F, and U+0041 LATIN CAPITAL
701: LETTER A .. U+0046 LATIN CAPITAL LETTER F, representing a
702: base-sixteen integer that itself is a Unicode code point that is
703: not U+0000, U+000D, in the range U+0080 .. U+009F, or in the range
704: 0xD800 .. 0xDFFF (surrogates). The digits must then be followed by
705: a U+003B SEMICOLON character (<code title="">;</code>).</dd>
706:
707: </dl><p>An <dfn id="syntax-ambiguous-ampersand" title="syntax-ambiguous-ampersand">ambiguous
708: ampersand</dfn> is a U+0026 AMPERSAND (<code>&</code>) character
709: that is followed by some <a href="#syntax-text" title="syntax-text">text</a> other
710: than a <a href="infrastructure.html#space-character">space character</a>, a U+003C LESS-THAN SIGN
711: character ('<'), or another U+0026 AMPERSAND (<code>&</code>)
1.123 mike 712: character.</p><h4 id="cdata-sections"><span class="secno">9.1.5 </span>CDATA sections</h4><p><dfn id="syntax-cdata" title="syntax-cdata">CDATA sections</dfn> must start with
1.1 mike 713: the character sequence U+003C LESS-THAN SIGN, U+0021 EXCLAMATION
714: MARK, U+005B LEFT SQUARE BRACKET, U+0043 LATIN CAPITAL LETTER C,
715: U+0044 LATIN CAPITAL LETTER D, U+0041 LATIN CAPITAL LETTER A, U+0054
716: LATIN CAPITAL LETTER T, U+0041 LATIN CAPITAL LETTER A, U+005B LEFT
717: SQUARE BRACKET (<code title=""><![CDATA[</code>). Following this
718: sequence, the CDATA section may have <a href="#syntax-text" title="syntax-text">text</a>, with the additional restriction
719: that the text must not contain the three character sequence U+005D
720: RIGHT SQUARE BRACKET, U+005D RIGHT SQUARE BRACKET, U+003E
721: GREATER-THAN SIGN (<code title="">]]></code>). Finally, the CDATA
722: section must be ended by the three character sequence U+005D RIGHT
723: SQUARE BRACKET, U+005D RIGHT SQUARE BRACKET, U+003E GREATER-THAN
1.159 mike 724: SIGN (<code title="">]]></code>).</p><h4 id="comments"><span class="secno">9.1.6 </span>Comments</h4><p class="XXX annotation"><b>Status: </b><i>Working draft</i></p><p><dfn id="syntax-comments" title="syntax-comments">Comments</dfn> must start with the
1.1 mike 725: four character sequence U+003C LESS-THAN SIGN, U+0021 EXCLAMATION
726: MARK, U+002D HYPHEN-MINUS, U+002D HYPHEN-MINUS (<code title=""><!--</code>). Following this sequence, the comment may
727: have <a href="#syntax-text" title="syntax-text">text</a>, with the additional
728: restriction that the text must not start with a single U+003E
729: GREATER-THAN SIGN ('>') character, nor start with a U+002D
730: HYPHEN-MINUS (<code title="">-</code>) character followed by a
731: U+003E GREATER-THAN SIGN ('>') character, nor contain two
732: consecutive U+002D HYPHEN-MINUS (<code title="">-</code>)
733: characters, nor end with a U+002D HYPHEN-MINUS (<code title="">-</code>) character. Finally, the comment must be ended by
734: the three character sequence U+002D HYPHEN-MINUS, U+002D
1.123 mike 735: HYPHEN-MINUS, U+003E GREATER-THAN SIGN (<code title="">--></code>).</p><div class="impl">
1.1 mike 736:
1.159 mike 737: <h3 id="parsing"><span class="secno">9.2 </span>Parsing HTML documents</h3><p class="XXX annotation"><b>Status: </b><i>Last call for comments</i></p>
1.1 mike 738:
739: <p><i>This section only applies to user agents, data mining tools,
740: and conformance checkers.</i></p>
741:
742: <p class="note">The rules for parsing XML documents into DOM trees
743: are covered by the next section, entitled "<a href="the-xhtml-syntax.html#the-xhtml-syntax">The XHTML
744: syntax</a>".</p>
745:
746: <p>For <a href="dom.html#html-documents">HTML documents</a>, user agents must use the parsing
747: rules described in this section to generate the DOM trees. Together,
748: these rules define what is referred to as the <dfn id="html-parser">HTML
749: parser</dfn>.</p>
750:
751: <div class="note">
752:
753: <p>While the HTML syntax described in this specification bears a
754: close resemblance to SGML and XML, it is a separate language with
755: its own parsing rules.</p>
756:
1.183 mike 757: <p>Some earlier versions of HTML (in particular from HTML2 to
758: HTML4) were based on SGML and used SGML parsing rules. However, few
1.1 mike 759: (if any) web browsers ever implemented true SGML parsing for HTML
760: documents; the only user agents to strictly handle HTML as an SGML
761: application have historically been validators. The resulting
762: confusion — with validators claiming documents to have one
763: representation while widely deployed Web browsers interoperably
764: implemented a different representation — has wasted decades
765: of productivity. This version of HTML thus returns to a non-SGML
766: basis.</p>
767:
768: <p>Authors interested in using SGML tools in their authoring
769: pipeline are encouraged to use XML tools and the XML serialization
770: of HTML.</p>
771:
772: </div>
773:
774: <p>This specification defines the parsing rules for HTML documents,
775: whether they are syntactically correct or not. Certain points in the
776: parsing algorithm are said to be <dfn id="parse-error" title="parse error">parse
777: errors</dfn>. The error handling for parse errors is well-defined:
778: user agents must either act as described below when encountering
779: such problems, or must abort processing at the first error that they
780: encounter for which they do not wish to apply the rules described
781: below.</p>
782:
783: <p>Conformance checkers must report at least one parse error
784: condition to the user if one or more parse error conditions exist in
785: the document and must not report parse error conditions if none
786: exist in the document. Conformance checkers may report more than one
1.174 mike 787: parse error condition if more than one parse error condition exists
1.1 mike 788: in the document. Conformance checkers are not required to recover
789: from parse errors.</p>
790:
791: <p class="note">Parse errors are only errors with the
792: <em>syntax</em> of HTML. In addition to checking for parse errors,
793: conformance checkers will also verify that the document obeys all
794: the other conformance requirements described in this
795: specification.</p>
796:
797: <p>For the purposes of conformance checkers, if a resource is
798: determined to be in <a href="#syntax">the HTML syntax</a>, then it is an
799: <a href="dom.html#html-documents" title="HTML documents">HTML document</a>.</p>
800:
801:
802: <h4 id="overview-of-the-parsing-model"><span class="secno">9.2.1 </span>Overview of the parsing model</h4>
803:
804: <p>The input to the HTML parsing process consists of a stream of
805: Unicode characters, which is passed through a
806: <a href="#tokenization">tokenization</a> stage followed by a <a href="#tree-construction">tree
807: construction</a> stage. The output is a <code>Document</code>
808: object.</p>
809:
810: <p class="note">Implementations that <a href="infrastructure.html#non-scripted">do not
811: support scripting</a> do not have to actually create a DOM
812: <code>Document</code> object, but the DOM tree in such cases is
813: still used as the model for the rest of the specification.</p>
814:
815: <p>In the common case, the data handled by the tokenization stage
1.159 mike 816: comes from the network, but <a href="embedded-content-0.html#dynamic-markup-insertion" title="dynamic markup
817: insertion">it can also come from script</a>, e.g. using the <code title="dom-document-write"><a href="embedded-content-0.html#dom-document-write">document.write()</a></code> API.</p>
1.1 mike 818:
819: <p><img alt="" src="images/parsing-model-overview.png"></p>
820:
821: <p id="nestedParsing">There is only one set of states for the
822: tokenizer stage and the tree construction stage, but the tree
823: construction stage is reentrant, meaning that while the tree
824: construction stage is handling one token, the tokenizer might be
825: resumed, causing further tokens to be emitted and processed before
826: the first token's processing is complete.</p>
827:
828: <div class="example">
829:
830: <p>In the following example, the tree construction stage will be
831: called upon to handle a "p" start tag token while handling the
832: "script" start tag token:</p>
833:
834: <pre>...
835: <script>
836: document.write('<p>');
837: </script>
838: ...</pre>
839:
840: </div>
841:
842: <p>To handle these cases, parsers have a <dfn id="script-nesting-level">script nesting
843: level</dfn>, which must be initially set to zero, and a <dfn id="parser-pause-flag">parser
844: pause flag</dfn>, which must be initially set to false.</p>
845:
846:
847:
848: <h4 id="the-input-stream"><span class="secno">9.2.2 </span>The <dfn>input stream</dfn></h4>
849:
850: <p>The stream of Unicode characters that comprises the input to the
851: tokenization stage will be initially seen by the user agent as a
852: stream of bytes (typically coming over the network or from the local
853: file system). The bytes encode the actual characters according to a
854: particular <em>character encoding</em>, which the user agent must
855: use to decode the bytes into characters.</p>
856:
857: <p class="note">For XML documents, the algorithm user agents must
858: use to determine the character encoding is given by the XML
1.104 mike 859: specification. This section does not apply to XML documents. <a href="references.html#refsXML">[XML]</a></p>
1.1 mike 860:
861:
1.168 mike 862: <h5 id="determining-the-character-encoding"><span class="secno">9.2.2.1 </span>Determining the character encoding</h5><p class="XXX annotation"><b>Status: </b><i>Working draft</i></p>
1.1 mike 863:
864: <p>In some cases, it might be impractical to unambiguously determine
865: the encoding before parsing the document. Because of this, this
866: specification provides for a two-pass mechanism with an optional
867: pre-scan. Implementations are allowed, as described below, to apply
868: a simplified parsing algorithm to whatever bytes they have available
869: before beginning to parse the document. Then, the real parser is
870: started, using a tentative encoding derived from this pre-parse and
871: other out-of-band metadata. If, while the document is being loaded,
872: the user agent discovers an encoding declaration that conflicts with
873: this information, then the parser can get reinvoked to perform a
874: parse of the document with the real encoding.</p>
875:
876: <p id="documentEncoding">User agents must use the following
877: algorithm (the <dfn id="encoding-sniffing-algorithm">encoding sniffing algorithm</dfn>) to determine
878: the character encoding to use when decoding a document in the first
879: pass. This algorithm takes as input any out-of-band metadata
880: available to the user agent (e.g. the <a href="infrastructure.html#content-type" title="Content-Type">Content-Type metadata</a> of the document)
881: and all the bytes available so far, and returns an encoding and a
882: <dfn id="concept-encoding-confidence" title="concept-encoding-confidence">confidence</dfn>. The
883: confidence is either <i>tentative</i>, <i>certain</i>, or
884: <i>irrelevant</i>. The encoding used, and whether the confidence in
885: that encoding is <i>tentative</i> or <i>certain</i>, is <a href="#meta-charset-during-parse">used during the parsing</a> to
886: determine whether to <a href="#change-the-encoding">change the encoding</a>. If no
887: encoding is necessary, e.g. because the parser is operating on a
888: stream of Unicode characters and doesn't have to use an encoding at
889: all, then the <a href="#concept-encoding-confidence" title="concept-encoding-confidence">confidence</a> is
890: <i>irrelevant</i>.</p>
891:
892: <ol><li><p>If the transport layer specifies an encoding, and it is
893: supported, return that encoding with the <a href="#concept-encoding-confidence" title="concept-encoding-confidence">confidence</a>
894: <i>certain</i>, and abort these steps.</p></li>
895:
896: <li><p>The user agent may wait for more bytes of the resource to be
897: available, either in this step or at any later step in this
898: algorithm. For instance, a user agent might wait 500ms or 512
899: bytes, whichever came first. In general preparsing the source to
900: find the encoding improves performance, as it reduces the need to
901: throw away the data structures used when parsing upon finding the
902: encoding information. However, if the user agent delays too long to
903: obtain data to determine the encoding, then the cost of the delay
904: could outweigh any performance improvements from the
905: preparse.</p></li>
906:
907: <li><p>For each of the rows in the following table, starting with
908: the first one and going down, if there are as many or more bytes
909: available than the number of bytes in the first column, and the
910: first bytes of the file match the bytes given in the first column,
911: then return the encoding given in the cell in the second column of
912: that row, with the <a href="#concept-encoding-confidence" title="concept-encoding-confidence">confidence</a>
913: <i>certain</i>, and abort these steps:</p>
914:
915: <!-- this table is present in several forms in this file; keep them in sync -->
916: <table><thead><tr><th>Bytes in Hexadecimal
917: </th><th>Encoding
918: </th></tr></thead><tbody><!-- nobody uses this
919: <tr>
920: <td>00 00 FE FF
921: <td>UTF-32BE
922: <tr>
923: <td>FF FE 00 00
924: <td>UTF-32LE
925: --><tr><td>FE FF
926: </td><td>UTF-16BE
927: </td></tr><tr><td>FF FE
928: </td><td>UTF-16LE
929: </td></tr><tr><td>EF BB BF
930: </td><td>UTF-8
931: <!-- nobody uses this
932: <tr>
933: <td>DD 73 66 73
934: <td>UTF-EBCDIC
935: -->
936: </td></tr></tbody></table><p class="note">This step looks for Unicode Byte Order Marks
937: (BOMs).</p></li>
938:
939: <li><p>Otherwise, the user agent will have to search for explicit
940: character encoding information in the file itself. This should
941: proceed as follows:
942:
943: </p><p>Let <var title="">position</var> be a pointer to a byte in the
944: input stream, initially pointing at the first byte. If at any
945: point during these substeps the user agent either runs out of
946: bytes or decides that scanning further bytes would not be
947: efficient, then skip to the next step of the overall character
948: encoding detection algorithm. User agents may decide that scanning
949: <em>any</em> bytes is not efficient, in which case these substeps
950: are entirely skipped.</p>
951:
952: <p>Now, repeat the following "two" steps until the algorithm
953: aborts (either because user agent aborts, as described above, or
954: because a character encoding is found):</p>
955:
956: <ol><li><p>If <var title="">position</var> points to:</p>
957:
958: <dl class="switch"><dt>A sequence of bytes starting with: 0x3C 0x21 0x2D 0x2D (ASCII '<!--')</dt>
959: <dd>
960:
961: <p>Advance the <var title="">position</var> pointer so that it
962: points at the first 0x3E byte which is preceded by two 0x2D
963: bytes (i.e. at the end of an ASCII '-->' sequence) and comes
964: after the 0x3C byte that was found. (The two 0x2D bytes can be
965: the same as the those in the '<!--' sequence.)</p>
966:
967: </dd>
968:
969: <dt>A sequence of bytes starting with: 0x3C, 0x4D or 0x6D, 0x45 or 0x65, 0x54 or 0x74, 0x41 or 0x61, and finally one of 0x09, 0x0A, 0x0C, 0x0D, 0x20, 0x2F (case-insensitive ASCII '<meta' followed by a space or slash)</dt>
970: <dd>
971:
972: <ol><li><p>Advance the <var title="">position</var> pointer so
973: that it points at the next 0x09, 0x0A, 0x0C, 0x0D, 0x20, or
974: 0x2F byte (the one in sequence of characters matched
975: above).</p></li>
976:
977: <li><p><a href="#concept-get-attributes-when-sniffing" title="concept-get-attributes-when-sniffing">Get
978: an attribute</a> and its value. If no attribute was
979: sniffed, then skip this inner set of steps, and jump to the
980: second step in the overall "two step" algorithm.</p></li>
981:
982: <li><p>If the attribute's name is neither "<code title="">charset</code>" nor "<code title="">content</code>",
983: then return to step 2 in these inner steps.</p></li>
984:
985: <li><p>If the attribute's name is "<code title="">charset</code>", let <var title="">charset</var> be
986: the attribute's value, interpreted as a character
987: encoding.</p></li>
988:
989: <li><p>Otherwise, the attribute's name is "<code title="">content</code>": apply the <a href="infrastructure.html#algorithm-for-extracting-an-encoding-from-a-content-type">algorithm for
990: extracting an encoding from a Content-Type</a>, giving the
991: attribute's value as the string to parse. If an encoding is
992: returned, let <var title="">charset</var> be that
993: encoding. Otherwise, return to step 2 in these inner
994: steps.</p></li>
995:
996: <li><p>If <var title="">charset</var> is a UTF-16 encoding,
997: change the value of <var title="">charset</var> to
998: UTF-8.</p></li>
999:
1000: <li><p>If <var title="">charset</var> is a supported
1001: character encoding, then return the given encoding, with
1002: <a href="#concept-encoding-confidence" title="concept-encoding-confidence">confidence</a>
1003: <i>tentative</i>, and abort all these steps.</p></li>
1004:
1005: <li><p>Otherwise, return to step 2 in these inner
1006: steps.</p></li>
1007:
1008: </ol></dd>
1009:
1010: <dt>A sequence of bytes starting with a 0x3C byte (ASCII '<'), optionally a 0x2F byte (ASCII '/'), and finally a byte in the range 0x41-0x5A or 0x61-0x7A (an ASCII letter)</dt>
1011: <dd>
1012:
1013: <ol><li><p>Advance the <var title="">position</var> pointer so
1014: that it points at the next 0x09 (ASCII TAB), 0x0A (ASCII LF),
1015: 0x0C (ASCII FF), 0x0D (ASCII CR), 0x20 (ASCII space), or 0x3E
1016: (ASCII '>') byte.</p></li>
1017:
1018: <li><p>Repeatedly <a href="#concept-get-attributes-when-sniffing" title="concept-get-attributes-when-sniffing">get an
1019: attribute</a> until no further attributes can be found,
1020: then jump to the second step in the overall "two step"
1021: algorithm.</p></li>
1022:
1023: </ol></dd>
1024:
1025: <dt>A sequence of bytes starting with: 0x3C 0x21 (ASCII '<!')</dt>
1026: <dt>A sequence of bytes starting with: 0x3C 0x2F (ASCII '</')</dt>
1027: <dt>A sequence of bytes starting with: 0x3C 0x3F (ASCII '<?')</dt>
1028: <dd>
1029:
1030: <p>Advance the <var title="">position</var> pointer so that it
1031: points at the first 0x3E byte (ASCII '>') that comes after the
1032: 0x3C byte that was found.</p>
1033:
1034: </dd>
1035:
1036: <dt>Any other byte</dt>
1037: <dd>
1038:
1039: <p>Do nothing with that byte.</p>
1040:
1041: </dd>
1042:
1043: </dl></li>
1044:
1045: <li>Move <var title="">position</var> so it points at the next
1046: byte in the input stream, and return to the first step of this
1047: "two step" algorithm.</li>
1048:
1049: </ol><p>When the above "two step" algorithm says to <dfn id="concept-get-attributes-when-sniffing" title="concept-get-attributes-when-sniffing">get an
1050: attribute</dfn>, it means doing this:</p>
1051:
1052: <ol><li><p>If the byte at <var title="">position</var> is one of 0x09
1053: (ASCII TAB), 0x0A (ASCII LF), 0x0C (ASCII FF), 0x0D (ASCII CR),
1054: 0x20 (ASCII space), or 0x2F (ASCII '/') then advance <var title="">position</var> to the next byte and redo this
1055: substep.</p></li>
1056:
1057: <li><p>If the byte at <var title="">position</var> is 0x3E (ASCII
1058: '>'), then abort the "get an attribute" algorithm. There isn't
1059: one.</p></li>
1060:
1061: <li><p>Otherwise, the byte at <var title="">position</var> is the
1062: start of the attribute name. Let <var title="">attribute
1063: name</var> and <var title="">attribute value</var> be the empty
1064: string.</p></li>
1065:
1066: <li><p><i>Attribute name</i>: Process the byte at <var title="">position</var> as follows:</p>
1067:
1068: <dl class="switch"><dt>If it is 0x3D (ASCII '='), and the <var title="">attribute
1069: name</var> is longer than the empty string</dt>
1070:
1071: <dd>Advance <var title="">position</var> to the next byte and
1072: jump to the step below labeled <i>value</i>.</dd>
1073:
1074: <dt>If it is 0x09 (ASCII TAB), 0x0A (ASCII LF), 0x0C (ASCII
1075: FF), 0x0D (ASCII CR), or 0x20 (ASCII space)</dt>
1076:
1077: <dd>Jump to the step below labeled <i>spaces</i>.</dd>
1078:
1079: <dt>If it is 0x2F (ASCII '/') or 0x3E (ASCII '>')</dt>
1080:
1081: <dd>Abort the "get an attribute" algorithm. The attribute's
1082: name is the value of <var title="">attribute name</var>, its
1083: value is the empty string.</dd>
1084:
1085: <dt>If it is in the range 0x41 (ASCII 'A') to 0x5A (ASCII
1086: 'Z')</dt>
1087:
1088: <dd>Append the Unicode character with code point <span title=""><var title="">b</var>+0x20</span> to <var title="">attribute
1089: name</var> (where <var title="">b</var> is the value of the
1090: byte at <var title="">position</var>).</dd>
1091:
1092: <dt>Anything else</dt>
1093:
1094: <dd>Append the Unicode character with the same code point as the
1095: value of the byte at <var title="">position</var>) to <var title="">attribute name</var>. (It doesn't actually matter how
1096: bytes outside the ASCII range are handled here, since only
1097: ASCII characters can contribute to the detection of a character
1098: encoding.)</dd>
1099:
1100: </dl></li>
1101:
1102: <li><p>Advance <var title="">position</var> to the next byte and
1103: return to the previous step.</p></li>
1104:
1105: <li><p><i>Spaces</i>: If the byte at <var title="">position</var> is one of 0x09 (ASCII TAB), 0x0A (ASCII
1106: LF), 0x0C (ASCII FF), 0x0D (ASCII CR), or 0x20 (ASCII space) then
1107: advance <var title="">position</var> to the next byte, then,
1108: repeat this step.</p></li>
1109:
1110: <li><p>If the byte at <var title="">position</var> is
1111: <em>not</em> 0x3D (ASCII '='), abort the "get an attribute"
1112: algorithm. The attribute's name is the value of <var title="">attribute name</var>, its value is the empty
1113: string.</p></li>
1114:
1115: <li><p>Advance <var title="">position</var> past the 0x3D (ASCII
1116: '=') byte.</p></li>
1117:
1118: <li><p><i>Value:</i> If the byte at <var title="">position</var> is one of 0x09 (ASCII TAB), 0x0A (ASCII
1119: LF), 0x0C (ASCII FF), 0x0D (ASCII CR), or 0x20 (ASCII space) then
1120: advance <var title="">position</var> to the next byte, then,
1121: repeat this step.</p></li>
1122:
1123: <li><p>Process the byte at <var title="">position</var> as
1124: follows:</p>
1125:
1126: <dl class="switch"><dt>If it is 0x22 (ASCII '"') or 0x27 ("'")</dt>
1127:
1128: <dd>
1129:
1130: <ol><li>Let <var title="">b</var> be the value of the byte at
1131: <var title="">position</var>.</li>
1132:
1133: <li>Advance <var title="">position</var> to the next
1134: byte.</li>
1135:
1136: <li>If the value of the byte at <var title="">position</var>
1137: is the value of <var title="">b</var>, then advance <var title="">position</var> to the next byte and abort the "get
1138: an attribute" algorithm. The attribute's name is the value of
1139: <var title="">attribute name</var>, and its value is the
1140: value of <var title="">attribute value</var>.</li>
1141:
1142: <li>Otherwise, if the value of the byte at <var title="">position</var> is in the range 0x41 (ASCII 'A') to
1143: 0x5A (ASCII 'Z'), then append a Unicode character to <var title="">attribute value</var> whose code point is 0x20 more
1144: than the value of the byte at <var title="">position</var>.</li>
1145:
1146: <li>Otherwise, append a Unicode character to <var title="">attribute value</var> whose code point is the same as
1147: the value of the byte at <var title="">position</var>.</li>
1148:
1149: <li>Return to the second step in these substeps.</li>
1150:
1151: </ol></dd>
1152:
1153: <dt>If it is 0x3E (ASCII '>')</dt>
1154:
1155: <dd>Abort the "get an attribute" algorithm. The attribute's
1156: name is the value of <var title="">attribute name</var>, its
1157: value is the empty string.</dd>
1158:
1159:
1160: <dt>If it is in the range 0x41 (ASCII 'A') to 0x5A (ASCII
1161: 'Z')</dt>
1162:
1163: <dd>Append the Unicode character with code point <span title=""><var title="">b</var>+0x20</span> to <var title="">attribute
1164: value</var> (where <var title="">b</var> is the value of the
1165: byte at <var title="">position</var>). Advance <var title="">position</var> to the next byte.</dd>
1166:
1167: <dt>Anything else</dt>
1168:
1169: <dd>Append the Unicode character with the same code point as the
1170: value of the byte at <var title="">position</var>) to <var title="">attribute value</var>. Advance <var title="">position</var> to the next byte.</dd>
1171:
1172: </dl></li>
1173:
1174: <li><p>Process the byte at <var title="">position</var> as
1175: follows:</p>
1176:
1177: <dl class="switch"><dt>If it is 0x09 (ASCII TAB), 0x0A (ASCII LF), 0x0C (ASCII
1178: FF), 0x0D (ASCII CR), 0x20 (ASCII space), or 0x3E (ASCII
1179: '>')</dt>
1180:
1181: <dd>Abort the "get an attribute" algorithm. The attribute's
1182: name is the value of <var title="">attribute name</var> and its
1183: value is the value of <var title="">attribute value</var>.</dd>
1184:
1185: <dt>If it is in the range 0x41 (ASCII 'A') to 0x5A (ASCII
1186: 'Z')</dt>
1187:
1188: <dd>Append the Unicode character with code point <span title=""><var title="">b</var>+0x20</span> to <var title="">attribute
1189: value</var> (where <var title="">b</var> is the value of the
1190: byte at <var title="">position</var>).</dd>
1191:
1192: <dt>Anything else</dt>
1193:
1194: <dd>Append the Unicode character with the same code point as the
1195: value of the byte at <var title="">position</var>) to <var title="">attribute value</var>.</dd>
1196:
1197: </dl></li>
1198:
1199: <li><p>Advance <var title="">position</var> to the next byte and
1200: return to the previous step.</p></li>
1201:
1202: </ol><p>For the sake of interoperability, user agents should not use a
1203: pre-scan algorithm that returns different results than the one
1204: described above. (But, if you do, please at least let us know, so
1205: that we can improve this algorithm and benefit everyone...)</p>
1206:
1207: </li>
1208:
1209: <li><p>If the user agent has information on the likely encoding for
1210: this page, e.g. based on the encoding of the page when it was last
1211: visited, then return that encoding, with the <a href="#concept-encoding-confidence" title="concept-encoding-confidence">confidence</a>
1212: <i>tentative</i>, and abort these steps.</p></li>
1213:
1214: <li><p>The user agent may attempt to autodetect the character
1215: encoding from applying frequency analysis or other algorithms to
1216: the data stream. If autodetection succeeds in determining a
1217: character encoding, then return that encoding, with the <a href="#concept-encoding-confidence" title="concept-encoding-confidence">confidence</a>
1.104 mike 1218: <i>tentative</i>, and abort these steps. <a href="references.html#refsUNIVCHARDET">[UNIVCHARDET]</a></p></li>
1.1 mike 1219:
1220: <li><p>Otherwise, return an implementation-defined or
1221: user-specified default character encoding, with the <a href="#concept-encoding-confidence" title="concept-encoding-confidence">confidence</a>
1222: <i>tentative</i>. In non-legacy environments, the more
1223: comprehensive <code title="">UTF-8</code> encoding is
1224: recommended. Due to its use in legacy content, <code title="">windows-1252</code> is recommended as a default in
1.187 mike 1225: predominantly Western demographics instead.</p></li>
1.1 mike 1226:
1227: </ol><p>The <a href="dom.html#document-s-character-encoding">document's character encoding</a> must immediately
1228: be set to the value returned from this algorithm, at the same time
1229: as the user agent uses the returned value to select the decoder to
1230: use for the input stream.</p>
1231:
1232:
1.237 mike 1233: <h5 id="character-encodings-0"><span class="secno">9.2.2.2 </span>Character encodings</h5><p class="XXX annotation"><b>Status: </b><i>Working draft</i></p>
1.1 mike 1234:
1.237 mike 1235: <p>User agents must at a minimum support the UTF-8 and Windows-1252
1236: encodings, but may support more.</p>
1237:
1238: <p class="note">It is not unusual for Web browsers to support dozens
1239: if not upwards of a hundred distinct character encodings.</p>
1240:
1241: <p>User agents must support the preferred MIME name of every
1242: character encoding they support that has a preferred MIME name, and
1243: should support all the IANA-registered aliases of every character
1244: encoding they support. <a href="references.html#refsIANACHARSET">[IANACHARSET]</a></p>
1245:
1246: <p>When comparing a string specifying a character encoding with the
1247: name or alias of a character encoding to determine if they are
1248: equal, user agents must remove any leading or trailing <a href="infrastructure.html#space-character" title="space character">space characters</a> in both names, and
1249: then perform the comparison in an <a href="infrastructure.html#ascii-case-insensitive">ASCII
1250: case-insensitive</a> manner.</p>
1251:
1252: <!-- this bit will be replaced by actual alias registrations in due course -->
1253:
1254: <p>In addition, user agents must support the aliases given in the
1255: following table for every character encoding they support, so that
1256: labels from the first column are treated as equivalent to the labels
1257: given in the corresponding cell from the second column on the same
1258: row.</p>
1259:
1260: <table><caption>Additional character encoding aliases</caption>
1261: <thead><tr><th> Alias </th><th> Corresponding encoding </th><th> References
1262: </th></tr></thead><tbody><tr><td> x-sjis </td><td> windows-31J </td><td>
1263: <a href="references.html#refsSHIFTJIS">[SHIFTJIS]</a>
1264: <a href="references.html#refsWIN31J">[WIN31J]</a>
1265: </td></tr><tr><td> windows-932 </td><td> windows-31J </td><td>
1266: <a href="references.html#refsWIN31J">[WIN31J]</a>
1267: </td></tr><tr><td> x-x-big5 </td><td> Big5 </td><td>
1268: <a href="references.html#refsBIG5">[BIG5]</a>
1269: </td></tr></tbody></table><!-- end of bit that will be replaced by actual alias registrations in due course --><hr><p>When a user agent would otherwise use an encoding given in the
1270: first column of the following table to either convert content to
1271: Unicode characters or convert Unicode characters to bytes, it must
1272: instead use the encoding given in the cell in the second column of
1273: the same row. When a byte or sequence of bytes is treated
1274: differently due to this encoding aliasing, it is said to have been
1275: <dfn id="misinterpreted-for-compatibility">misinterpreted for compatibility</dfn>.</p>
1276:
1277: <table><caption>Character encoding overrides</caption>
1278: <thead><tr><th> Input encoding </th><th> Replacement encoding </th><th> References
1279: </th></tr></thead><tbody><!-- how about EUC-JP? --><tr><td> EUC-KR </td><td> windows-949 </td><td>
1280: <a href="references.html#refsEUCKR">[EUCKR]</a>
1281: <a href="references.html#refsWIN949">[WIN949]</a>
1282: </td></tr><tr><td> GB2312 </td><td> GBK </td><td>
1283: <a href="references.html#refsRFC1345">[RFC1345]</a>
1284: <a href="references.html#refsGBK">[GBK]</a>
1285: </td></tr><tr><td> GB_2312-80 </td><td> GBK </td><td>
1286: <a href="references.html#refsRFC1345">[RFC1345]</a>
1287: <a href="references.html#refsGBK">[GBK]</a>
1288: </td></tr><tr><td> ISO-8859-1 </td><td> windows-1252 </td><td>
1289: <a href="references.html#refsRFC1345">[RFC1345]</a>
1290: <a href="references.html#refsWIN1252">[WIN1252]</a>
1291: </td></tr><tr><td> ISO-8859-9 </td><td> windows-1254 </td><td>
1292: <a href="references.html#refsRFC1345">[RFC1345]</a>
1293: <a href="references.html#refsWIN1254">[WIN1254]</a>
1294: </td></tr><tr><td> ISO-8859-11 </td><td> windows-874 </td><td>
1295: <a href="references.html#refsISO885911">[ISO885911]</a>
1296: <a href="references.html#refsWIN874">[WIN874]</a>
1297: </td></tr><tr><td> KS_C_5601-1987 </td><td> windows-949 </td><td>
1298: <a href="references.html#refsRFC1345">[RFC1345]</a>
1299: <a href="references.html#refsWIN949">[WIN949]</a>
1300: </td></tr><tr><td> Shift_JIS </td><td> windows-31J </td><td>
1301: <a href="references.html#refsSHIFTJIS">[SHIFTJIS]</a>
1302: <a href="references.html#refsWIN31J">[WIN31J]</a>
1303: </td></tr><tr><td> TIS-620 </td><td> windows-874 </td><td>
1304: <a href="references.html#refsTIS620">[TIS620]</a>
1305: <a href="references.html#refsWIN874">[WIN874]</a>
1306: </td></tr><tr><td> US-ASCII </td><td> windows-1252 </td><td>
1307: <a href="references.html#refsRFC1345">[RFC1345]</a>
1308: <a href="references.html#refsWIN1252">[WIN1252]</a>
1309: </td></tr></tbody></table><p class="note">The requirement to treat certain encodings as other
1310: encodings according to the table above is a <a href="introduction.html#willful-violation">willful
1311: violation</a> of the W3C Character Model specification, motivated
1312: by a desire for compatibility with legacy content. <a href="references.html#refsCHARMOD">[CHARMOD]</a></p>
1313:
1314: <p>When a user agent is to use the UTF-16 encoding but no BOM has
1315: been found, user agents must default to UTF-16LE.</p>
1316:
1317: <p class="note">The requirement to default UTF-16 to LE rather than
1318: BE is a <a href="introduction.html#willful-violation">willful violation</a> of RFC 2781, motivated by a
1319: desire for compatibility with legacy content. <a href="references.html#refsCHARMOD">[CHARMOD]</a></p>
1320:
1321: <hr><p>User agents must not support the CESU-8, UTF-7, BOCU-1 and SCSU
1322: encodings. <a href="references.html#refsCESU8">[CESU8]</a> <a href="references.html#refsUTF7">[UTF7]</a> <a href="references.html#refsBOCU1">[BOCU1]</a> <a href="references.html#refsSCSU">[SCSU]</a></p>
1323:
1324: <p>Support for encodings based on EBCDIC is not recommended. This
1325: encoding is rarely used for publicly-facing Web content.</p>
1326:
1327: <p>Support for UTF-32 is not recommended. This encoding is rarely
1328: used, and frequently implemented incorrectly.</p>
1329:
1330: <p class="note">This specification does not make any attempt to
1331: support EBCDIC-based encodings and UTF-32 in its algorithms; support
1332: and use of these encodings can thus lead to unexpected behavior in
1333: implementations of this specification.</p>
1334:
1335:
1336:
1337: <h5 id="preprocessing-the-input-stream"><span class="secno">9.2.2.3 </span>Preprocessing the input stream</h5>
1.1 mike 1338:
1339: <p>Given an encoding, the bytes in the input stream must be
1340: converted to Unicode characters for the tokenizer, as described by
1341: the rules for that encoding, except that the leading U+FEFF BYTE
1342: ORDER MARK character, if any, must not be stripped by the encoding
1343: layer (it is stripped by the rule below).</p> <!-- this is to
1344: prevent two leading BOMs from being both stripped, once by the
1345: decoder, and once by the parser -->
1346:
1347: <p>Bytes or sequences of bytes in the original byte stream that
1348: could not be converted to Unicode characters must be converted to
1349: U+FFFD REPLACEMENT CHARACTER code points.</p>
1350:
1351: <p class="note">Bytes or sequences of bytes in the original byte
1352: stream that did not conform to the encoding specification
1353: (e.g. invalid UTF-8 byte sequences in a UTF-8 input stream) are
1354: errors that conformance checkers are expected to report.</p>
1355:
1356: <p>Any byte or sequences of bytes in the original byte stream that
1.237 mike 1357: is <a href="#misinterpreted-for-compatibility">misinterpreted for compatibility</a> is a <a href="#parse-error">parse
1.1 mike 1358: error</a>.</p>
1359:
1360: <p>One leading U+FEFF BYTE ORDER MARK character must be ignored if
1361: any are present.</p>
1362:
1363: <p>All U+0000 NULL characters in the input must be replaced by
1364: U+FFFD REPLACEMENT CHARACTERs. Any occurrences of such characters is
1365: a <a href="#parse-error">parse error</a>.</p>
1366:
1367: <p>Any occurrences of any characters in the ranges U+0001 to U+0008,
1368: <!-- HT, LF allowed --> <!-- U+000B is in the next list --> <!-- FF,
1369: CR allowed --> U+000E to U+001F, <!-- ASCII allowed --> U+007F
1370: <!--to U+0084, (U+0085 NEL not allowed), U+0086--> to U+009F, U+D800
1371: to U+DFFF<!-- surrogates not allowed -->, U+FDD0 to U+FDEF, and
1372: characters U+000B, U+FFFE, U+FFFF, U+1FFFE, U+1FFFF, U+2FFFE,
1373: U+2FFFF, U+3FFFE, U+3FFFF, U+4FFFE, U+4FFFF, U+5FFFE, U+5FFFF,
1374: U+6FFFE, U+6FFFF, U+7FFFE, U+7FFFF, U+8FFFE, U+8FFFF, U+9FFFE,
1375: U+9FFFF, U+AFFFE, U+AFFFF, U+BFFFE, U+BFFFF, U+CFFFE, U+CFFFF,
1376: U+DFFFE, U+DFFFF, U+EFFFE, U+EFFFF, U+FFFFE, U+FFFFF, U+10FFFE, and
1377: U+10FFFF are <a href="#parse-error" title="parse error">parse errors</a>. (These
1378: are all control characters or permanently undefined Unicode
1379: characters.)</p>
1380:
1381: <p>U+000D CARRIAGE RETURN (CR) characters and U+000A LINE FEED (LF)
1382: characters are treated specially. Any CR characters that are
1383: followed by LF characters must be removed, and any CR characters not
1384: followed by LF characters must be converted to LF characters. Thus,
1385: newlines in HTML DOMs are represented by LF characters, and there
1386: are never any CR characters in the input to the
1387: <a href="#tokenization">tokenization</a> stage.</p>
1388:
1389: <p>The <dfn id="next-input-character">next input character</dfn> is the first character in the
1390: input stream that has not yet been <dfn id="consumed">consumed</dfn>. Initially,
1391: the <i><a href="#next-input-character">next input character</a></i> is the first character in the
1392: input. The <dfn id="current-input-character">current input character</dfn> is the last character
1393: to have been <i><a href="#consumed">consumed</a></i>.</p>
1394:
1395: <p>The <dfn id="insertion-point">insertion point</dfn> is the position (just before a
1396: character or just before the end of the input stream) where content
1.159 mike 1397: inserted using <code title="dom-document-write"><a href="embedded-content-0.html#dom-document-write">document.write()</a></code> is actually
1.1 mike 1398: inserted. The insertion point is relative to the position of the
1399: character immediately after it, it is not an absolute offset into
1400: the input stream. Initially, the insertion point is
1.10 mike 1401: undefined.</p>
1.1 mike 1402:
1403: <p>The "EOF" character in the tables below is a conceptual character
1404: representing the end of the <a href="#the-input-stream">input stream</a>. If the parser
1.159 mike 1405: is a <a href="embedded-content-0.html#script-created-parser">script-created parser</a>, then the end of the
1.1 mike 1406: <a href="#the-input-stream">input stream</a> is reached when an <dfn id="explicit-eof-character">explicit "EOF"
1.159 mike 1407: character</dfn> (inserted by the <code title="dom-document-close"><a href="embedded-content-0.html#dom-document-close">document.close()</a></code> method) is
1.1 mike 1408: consumed. Otherwise, the "EOF" character is not a real character in
1409: the stream, but rather the lack of any further characters.</p>
1410:
1411:
1.237 mike 1412: <h5 id="changing-the-encoding-while-parsing"><span class="secno">9.2.2.4 </span>Changing the encoding while parsing</h5>
1.1 mike 1413:
1414: <p>When the parser requires the user agent to <dfn id="change-the-encoding">change the
1415: encoding</dfn>, it must run the following steps. This might happen
1416: if the <a href="#encoding-sniffing-algorithm">encoding sniffing algorithm</a> described above
1417: failed to find an encoding, or if it found an encoding that was not
1418: the actual encoding of the file.</p>
1419:
1420: <ol><li>If the new encoding is identical or equivalent to the encoding
1421: that is already being used to interpret the input stream, then set
1422: the <a href="#concept-encoding-confidence" title="concept-encoding-confidence">confidence</a> to
1423: <i>certain</i> and abort these steps. This happens when the
1424: encoding information found in the file matches what the
1425: <a href="#encoding-sniffing-algorithm">encoding sniffing algorithm</a> determined to be the
1426: encoding, and in the second pass through the parser if the first
1427: pass found that the encoding sniffing algorithm described in the
1428: earlier section failed to find the right encoding.</li>
1429:
1430: <li>If the encoding that is already being used to interpret the
1431: input stream is a UTF-16 encoding, then set the <a href="#concept-encoding-confidence" title="concept-encoding-confidence">confidence</a> to
1432: <i>certain</i> and abort these steps. The new encoding is ignored;
1433: if it was anything but the same encoding, then it would be clearly
1434: incorrect.</li>
1435:
1436: <li>If the new encoding is a UTF-16 encoding, change it to
1437: UTF-8.</li>
1438:
1439: <li>If all the bytes up to the last byte converted by the current
1440: decoder have the same Unicode interpretations in both the current
1441: encoding and the new encoding, and if the user agent supports
1442: changing the converter on the fly, then the user agent may change
1443: to the new converter for the encoding on the fly. Set the
1444: <a href="dom.html#document-s-character-encoding">document's character encoding</a> and the encoding used to
1445: convert the input stream to the new encoding, set the <a href="#concept-encoding-confidence" title="concept-encoding-confidence">confidence</a> to
1446: <i>certain</i>, and abort these steps.</li>
1447:
1448: <li>Otherwise, <a href="history.html#navigate">navigate</a> to the document again, with
1449: <a href="history.html#replacement-enabled">replacement enabled</a>, and using the same <a href="history.html#source-browsing-context">source
1450: browsing context</a>, but this time skip the <a href="#encoding-sniffing-algorithm">encoding
1451: sniffing algorithm</a> and instead just set the encoding to the
1452: new encoding and the <a href="#concept-encoding-confidence" title="concept-encoding-confidence">confidence</a> to
1453: <i>certain</i>. Whenever possible, this should be done without
1454: actually contacting the network layer (the bytes should be
1455: re-parsed from memory), even if, e.g., the document is marked as
1456: not being cacheable. If this is not possible and contacting the
1457: network layer would involve repeating a request that uses a method
1458: other than HTTP GET (<a href="infrastructure.html#concept-http-equivalent-get" title="concept-http-equivalent-get">or
1459: equivalent</a> for non-HTTP URLs), then instead set the <a href="#concept-encoding-confidence" title="concept-encoding-confidence">confidence</a> to
1460: <i>certain</i> and ignore the new encoding. The resource will be
1461: misinterpreted. User agents may notify the user of the situation,
1462: to aid in application development.</li>
1463:
1464: </ol><h4 id="parse-state"><span class="secno">9.2.3 </span>Parse state</h4>
1465:
1466: <h5 id="the-insertion-mode"><span class="secno">9.2.3.1 </span>The insertion mode</h5>
1467:
1468: <p>The <dfn id="insertion-mode">insertion mode</dfn> is a state variable that controls
1469: the primary operation of the tree construction stage.</p>
1470:
1471: <p>Initially, the <a href="#insertion-mode">insertion mode</a> is "<a href="#the-initial-insertion-mode" title="insertion mode: initial">initial</a>". It can change to
1472: "<a href="#the-before-html-insertion-mode" title="insertion mode: before html">before html</a>",
1473: "<a href="#the-before-head-insertion-mode" title="insertion mode: before head">before head</a>",
1474: "<a href="#parsing-main-inhead" title="insertion mode: in head">in head</a>", "<a href="#parsing-main-inheadnoscript" title="insertion mode: in head noscript">in head noscript</a>",
1.91 mike 1475: "<a href="#the-after-head-insertion-mode" title="insertion mode: after head">after head</a>", "<a href="#parsing-main-inbody" title="insertion mode: in body">in body</a>", "<a href="#parsing-main-incdata" title="insertion mode: in RAWTEXT/RCDATA">in RAWTEXT/RCDATA</a>",
1.1 mike 1476: "<a href="#parsing-main-intable" title="insertion mode: in table">in table</a>", "<a href="#parsing-main-intabletext" title="insertion mode: in table text">in table text</a>", "<a href="#parsing-main-incaption" title="insertion mode: in caption">in caption</a>", "<a href="#parsing-main-incolgroup" title="insertion mode: in column group">in column group</a>",
1477: "<a href="#parsing-main-intbody" title="insertion mode: in table body">in table body</a>",
1478: "<a href="#parsing-main-intr" title="insertion mode: in row">in row</a>", "<a href="#parsing-main-intd" title="insertion mode: in cell">in cell</a>", "<a href="#parsing-main-inselect" title="insertion mode: in select">in select</a>", "<a href="#parsing-main-inselectintable" title="insertion mode: in select in table">in select in
1479: table</a>", "<a href="#parsing-main-inforeign" title="insertion mode: in foreign content">in
1480: foreign content</a>", "<a href="#parsing-main-afterbody" title="insertion mode: after
1481: body">after body</a>", "<a href="#parsing-main-inframeset" title="insertion mode: in
1482: frameset">in frameset</a>", "<a href="#parsing-main-afterframeset" title="insertion mode: after
1483: frameset">after frameset</a>", "<a href="#the-after-after-body-insertion-mode" title="insertion mode:
1484: after after body">after after body</a>", and "<a href="#the-after-after-frameset-insertion-mode" title="insertion mode: after after frameset">after after
1485: frameset</a>" during the course of the parsing, as described in
1486: the <a href="#tree-construction">tree construction</a> stage. The insertion mode affects
1487: how tokens are processed and whether CDATA sections are
1488: supported.</p>
1489:
1490: <p>Seven of these modes, namely "<a href="#parsing-main-inhead" title="insertion mode: in
1491: head">in head</a>", "<a href="#parsing-main-inbody" title="insertion mode: in body">in
1492: body</a>", "<a href="#parsing-main-intable" title="insertion mode: in table">in
1493: table</a>", "<a href="#parsing-main-intbody" title="insertion mode: in table body">in table
1494: body</a>", "<a href="#parsing-main-intr" title="insertion mode: in row">in row</a>",
1495: "<a href="#parsing-main-intd" title="insertion mode: in cell">in cell</a>", and "<a href="#parsing-main-inselect" title="insertion mode: in select">in select</a>", are special, in
1496: that the other modes defer to them at various times. When the
1497: algorithm below says that the user agent is to do something
1498: "<dfn id="using-the-rules-for">using the rules for</dfn> the <var title="">m</var> insertion
1499: mode", where <var title="">m</var> is one of these modes, the user
1500: agent must use the rules described under the <var title="">m</var>
1501: <a href="#insertion-mode">insertion mode</a>'s section, but must leave the
1502: <a href="#insertion-mode">insertion mode</a> unchanged unless the rules in <var title="">m</var> themselves switch the <a href="#insertion-mode">insertion mode</a>
1503: to a new value.</p>
1504:
1505: <p>When the insertion mode is switched to "<a href="#parsing-main-incdata" title="insertion
1.91 mike 1506: mode: in RAWTEXT/RCDATA">in RAWTEXT/RCDATA</a>" or "<a href="#parsing-main-intabletext" title="insertion mode: in table text">in table text</a>", the
1.1 mike 1507: <dfn id="original-insertion-mode">original insertion mode</dfn> is also set. This is the
1508: insertion mode to which the tree construction stage will return.</p>
1509:
1510: <p>When the insertion mode is switched to "<a href="#parsing-main-inforeign" title="insertion
1511: mode: in foreign content">in foreign content</a>", the
1512: <dfn id="secondary-insertion-mode">secondary insertion mode</dfn> is also set. This secondary mode
1513: is used within the rules for the "<a href="#parsing-main-inforeign" title="insertion mode: in
1514: foreign content">in foreign content</a>" mode to handle HTML
1515: (i.e. not foreign) content.</p>
1516:
1517: <hr><p>When the steps below require the UA to <dfn id="reset-the-insertion-mode-appropriately">reset the insertion
1518: mode appropriately</dfn>, it means the UA must follow these
1519: steps:</p>
1520:
1521: <ol><li>Let <var title="">last</var> be false.</li>
1522:
1523: <li>Let <var title="">node</var> be the last node in the
1524: <a href="#stack-of-open-elements">stack of open elements</a>.</li>
1525:
1526: <li>If <var title="">node</var> is the first node in the stack of
1527: open elements, then set <var title="">last</var> to true and set
1528: <var title="">node</var> to the <var title="">context</var>
1529: element. (<a href="#fragment-case">fragment case</a>)</li>
1530:
1531: <li>If <var title="">node</var> is a <code><a href="forms.html#the-select-element">select</a></code> element,
1532: then switch the <a href="#insertion-mode">insertion mode</a> to "<a href="#parsing-main-inselect" title="insertion mode: in select">in select</a>" and abort these
1533: steps. (<a href="#fragment-case">fragment case</a>)</li>
1534:
1535: <li>If <var title="">node</var> is a <code><a href="tabular-data.html#the-td-element">td</a></code> or
1536: <code><a href="tabular-data.html#the-th-element">th</a></code> element and <var title="">last</var> is false, then
1537: switch the <a href="#insertion-mode">insertion mode</a> to "<a href="#parsing-main-intd" title="insertion
1538: mode: in cell">in cell</a>" and abort these steps.</li>
1539:
1540: <li>If <var title="">node</var> is a <code><a href="tabular-data.html#the-tr-element">tr</a></code> element, then
1541: switch the <a href="#insertion-mode">insertion mode</a> to "<a href="#parsing-main-intr" title="insertion
1542: mode: in row">in row</a>" and abort these steps.</li>
1543:
1544: <li>If <var title="">node</var> is a <code><a href="tabular-data.html#the-tbody-element">tbody</a></code>,
1545: <code><a href="tabular-data.html#the-thead-element">thead</a></code>, or <code><a href="tabular-data.html#the-tfoot-element">tfoot</a></code> element, then switch the
1546: <a href="#insertion-mode">insertion mode</a> to "<a href="#parsing-main-intbody" title="insertion mode: in
1547: table body">in table body</a>" and abort these steps.</li>
1548:
1549: <li>If <var title="">node</var> is a <code><a href="tabular-data.html#the-caption-element">caption</a></code> element,
1550: then switch the <a href="#insertion-mode">insertion mode</a> to "<a href="#parsing-main-incaption" title="insertion mode: in caption">in caption</a>" and abort
1551: these steps.</li>
1552:
1553: <li>If <var title="">node</var> is a <code><a href="tabular-data.html#the-colgroup-element">colgroup</a></code> element,
1554: then switch the <a href="#insertion-mode">insertion mode</a> to "<a href="#parsing-main-incolgroup" title="insertion mode: in column group">in column group</a>" and
1555: abort these steps. (<a href="#fragment-case">fragment case</a>)</li>
1556:
1557: <li>If <var title="">node</var> is a <code><a href="tabular-data.html#the-table-element">table</a></code> element,
1558: then switch the <a href="#insertion-mode">insertion mode</a> to "<a href="#parsing-main-intable" title="insertion mode: in table">in table</a>" and abort these
1559: steps.</li>
1560:
1561: <li>If <var title="">node</var> is an element from the <a href="#mathml-namespace">MathML
1562: namespace</a> or the <a href="#svg-namespace">SVG namespace</a>, then switch the
1563: <a href="#insertion-mode">insertion mode</a> to "<a href="#parsing-main-inforeign" title="insertion mode: in
1564: foreign content">in foreign content</a>", let the
1565: <a href="#secondary-insertion-mode">secondary insertion mode</a> be "<a href="#parsing-main-inbody" title="insertion
1566: mode: in body">in body</a>", and abort these steps.</li>
1567:
1.159 mike 1568: <li>If <var title="">node</var> is a <code><a href="semantics.html#the-head-element-0">head</a></code> element,
1.1 mike 1569: then switch the <a href="#insertion-mode">insertion mode</a> to "<a href="#parsing-main-inbody" title="insertion mode: in body">in body</a>" ("<a href="#parsing-main-inbody" title="insertion mode: in body">in body</a>"! <em> not "<a href="#parsing-main-inhead" title="insertion mode: in head">in head</a>"</em>!) and abort
1570: these steps. (<a href="#fragment-case">fragment case</a>)</li> <!-- This is only
1571: here for now in case people think that the spec accidentally
1572: omitted it and try to "fix" it. Note that noscript-in-head is also
1573: handled this way. This is all intentional. The only thing it
1574: doesn't handle is the scripting-disabled fragment parsing case for
1575: a <head> element containing a <noscript> which itself contains
1576: something other than a <link> or a <style> element; you'd expect
1577: that to break out of the <noscript> but it doesn't. This is an edge
1578: case that doesn't affect the spec, since the algorithm for fragment
1579: parsing is only used for innerHTML/outerHTML/insertAdjacentHTML(),
1580: where we know scripting is enabled. (XXX except maybe if innerHTML
1581: is set from another browsing context on a document with designMode
1582: set?) -->
1583:
1.159 mike 1584: <li>If <var title="">node</var> is a <code><a href="semantics.html#the-body-element-0">body</a></code> element,
1.1 mike 1585: then switch the <a href="#insertion-mode">insertion mode</a> to "<a href="#parsing-main-inbody" title="insertion mode: in body">in body</a>" and abort these
1586: steps.</li>
1587:
1.5 mike 1588: <li>If <var title="">node</var> is a <code><a href="obsolete.html#frameset">frameset</a></code> element,
1.1 mike 1589: then switch the <a href="#insertion-mode">insertion mode</a> to "<a href="#parsing-main-inframeset" title="insertion mode: in frameset">in frameset</a>" and abort
1590: these steps. (<a href="#fragment-case">fragment case</a>)</li>
1591:
1.159 mike 1592: <li>If <var title="">node</var> is an <code><a href="semantics.html#the-html-element-0">html</a></code> element,
1.1 mike 1593: then: if the <a href="#head-element-pointer"><code title="">head</code> element
1594: pointer</a> is null, switch the <a href="#insertion-mode">insertion mode</a> to
1595: "<a href="#the-before-head-insertion-mode" title="insertion mode: before head">before head</a>",
1596: otherwise, switch the <a href="#insertion-mode">insertion mode</a> to "<a href="#the-after-head-insertion-mode" title="insertion mode: after head">after head</a>". In either
1597: case, abort these steps. (<a href="#fragment-case">fragment case</a>)</li> <!-- XXX
1598: can the head element pointer ever be non-null when we're going
1599: through these steps? -->
1600:
1601: <li>If <var title="">last</var> is true, then switch the
1602: <a href="#insertion-mode">insertion mode</a> to "<a href="#parsing-main-inbody" title="insertion mode: in
1603: body">in body</a>" and abort these steps. (<a href="#fragment-case">fragment
1604: case</a>)</li>
1605:
1606: <li>Let <var title="">node</var> now be the node before <var title="">node</var> in the <a href="#stack-of-open-elements">stack of open
1607: elements</a>.</li>
1608:
1609: <li>Return to step 3.</li>
1610:
1611: </ol><h5 id="the-stack-of-open-elements"><span class="secno">9.2.3.2 </span>The stack of open elements</h5>
1612:
1613: <p>Initially, the <dfn id="stack-of-open-elements">stack of open elements</dfn> is empty. The
1614: stack grows downwards; the topmost node on the stack is the first
1615: one added to the stack, and the bottommost node of the stack is the
1616: most recently added node in the stack (notwithstanding when the
1617: stack is manipulated in a random access fashion as part of <a href="#adoptionAgency">the handling for misnested tags</a>).</p>
1618:
1619: <p>The "<a href="#the-before-html-insertion-mode" title="insertion mode: before html">before
1620: html</a>" <a href="#insertion-mode">insertion mode</a> creates the
1.159 mike 1621: <code><a href="semantics.html#the-html-element-0">html</a></code> root element node, which is then added to the
1.1 mike 1622: stack.</p>
1623:
1624: <p>In the <a href="#fragment-case">fragment case</a>, the <a href="#stack-of-open-elements">stack of open
1.159 mike 1625: elements</a> is initialized to contain an <code><a href="semantics.html#the-html-element-0">html</a></code>
1.1 mike 1626: element that is created as part of <a href="#html-fragment-parsing-algorithm" title="html fragment
1627: parsing algorithm">that algorithm</a>. (The <a href="#fragment-case">fragment
1628: case</a> skips the "<a href="#the-before-html-insertion-mode" title="insertion mode: before
1629: html">before html</a>" <a href="#insertion-mode">insertion mode</a>.)</p>
1630:
1.159 mike 1631: <p>The <code><a href="semantics.html#the-html-element-0">html</a></code> node, however it is created, is the topmost
1.1 mike 1632: node of the stack. It never gets popped off the stack.</p>
1633:
1634: <p>The <dfn id="current-node">current node</dfn> is the bottommost node in this
1635: stack.</p>
1636:
1637: <p>The <dfn id="current-table">current table</dfn> is the last <code><a href="tabular-data.html#the-table-element">table</a></code>
1638: element in the <a href="#stack-of-open-elements">stack of open elements</a>, if there is
1639: one. If there is no <code><a href="tabular-data.html#the-table-element">table</a></code> element in the <a href="#stack-of-open-elements">stack of
1640: open elements</a> (<a href="#fragment-case">fragment case</a>), then the
1641: <a href="#current-table">current table</a> is the first element in the <a href="#stack-of-open-elements">stack
1.159 mike 1642: of open elements</a> (the <code><a href="semantics.html#the-html-element-0">html</a></code> element).</p>
1.1 mike 1643:
1644: <p>Elements in the stack fall into the following categories:</p>
1645:
1646: <dl><dt><dfn id="special">Special</dfn></dt>
1647:
1648: <dd><p>The following HTML elements have varying levels of special
1649: parsing rules: <code><a href="semantics.html#the-address-element">address</a></code>, <code><a href="the-canvas-element.html#the-area-element">area</a></code>,
1650: <code><a href="semantics.html#the-article-element">article</a></code>, <code><a href="semantics.html#the-aside-element">aside</a></code>, <code><a href="semantics.html#the-base-element">base</a></code>,
1.5 mike 1651: <code><a href="obsolete.html#basefont">basefont</a></code>, <code>bgsound</code>,
1.159 mike 1652: <code><a href="semantics.html#the-blockquote-element">blockquote</a></code>, <code><a href="semantics.html#the-body-element-0">body</a></code>, <code><a href="semantics.html#the-br-element">br</a></code>,
1.5 mike 1653: <code><a href="obsolete.html#center">center</a></code>, <code><a href="tabular-data.html#the-col-element">col</a></code>, <code><a href="tabular-data.html#the-colgroup-element">colgroup</a></code>,
1.86 mike 1654: <code><a href="interactive-elements.html#the-command">command</a></code>, <!--v2DDATAGRID--><code>datagrid</code>,
1655: <code><a href="semantics.html#the-dd-element">dd</a></code>, <code><a href="interactive-elements.html#the-details-element">details</a></code>, <code><a href="semantics.html#the-dialog-element">dialog</a></code>,
1656: <code><a href="obsolete.html#dir">dir</a></code>, <code><a href="interactive-elements.html#the-div-element">div</a></code>, <code><a href="semantics.html#the-dl-element">dl</a></code>,
1.159 mike 1657: <code><a href="semantics.html#the-dt-element">dt</a></code>, <code><a href="text-level-semantics.html#the-embed-element">embed</a></code>, <code><a href="forms.html#the-fieldset-element">fieldset</a></code>,
1658: <code><a href="text-level-semantics.html#the-figure-element">figure</a></code>, <code><a href="semantics.html#the-footer-element">footer</a></code>, <code><a href="forms.html#the-form-element">form</a></code>,
1.86 mike 1659: <code><a href="obsolete.html#frame">frame</a></code>, <code><a href="obsolete.html#frameset">frameset</a></code>, <code><a href="semantics.html#the-h1-h2-h3-h4-h5-and-h6-elements">h1</a></code>,
1660: <code><a href="semantics.html#the-h1-h2-h3-h4-h5-and-h6-elements">h2</a></code>, <code><a href="semantics.html#the-h1-h2-h3-h4-h5-and-h6-elements">h3</a></code>, <code><a href="semantics.html#the-h1-h2-h3-h4-h5-and-h6-elements">h4</a></code>, <code><a href="semantics.html#the-h1-h2-h3-h4-h5-and-h6-elements">h5</a></code>,
1.159 mike 1661: <code><a href="semantics.html#the-h1-h2-h3-h4-h5-and-h6-elements">h6</a></code>, <code><a href="semantics.html#the-head-element-0">head</a></code>, <code><a href="semantics.html#the-header-element">header</a></code>,
1662: <code><a href="semantics.html#the-hgroup-element">hgroup</a></code>, <code><a href="semantics.html#the-hr-element">hr</a></code>, <code><a href="text-level-semantics.html#the-iframe-element">iframe</a></code>, <!--
1.86 mike 1663: <code>image</code>, (commented out because this isn't an element
1664: that can end up on the stack, so it doesn't matter) -->
1.159 mike 1665: <code><a href="text-level-semantics.html#the-img-element">img</a></code>, <code><a href="forms.html#the-input-element">input</a></code>, <code><a href="obsolete.html#isindex-0">isindex</a></code>,
1.86 mike 1666: <code><a href="semantics.html#the-li-element">li</a></code>, <code><a href="semantics.html#the-link-element">link</a></code>, <code><a href="obsolete.html#listing">listing</a></code>,
1667: <code><a href="interactive-elements.html#menus">menu</a></code>, <code><a href="semantics.html#meta">meta</a></code>, <code><a href="semantics.html#the-nav-element">nav</a></code>,
1668: <code><a href="obsolete.html#noembed">noembed</a></code>, <code><a href="obsolete.html#noframes">noframes</a></code>, <code><a href="semantics.html#the-noscript-element">noscript</a></code>,
1.159 mike 1669: <code><a href="semantics.html#the-ol-element">ol</a></code>, <code><a href="semantics.html#the-p-element">p</a></code>, <code><a href="text-level-semantics.html#the-param-element">param</a></code>,
1.86 mike 1670: <code><a href="obsolete.html#plaintext">plaintext</a></code>, <code><a href="semantics.html#the-pre-element">pre</a></code>, <code><a href="semantics.html#script">script</a></code>,
1671: <code><a href="semantics.html#the-section-element">section</a></code>, <code><a href="forms.html#the-select-element">select</a></code>, <code><a href="obsolete.html#spacer">spacer</a></code>,
1672: <code><a href="semantics.html#the-style-element">style</a></code>, <code><a href="tabular-data.html#the-tbody-element">tbody</a></code>, <code><a href="forms.html#the-textarea-element">textarea</a></code>,
1.159 mike 1673: <code><a href="tabular-data.html#the-tfoot-element">tfoot</a></code>, <code><a href="tabular-data.html#the-thead-element">thead</a></code>, <code><a href="semantics.html#the-title-element-0">title</a></code>,
1.174 mike 1674: <code><a href="tabular-data.html#the-tr-element">tr</a></code>, <code><a href="semantics.html#the-ul-element">ul</a></code>, <code>wbr</code>, and
1675: <code><a href="obsolete.html#xmp">xmp</a></code>.</p></dd>
1.1 mike 1676:
1677: <dt><dfn id="scoping">Scoping</dfn></dt>
1678: <dd><p>The following HTML elements introduce new <a href="#has-an-element-in-scope" title="has
1679: an element in scope">scopes</a> for various parts of the
1.5 mike 1680: parsing: <code><a href="obsolete.html#the-applet-element">applet</a></code>, <code><a href="forms.html#the-button-element">button</a></code>,
1.159 mike 1681: <code><a href="tabular-data.html#the-caption-element">caption</a></code>, <code><a href="semantics.html#the-html-element-0">html</a></code>, <code><a href="obsolete.html#the-marquee-element">marquee</a></code>,
1682: <code><a href="text-level-semantics.html#the-object-element">object</a></code>, <code><a href="tabular-data.html#the-table-element">table</a></code>, <code><a href="tabular-data.html#the-td-element">td</a></code>,
1.1 mike 1683: <code><a href="tabular-data.html#the-th-element">th</a></code>, and SVG's <code>foreignObject</code>.</p></dd>
1684:
1685: <dt><dfn id="formatting">Formatting</dfn></dt>
1686: <dd><p>The following HTML elements are those that end up in the
1687: <a href="#list-of-active-formatting-elements">list of active formatting elements</a>: <code><a href="text-level-semantics.html#the-a-element">a</a></code>,
1.5 mike 1688: <code><a href="text-level-semantics.html#the-b-element">b</a></code>, <code><a href="obsolete.html#big">big</a></code>, <code><a href="text-level-semantics.html#the-code-element">code</a></code>,
1689: <code><a href="text-level-semantics.html#the-em-element">em</a></code>, <code><a href="obsolete.html#font">font</a></code>, <code><a href="text-level-semantics.html#the-i-element">i</a></code>,
1690: <code>nobr</code>, <code><a href="obsolete.html#s">s</a></code>, <code><a href="text-level-semantics.html#the-small-element">small</a></code>,
1691: <code><a href="obsolete.html#strike">strike</a></code>, <code><a href="text-level-semantics.html#the-strong-element">strong</a></code>, <code><a href="obsolete.html#tt">tt</a></code>, and
1692: <code><a href="obsolete.html#u">u</a></code>.</p></dd>
1.1 mike 1693:
1694: <dt><dfn id="phrasing">Phrasing</dfn></dt>
1695: <dd><p>All other elements found while parsing an HTML
1696: document.</p></dd>
1697:
1.234 mike 1698: </dl><p>The <a href="#stack-of-open-elements">stack of open elements</a> is said to <dfn id="has-an-element-in-the-specific-scope" title="has an element in the specific scope">have an element in a
1699: specific scope</dfn> consisting of a list of element types <var title="">list</var> when the following algorithm terminates in a
1700: match state:</p>
1.1 mike 1701:
1702: <ol><li><p>Initialize <var title="">node</var> to be the <a href="#current-node">current
1703: node</a> (the bottommost node of the stack).</p></li>
1704:
1705: <li><p>If <var title="">node</var> is the target node, terminate in
1706: a match state.</p></li>
1707:
1.234 mike 1708: <li><p>Otherwise, if <var title="">node</var> is one of the element
1709: types in <var title="">list</var>, terminate in a failure
1710: state.</p></li>
1.1 mike 1711:
1712: <li><p>Otherwise, set <var title="">node</var> to the previous
1713: entry in the <a href="#stack-of-open-elements">stack of open elements</a> and return to step
1714: 2. (This will never fail, since the loop will always terminate in
1715: the previous step if the top of the stack — an
1.159 mike 1716: <code><a href="semantics.html#the-html-element-0">html</a></code> element — is reached.)</p></li>
1.1 mike 1717:
1.234 mike 1718: </ol><p>The <a href="#stack-of-open-elements">stack of open elements</a> is said to <dfn id="has-an-element-in-scope" title="has an element in scope">have an element in scope</dfn> when
1719: it <a href="#has-an-element-in-the-specific-scope">has an element in the specific scope</a> consisting
1720: of the following element types:</p>
1721:
1722: <ul class="brief"><li><code><a href="obsolete.html#the-applet-element">applet</a></code> in the HTML namespace</li>
1723: <li><code><a href="tabular-data.html#the-caption-element">caption</a></code> in the HTML namespace</li>
1724: <li><code><a href="semantics.html#the-html-element-0">html</a></code> in the HTML namespace</li> <!-- (This can only happen if the <var title="">node</var> is the topmost node of the <span>stack of open elements</span>, and prevents the next step from being invoked if there are no more elements in the stack.) -->
1725: <li><code><a href="tabular-data.html#the-table-element">table</a></code> in the HTML namespace</li>
1726: <li><code><a href="tabular-data.html#the-td-element">td</a></code> in the HTML namespace</li>
1727: <li><code><a href="tabular-data.html#the-th-element">th</a></code> in the HTML namespace</li>
1728: <li><code><a href="forms.html#the-button-element">button</a></code> in the HTML namespace</li>
1729: <li><code><a href="obsolete.html#the-marquee-element">marquee</a></code> in the HTML namespace</li>
1730: <li><code><a href="text-level-semantics.html#the-object-element">object</a></code> in the HTML namespace</li>
1731: <li><code title="">foreignObject</code> in the SVG namespace</li>
1732: </ul><p>The <a href="#stack-of-open-elements">stack of open elements</a> is said to <dfn id="has-an-element-in-list-item-scope" title="has an element in list item scope">have an element in list
1733: item scope</dfn> when it <a href="#has-an-element-in-the-specific-scope">has an element in the specific
1734: scope</a> consisting of the following element types:</p>
1735:
1736: <ul class="brief"><li>All the element types listed above for the <i><a href="#has-an-element-in-scope">has an element
1737: in scope</a></i> algorithm.</li>
1738: <li><code><a href="semantics.html#the-ol-element">ol</a></code> in the HTML namespace</li>
1739: <li><code><a href="semantics.html#the-ul-element">ul</a></code> in the HTML namespace</li>
1740: </ul><p>The <a href="#stack-of-open-elements">stack of open elements</a> is said to <dfn id="has-an-element-in-table-scope" title="has an element in table scope">have an element in <em>table
1741: scope</em></dfn> when it <a href="#has-an-element-in-the-specific-scope">has an element in the specific
1742: scope</a> consisting of the following element types:</p>
1743:
1744: <ul class="brief"><li><code><a href="semantics.html#the-html-element-0">html</a></code> in the HTML namespace</li> <!-- (This can only happen if the <var title="">node</var> is the topmost node of the <span>stack of open elements</span>, and prevents the next step from being invoked if there are no more elements in the stack.) -->
1745: <li><code><a href="tabular-data.html#the-table-element">table</a></code> in the HTML namespace</li>
1746: </ul><p>Nothing happens if at any time any of the elements in the
1.1 mike 1747: <a href="#stack-of-open-elements">stack of open elements</a> are moved to a new location in,
1748: or removed from, the <code>Document</code> tree. In particular, the
1749: stack is not changed in this situation. This can cause, amongst
1750: other strange effects, content to be appended to nodes that are no
1751: longer in the DOM.</p>
1752:
1753: <p class="note">In some cases (namely, when <a href="#adoptionAgency">closing misnested formatting elements</a>),
1754: the stack is manipulated in a random-access fashion.</p>
1755:
1756:
1757: <h5 id="the-list-of-active-formatting-elements"><span class="secno">9.2.3.3 </span>The list of active formatting elements</h5>
1758:
1759: <p>Initially, the <dfn id="list-of-active-formatting-elements">list of active formatting elements</dfn> is
1760: empty. It is used to handle mis-nested <a href="#formatting" title="formatting">formatting element tags</a>.</p>
1761:
1762: <p>The list contains elements in the <a href="#formatting">formatting</a>
1763: category, and scope markers. The scope markers are inserted when
1.159 mike 1764: entering <code><a href="obsolete.html#the-applet-element">applet</a></code> elements, buttons, <code><a href="text-level-semantics.html#the-object-element">object</a></code>
1.1 mike 1765: elements, marquees, table cells, and table captions, and are used to
1.5 mike 1766: prevent formatting from "leaking" <em>into</em> <code><a href="obsolete.html#the-applet-element">applet</a></code>
1.159 mike 1767: elements, buttons, <code><a href="text-level-semantics.html#the-object-element">object</a></code> elements, marquees, and
1.1 mike 1768: tables.</p>
1769:
1770: <p class="note">The scope markers are unrelated to the concept of an
1771: element being <a href="#has-an-element-in-scope" title="has an element in scope">in
1772: scope</a>.</p>
1773:
1774: <p>In addition, each element in the <a href="#list-of-active-formatting-elements">list of active formatting
1775: elements</a> is associated with the token for which it was
1776: created, so that further elements can be created for that token if
1777: necessary.</p>
1778:
1779: <p>When the steps below require the UA to <dfn id="reconstruct-the-active-formatting-elements">reconstruct the
1780: active formatting elements</dfn>, the UA must perform the following
1781: steps:</p>
1782:
1783: <ol><li>If there are no entries in the <a href="#list-of-active-formatting-elements">list of active formatting
1784: elements</a>, then there is nothing to reconstruct; stop this
1785: algorithm.</li>
1786:
1787: <li>If the last (most recently added) entry in the <a href="#list-of-active-formatting-elements">list of
1788: active formatting elements</a> is a marker, or if it is an
1789: element that is in the <a href="#stack-of-open-elements">stack of open elements</a>, then
1790: there is nothing to reconstruct; stop this algorithm.</li>
1791:
1792: <li>Let <var title="">entry</var> be the last (most recently added)
1793: element in the <a href="#list-of-active-formatting-elements">list of active formatting
1794: elements</a>.</li>
1795:
1796: <li>If there are no entries before <var title="">entry</var> in the
1797: <a href="#list-of-active-formatting-elements">list of active formatting elements</a>, then jump to step
1798: 8.</li>
1799:
1800: <li>Let <var title="">entry</var> be the entry one earlier than
1801: <var title="">entry</var> in the <a href="#list-of-active-formatting-elements">list of active formatting
1802: elements</a>.</li>
1803:
1804: <li>If <var title="">entry</var> is neither a marker nor an element
1805: that is also in the <a href="#stack-of-open-elements">stack of open elements</a>, go to step
1806: 4.</li>
1807:
1808: <li>Let <var title="">entry</var> be the element one later than
1809: <var title="">entry</var> in the <a href="#list-of-active-formatting-elements">list of active formatting
1810: elements</a>.</li>
1811:
1812: <li><a href="#create-an-element-for-the-token">Create an element for the token</a> for which the
1813: element <var title="">entry</var> was created, to obtain <var title="">new element</var>.</li>
1814:
1815: <li>Append <var title="">new element</var> to the <a href="#current-node">current
1816: node</a> and push it onto the <a href="#stack-of-open-elements">stack of open
1817: elements</a> so that it is the new <a href="#current-node">current
1818: node</a>.</li>
1819:
1820: <li>Replace the entry for <var title="">entry</var> in the list
1821: with an entry for <var title="">new element</var>.</li>
1822:
1823: <li>If the entry for <var title="">new element</var> in the
1824: <a href="#list-of-active-formatting-elements">list of active formatting elements</a> is not the last
1825: entry in the list, return to step 7.</li>
1826:
1827: </ol><p>This has the effect of reopening all the formatting elements that
1828: were opened in the current body, cell, or caption (whichever is
1829: youngest) that haven't been explicitly closed.</p>
1830:
1831: <p class="note">The way this specification is written, the
1832: <a href="#list-of-active-formatting-elements">list of active formatting elements</a> always consists of
1833: elements in chronological order with the least recently added
1834: element first and the most recently added element last (except for
1835: while steps 8 to 11 of the above algorithm are being executed, of
1836: course).</p>
1837:
1838: <p>When the steps below require the UA to <dfn id="clear-the-list-of-active-formatting-elements-up-to-the-last-marker">clear the list of
1839: active formatting elements up to the last marker</dfn>, the UA must
1840: perform the following steps:</p>
1841:
1842: <ol><li>Let <var title="">entry</var> be the last (most recently added)
1843: entry in the <a href="#list-of-active-formatting-elements">list of active formatting elements</a>.</li>
1844:
1845: <li>Remove <var title="">entry</var> from the <a href="#list-of-active-formatting-elements">list of active
1846: formatting elements</a>.</li>
1847:
1848: <li>If <var title="">entry</var> was a marker, then stop the
1849: algorithm at this point. The list has been cleared up to the last
1850: marker.</li>
1851:
1852: <li>Go to step 1.</li>
1853:
1854: </ol><h5 id="the-element-pointers"><span class="secno">9.2.3.4 </span>The element pointers</h5>
1855:
1856: <p>Initially, the <dfn id="head-element-pointer"><code title="">head</code> element
1857: pointer</dfn> and the <dfn id="form-element-pointer"><code title="">form</code> element
1858: pointer</dfn> are both null.</p>
1859:
1.159 mike 1860: <p>Once a <code><a href="semantics.html#the-head-element-0">head</a></code> element has been parsed (whether
1.1 mike 1861: implicitly or explicitly) the <a href="#head-element-pointer"><code title="">head</code>
1862: element pointer</a> gets set to point to this node.</p>
1863:
1864: <p>The <a href="#form-element-pointer"><code title="">form</code> element pointer</a>
1865: points to the last <code><a href="forms.html#the-form-element">form</a></code> element that was opened and
1866: whose end tag has not yet been seen. It is used to make form
1867: controls associate with forms in the face of dramatically bad
1868: markup, for historical reasons.</p>
1869:
1870:
1871: <h5 id="other-parsing-state-flags"><span class="secno">9.2.3.5 </span>Other parsing state flags</h5>
1872:
1873: <p>The <dfn id="scripting-flag">scripting flag</dfn> is set to "enabled" if <a href="browsers.html#concept-n-script" title="concept-n-script">scripting was enabled</a> for the
1874: <code>Document</code> with which the parser is associated when the
1875: parser was created, and "disabled" otherwise.</p>
1876:
1877: <p class="note">The <a href="#scripting-flag">scripting flag</a> can be enabled even
1878: when the parser was originally created for the <a href="#html-fragment-parsing-algorithm">HTML fragment
1879: parsing algorithm</a>, even though <code><a href="semantics.html#script">script</a></code> elements
1880: don't execute in that case.</p>
1881:
1882: <p>The <dfn id="frameset-ok-flag">frameset-ok flag</dfn> is set to "ok" when the parser is
1883: created. It is set to "not ok" after certain tokens are seen.</p>
1884:
1885:
1.159 mike 1886: <h4 id="tokenization"><span class="secno">9.2.4 </span><dfn>Tokenization</dfn></h4><p class="XXX annotation"><b>Status: </b><i>Last call for comments</i></p>
1.1 mike 1887:
1888: <p>Implementations must act as if they used the following state
1889: machine to tokenize HTML. The state machine must start in the
1890: <a href="#data-state">data state</a>. Most states consume a single character,
1891: which may have various side-effects, and either switches the state
1892: machine to a new state to <em>reconsume</em> the same character, or
1893: switches it to a new state (to consume the next character), or
1894: repeats the same state (to consume the next character). Some states
1895: have more complicated behavior and can consume several characters
1896: before switching to another state.</p>
1897:
1898: <p>The exact behavior of certain states depends on a <dfn id="content-model-flag">content
1899: model flag</dfn> that is set after certain tokens are emitted. The
1.91 mike 1900: flag has several states: <i title="">PCDATA</i>, <i title="">RCDATA</i>, <i title="">RAWTEXT</i>, and <i title="">PLAINTEXT</i>. Initially, it must be in the PCDATA
1901: state. In the RCDATA and RAWTEXT states, a further <dfn id="escape-flag">escape
1.1 mike 1902: flag</dfn> is used to control the behavior of the tokenizer. It is
1903: either true or false, and initially must be set to the false
1904: state. The <a href="#insertion-mode">insertion mode</a> and the <a href="#stack-of-open-elements">stack of open
1905: elements</a> also affects tokenization.</p>
1906:
1907: <p>The output of the tokenization step is a series of zero or more
1908: of the following tokens: DOCTYPE, start tag, end tag, comment,
1909: character, end-of-file. DOCTYPE tokens have a name, a public
1910: identifier, a system identifier, and a <i>force-quirks
1911: flag</i>. When a DOCTYPE token is created, its name, public
1912: identifier, and system identifier must be marked as missing (which
1913: is a distinct state from the empty string), and the <i>force-quirks
1914: flag</i> must be set to <i>off</i> (its other state is
1915: <i>on</i>). Start and end tag tokens have a tag name, a
1916: <i>self-closing flag</i>, and a list of attributes, each of which
1917: has a name and a value. When a start or end tag token is created,
1918: its <i>self-closing flag</i> must be unset (its other state is that
1919: it be set), and its attributes list must be empty. Comment and
1920: character tokens have data.</p>
1921:
1922: <p>When a token is emitted, it must immediately be handled by the
1923: <a href="#tree-construction">tree construction</a> stage. The tree construction stage
1924: can affect the state of the <a href="#content-model-flag">content model flag</a>, and can
1925: insert additional characters into the stream. (For example, the
1926: <code><a href="semantics.html#script">script</a></code> element can result in scripts executing and
1.159 mike 1927: using the <a href="embedded-content-0.html#dynamic-markup-insertion">dynamic markup insertion</a> APIs to insert
1.1 mike 1928: characters into the stream being tokenized.)</p>
1929:
1930: <p>When a start tag token is emitted with its <i>self-closing
1931: flag</i> set, if the flag is not <dfn id="acknowledge-self-closing-flag" title="acknowledge
1932: self-closing flag">acknowledged</dfn> when it is processed by the
1933: tree construction stage, that is a <a href="#parse-error">parse error</a>.</p>
1934:
1935: <p>When an end tag token is emitted, the <a href="#content-model-flag">content model
1936: flag</a> must be switched to the PCDATA state.</p>
1937:
1938: <p>When an end tag token is emitted with attributes, that is a
1939: <a href="#parse-error">parse error</a>.</p>
1940:
1941: <p>When an end tag token is emitted with its <i>self-closing
1942: flag</i> set, that is a <a href="#parse-error">parse error</a>.</p>
1943:
1944: <p>Before each step of the tokenizer, the user agent must first
1945: check the <a href="#parser-pause-flag">parser pause flag</a>. If it is true, then the
1946: tokenizer must abort the processing of any nested invocations of the
1947: tokenizer, yielding control back to the caller. If it is false, then
1948: the user agent may then check to see if either one of the scripts in
1949: the <a href="semantics.html#list-of-scripts-that-will-execute-as-soon-as-possible">list of scripts that will execute as soon as
1950: possible</a> or the first script in the <a href="semantics.html#list-of-scripts-that-will-execute-asynchronously">list of scripts
1951: that will execute asynchronously</a>, has <a href="semantics.html#completed-loading">completed
1952: loading</a>. If one has, then it must be <a href="semantics.html#executing-a-script-block" title="executing a
1953: script block">executed</a> and removed from its list.</p>
1954:
1955: <p>The tokenizer state machine consists of the states defined in the
1956: following subsections.</p>
1957:
1958: <!-- XXX should go through these reordering the entries so that
1959: they're in some consistent order, like, by Unicode, errors last, or
1960: something -->
1961:
1962: <h5 id="data-state"><span class="secno">9.2.4.1 </span><dfn>Data state</dfn></h5>
1963:
1964: <p>Consume the <a href="#next-input-character">next input character</a>:</p>
1965:
1966: <dl class="switch"><dt>U+0026 AMPERSAND (&)</dt>
1967: <dd>When the <a href="#content-model-flag">content model flag</a> is set to one of the
1968: PCDATA or RCDATA states and the <a href="#escape-flag">escape flag</a> is
1969: false: switch to the <a href="#character-reference-data-state">character reference data
1.56 mike 1970: state</a>.</dd>
1971: <dd>Otherwise: treat it as per the "anything else" entry
1972: below.</dd>
1.1 mike 1973:
1974: <dt>U+002D HYPHEN-MINUS (-)</dt>
1975: <dd>
1976:
1977: <p>If the <a href="#content-model-flag">content model flag</a> is set to either the
1.91 mike 1978: RCDATA state or the RAWTEXT state, and the <a href="#escape-flag">escape flag</a>
1.1 mike 1979: is false, and there are at least three characters before this
1980: one in the input stream, and the last four characters in the
1981: input stream, including this one, are U+003C LESS-THAN SIGN,
1982: U+0021 EXCLAMATION MARK, U+002D HYPHEN-MINUS, and U+002D
1983: HYPHEN-MINUS ("<!--"), then set the <a href="#escape-flag">escape flag</a>
1984: to true.</p>
1985:
1986: <p>In any case, emit the input character as a character
1987: token. Stay in the <a href="#data-state">data state</a>.</p>
1988:
1989: </dd>
1990:
1991: <dt>U+003C LESS-THAN SIGN (<)</dt>
1992: <dd>When the <a href="#content-model-flag">content model flag</a> is set to the PCDATA
1993: state: switch to the <a href="#tag-open-state">tag open state</a>.</dd>
1994: <dd>When the <a href="#content-model-flag">content model flag</a> is set to either the
1.91 mike 1995: RCDATA state or the RAWTEXT state, and the <a href="#escape-flag">escape flag</a>
1.1 mike 1996: is false: switch to the <a href="#tag-open-state">tag open state</a>.</dd>
1997: <dd>Otherwise: treat it as per the "anything else" entry
1998: below.</dd>
1999:
2000: <dt>U+003E GREATER-THAN SIGN (>)</dt>
2001: <dd>
2002:
2003: <p>If the <a href="#content-model-flag">content model flag</a> is set to either the
1.91 mike 2004: RCDATA state or the RAWTEXT state, and the <a href="#escape-flag">escape
1.1 mike 2005: flag</a> is true, and the last three characters in the input
2006: stream including this one are U+002D HYPHEN-MINUS, U+002D
2007: HYPHEN-MINUS, U+003E GREATER-THAN SIGN ("-->"), set the
2008: <a href="#escape-flag">escape flag</a> to false.</p> <!-- no need to check
2009: that there are enough characters, since you can only run into
2010: this if the flag is true in the first place, which requires four
2011: characters. -->
2012:
2013: <p>In any case, emit the input character as a character
2014: token. Stay in the <a href="#data-state">data state</a>.</p>
2015:
2016: </dd>
2017:
2018: <dt>EOF</dt>
2019: <dd>Emit an end-of-file token.</dd>
2020:
2021: <dt>Anything else</dt>
2022: <dd>Emit the input character as a character token. Stay in the
2023: <a href="#data-state">data state</a>.</dd>
2024:
2025: </dl><h5 id="character-reference-data-state"><span class="secno">9.2.4.2 </span><dfn>Character reference data state</dfn></h5>
2026:
2027: <p><i>(This cannot happen if the <a href="#content-model-flag">content model flag</a>
1.91 mike 2028: is set to the RAWTEXT state.)</i></p>
1.1 mike 2029:
2030: <p>Attempt to <a href="#consume-a-character-reference">consume a character reference</a>, with no
2031: <a href="#additional-allowed-character">additional allowed character</a>.</p>
2032:
2033: <p>If nothing is returned, emit a U+0026 AMPERSAND character
2034: token.</p>
2035:
2036: <p>Otherwise, emit the character token that was returned.</p>
2037:
2038: <p>Finally, switch to the <a href="#data-state">data state</a>.</p>
2039:
2040:
2041: <h5 id="tag-open-state"><span class="secno">9.2.4.3 </span><dfn>Tag open state</dfn></h5>
2042:
2043: <p>The behavior of this state depends on the <a href="#content-model-flag">content model
2044: flag</a>.</p>
2045:
2046: <dl><dt>If the <a href="#content-model-flag">content model flag</a> is set to the RCDATA
1.91 mike 2047: or RAWTEXT states</dt>
1.1 mike 2048:
2049: <dd>
2050:
2051: <p>Consume the <a href="#next-input-character">next input character</a>. If it is a
2052: U+002F SOLIDUS (/) character, switch to the <a href="#close-tag-open-state">close tag open
2053: state</a>. Otherwise, emit a U+003C LESS-THAN SIGN character
2054: token and reconsume the <a href="#current-input-character">current input character</a> in the
2055: <a href="#data-state">data state</a>.</p>
2056:
2057: </dd>
2058:
2059: <dt>If the <a href="#content-model-flag">content model flag</a> is set to the PCDATA
2060: state</dt>
2061:
2062: <dd>
2063:
2064: <p>Consume the <a href="#next-input-character">next input character</a>:</p>
2065:
2066: <dl class="switch"><dt>U+0021 EXCLAMATION MARK (!)</dt>
2067: <dd>Switch to the <a href="#markup-declaration-open-state">markup declaration open state</a>.</dd>
2068:
2069: <dt>U+002F SOLIDUS (/)</dt>
2070: <dd>Switch to the <a href="#close-tag-open-state">close tag open state</a>.</dd>
2071:
2072: <dt>U+0041 LATIN CAPITAL LETTER A through to U+005A LATIN CAPITAL LETTER Z</dt>
2073: <dd>Create a new start tag token, set its tag name to the
2074: lowercase version of the input character (add 0x0020 to the
2075: character's code point), then switch to the <a href="#tag-name-state">tag name
2076: state</a>. (Don't emit the token yet; further details will
2077: be filled in before it is emitted.)</dd>
2078:
2079: <dt>U+0061 LATIN SMALL LETTER A through to U+007A LATIN SMALL LETTER Z</dt>
2080: <dd>Create a new start tag token, set its tag name to the input
2081: character, then switch to the <a href="#tag-name-state">tag name
2082: state</a>. (Don't emit the token yet; further details will
2083: be filled in before it is emitted.)</dd>
2084:
2085: <dt>U+003E GREATER-THAN SIGN (>)</dt>
2086: <dd><a href="#parse-error">Parse error</a>. Emit a U+003C LESS-THAN SIGN
2087: character token and a U+003E GREATER-THAN SIGN character
2088: token. Switch to the <a href="#data-state">data state</a>.</dd>
2089:
2090: <dt>U+003F QUESTION MARK (?)</dt>
2091: <dd><a href="#parse-error">Parse error</a>. Switch to the <a href="#bogus-comment-state">bogus
2092: comment state</a>.</dd>
2093:
2094: <dt>Anything else</dt>
2095: <dd><a href="#parse-error">Parse error</a>. Emit a U+003C LESS-THAN SIGN
2096: character token and reconsume the <a href="#current-input-character">current input character</a> in the
2097: <a href="#data-state">data state</a>.</dd>
2098:
2099: </dl></dd>
2100:
2101: </dl><h5 id="close-tag-open-state"><span class="secno">9.2.4.4 </span><dfn>Close tag open state</dfn></h5>
2102:
2103: <p>If the <a href="#content-model-flag">content model flag</a> is set to the RCDATA or
1.91 mike 2104: RAWTEXT states but no start tag token has ever been emitted by this
1.1 mike 2105: instance of the tokenizer (<a href="#fragment-case">fragment case</a>), or, if the
1.91 mike 2106: <a href="#content-model-flag">content model flag</a> is set to the RCDATA or RAWTEXT states
1.1 mike 2107: and the next few characters do not match the tag name of the last
2108: start tag token emitted (compared in an <a href="infrastructure.html#ascii-case-insensitive">ASCII
2109: case-insensitive</a> manner), or if they do but they are not
2110: immediately followed by one of the following characters:</p>
2111:
2112: <ul class="brief"><li>U+0009 CHARACTER TABULATION</li>
2113: <li>U+000A LINE FEED (LF)</li>
2114: <li>U+000C FORM FEED (FF)</li>
2115: <!--<li>U+000D CARRIAGE RETURN (CR)</li>-->
2116: <li>U+0020 SPACE</li>
2117: <li>U+003E GREATER-THAN SIGN (>)</li>
2118: <li>U+002F SOLIDUS (/)</li>
2119: <li>EOF</li>
2120: </ul><p>...then emit a U+003C LESS-THAN SIGN character token, a U+002F
2121: SOLIDUS character token, and switch to the <a href="#data-state">data state</a>
2122: to process the <a href="#next-input-character">next input character</a>.</p>
2123:
2124: <p>Otherwise, if the <a href="#content-model-flag">content model flag</a> is set to the
2125: PCDATA state, or if the next few characters <em>do</em> match that tag
2126: name, consume the <a href="#next-input-character">next input character</a>:</p>
2127:
2128: <dl class="switch"><dt>U+0041 LATIN CAPITAL LETTER A through to U+005A LATIN CAPITAL LETTER Z</dt>
2129: <dd>Create a new end tag token, set its tag name to the lowercase
2130: version of the input character (add 0x0020 to the character's
2131: code point), then switch to the <a href="#tag-name-state">tag name
2132: state</a>. (Don't emit the token yet; further details will be
2133: filled in before it is emitted.)</dd>
2134:
2135: <dt>U+0061 LATIN SMALL LETTER A through to U+007A LATIN SMALL LETTER Z</dt>
2136: <dd>Create a new end tag token, set its tag name to the input
2137: character, then switch to the <a href="#tag-name-state">tag name state</a>. (Don't
2138: emit the token yet; further details will be filled in before it
2139: is emitted.)</dd>
2140:
2141: <dt>U+003E GREATER-THAN SIGN (>)</dt>
2142: <dd><a href="#parse-error">Parse error</a>. Switch to the <a href="#data-state">data
2143: state</a>.</dd>
2144:
2145: <dt>EOF</dt>
2146: <dd><a href="#parse-error">Parse error</a>. Emit a U+003C LESS-THAN SIGN
2147: character token and a U+002F SOLIDUS character token. Reconsume
2148: the EOF character in the <a href="#data-state">data state</a>.</dd>
2149:
2150: <dt>Anything else</dt>
2151: <dd><a href="#parse-error">Parse error</a>. Switch to the <a href="#bogus-comment-state">bogus
2152: comment state</a>.</dd>
2153:
2154: </dl><h5 id="tag-name-state"><span class="secno">9.2.4.5 </span><dfn>Tag name state</dfn></h5>
2155:
2156: <p>Consume the <a href="#next-input-character">next input character</a>:</p>
2157:
2158: <dl class="switch"><dt>U+0009 CHARACTER TABULATION</dt>
2159: <dt>U+000A LINE FEED (LF)</dt>
2160: <dt>U+000C FORM FEED (FF)</dt>
2161: <!--<dt>U+000D CARRIAGE RETURN (CR)</dt>-->
2162: <dt>U+0020 SPACE</dt>
2163: <dd>Switch to the <a href="#before-attribute-name-state">before attribute name state</a>.</dd>
2164:
2165: <dt>U+002F SOLIDUS (/)</dt>
2166: <dd>Switch to the <a href="#self-closing-start-tag-state">self-closing start tag state</a>.</dd>
2167:
2168: <dt>U+003E GREATER-THAN SIGN (>)</dt>
2169: <dd>Emit the current tag token. Switch to the <a href="#data-state">data
2170: state</a>.</dd>
2171:
2172: <dt>U+0041 LATIN CAPITAL LETTER A through to U+005A LATIN CAPITAL LETTER Z</dt>
2173: <dd>Append the lowercase version of the <a href="#current-input-character">current input character</a>
2174: (add 0x0020 to the character's code point) to the current tag
2175: token's tag name. Stay in the <a href="#tag-name-state">tag name state</a>.</dd>
2176:
2177: <dt>EOF</dt>
2178: <dd><a href="#parse-error">Parse error</a>. Reconsume the EOF character in the
2179: <a href="#data-state">data state</a>.</dd>
2180:
2181: <dt>Anything else</dt>
2182: <dd>Append the <a href="#current-input-character">current input character</a> to the current tag token's
2183: tag name. Stay in the <a href="#tag-name-state">tag name state</a>.</dd>
2184:
2185: </dl><h5 id="before-attribute-name-state"><span class="secno">9.2.4.6 </span><dfn>Before attribute name state</dfn></h5>
2186:
2187: <p>Consume the <a href="#next-input-character">next input character</a>:</p>
2188:
2189: <dl class="switch"><dt>U+0009 CHARACTER TABULATION</dt>
2190: <dt>U+000A LINE FEED (LF)</dt>
2191: <dt>U+000C FORM FEED (FF)</dt>
2192: <!--<dt>U+000D CARRIAGE RETURN (CR)</dt>-->
2193: <dt>U+0020 SPACE</dt>
2194: <dd>Stay in the <a href="#before-attribute-name-state">before attribute name state</a>.</dd>
2195:
2196: <dt>U+002F SOLIDUS (/)</dt>
2197: <dd>Switch to the <a href="#self-closing-start-tag-state">self-closing start tag state</a>.</dd>
2198:
2199: <dt>U+003E GREATER-THAN SIGN (>)</dt>
2200: <dd>Emit the current tag token. Switch to the <a href="#data-state">data
2201: state</a>.</dd>
2202:
2203: <dt>U+0041 LATIN CAPITAL LETTER A through to U+005A LATIN CAPITAL LETTER Z</dt>
2204: <dd>Start a new attribute in the current tag token. Set that
2205: attribute's name to the lowercase version of the <a href="#current-input-character">current input
2206: character</a> (add 0x0020 to the character's code point), and its
2207: value to the empty string. Switch to the <a href="#attribute-name-state">attribute name
2208: state</a>.</dd>
2209:
2210: <dt>U+0022 QUOTATION MARK (")</dt>
2211: <dt>U+0027 APOSTROPHE (')</dt>
2212: <dt>U+003C LESS-THAN SIGN (<)</dt>
2213: <dt>U+003D EQUALS SIGN (=)</dt>
2214: <dd><a href="#parse-error">Parse error</a>. Treat it as per the "anything else"
2215: entry below.</dd>
2216:
2217: <dt>EOF</dt>
2218: <dd><a href="#parse-error">Parse error</a>. Reconsume the EOF character in the
2219: <a href="#data-state">data state</a>.</dd>
2220:
2221: <dt>Anything else</dt>
2222: <dd>Start a new attribute in the current tag token. Set that
2223: attribute's name to the <a href="#current-input-character">current input character</a>, and its value to
2224: the empty string. Switch to the <a href="#attribute-name-state">attribute name
2225: state</a>.</dd>
2226:
2227: </dl><h5 id="attribute-name-state"><span class="secno">9.2.4.7 </span><dfn>Attribute name state</dfn></h5>
2228:
2229: <p>Consume the <a href="#next-input-character">next input character</a>:</p>
2230:
2231: <dl class="switch"><dt>U+0009 CHARACTER TABULATION</dt>
2232: <dt>U+000A LINE FEED (LF)</dt>
2233: <dt>U+000C FORM FEED (FF)</dt>
2234: <!--<dt>U+000D CARRIAGE RETURN (CR)</dt>-->
2235: <dt>U+0020 SPACE</dt>
2236: <dd>Switch to the <a href="#after-attribute-name-state">after attribute name state</a>.</dd>
2237:
2238: <dt>U+002F SOLIDUS (/)</dt>
2239: <dd>Switch to the <a href="#self-closing-start-tag-state">self-closing start tag state</a>.</dd>
2240:
2241: <dt>U+003D EQUALS SIGN (=)</dt>
2242: <dd>Switch to the <a href="#before-attribute-value-state">before attribute value state</a>.</dd>
2243:
2244: <dt>U+003E GREATER-THAN SIGN (>)</dt>
2245: <dd>Emit the current tag token. Switch to the <a href="#data-state">data
2246: state</a>.</dd>
2247:
2248: <dt>U+0041 LATIN CAPITAL LETTER A through to U+005A LATIN CAPITAL LETTER Z</dt>
2249: <dd>Append the lowercase version of the <a href="#current-input-character">current input character</a>
2250: (add 0x0020 to the character's code point) to the current
2251: attribute's name. Stay in the <a href="#attribute-name-state">attribute name
2252: state</a>.</dd>
2253:
2254: <dt>U+0022 QUOTATION MARK (")</dt>
2255: <dt>U+0027 APOSTROPHE (')</dt>
2256: <dt>U+003C LESS-THAN SIGN (<)</dt>
2257: <dd><a href="#parse-error">Parse error</a>. Treat it as per the "anything else"
2258: entry below.</dd>
2259:
2260: <dt>EOF</dt>
2261: <dd><a href="#parse-error">Parse error</a>. Reconsume the EOF character in the
2262: <a href="#data-state">data state</a>.</dd>
2263:
2264: <dt>Anything else</dt>
2265: <dd>Append the <a href="#current-input-character">current input character</a> to the current attribute's
2266: name. Stay in the <a href="#attribute-name-state">attribute name state</a>.</dd>
2267:
2268: </dl><p>When the user agent leaves the attribute name state (and before
2269: emitting the tag token, if appropriate), the complete attribute's
2270: name must be compared to the other attributes on the same token;
2271: if there is already an attribute on the token with the exact same
2272: name, then this is a <a href="#parse-error">parse error</a> and the new
2273: attribute must be dropped, along with the value that gets
2274: associated with it (if any).</p>
2275:
2276:
2277: <h5 id="after-attribute-name-state"><span class="secno">9.2.4.8 </span><dfn>After attribute name state</dfn></h5>
2278:
2279: <p>Consume the <a href="#next-input-character">next input character</a>:</p>
2280:
2281: <dl class="switch"><dt>U+0009 CHARACTER TABULATION</dt>
2282: <dt>U+000A LINE FEED (LF)</dt>
2283: <dt>U+000C FORM FEED (FF)</dt>
2284: <!--<dt>U+000D CARRIAGE RETURN (CR)</dt>-->
2285: <dt>U+0020 SPACE</dt>
2286: <dd>Stay in the <a href="#after-attribute-name-state">after attribute name state</a>.</dd>
2287:
2288: <dt>U+002F SOLIDUS (/)</dt>
2289: <dd>Switch to the <a href="#self-closing-start-tag-state">self-closing start tag state</a>.</dd>
2290:
2291: <dt>U+003D EQUALS SIGN (=)</dt>
2292: <dd>Switch to the <a href="#before-attribute-value-state">before attribute value state</a>.</dd>
2293:
2294: <dt>U+003E GREATER-THAN SIGN (>)</dt>
2295: <dd>Emit the current tag token. Switch to the <a href="#data-state">data
2296: state</a>.</dd>
2297:
2298: <dt>U+0041 LATIN CAPITAL LETTER A through to U+005A LATIN CAPITAL LETTER Z</dt>
2299: <dd>Start a new attribute in the current tag token. Set that
2300: attribute's name to the lowercase version of the <a href="#current-input-character">current input character</a>
2301: (add 0x0020 to the character's code point), and its value to
2302: the empty string. Switch to the <a href="#attribute-name-state">attribute name
2303: state</a>.</dd>
2304:
2305: <dt>U+0022 QUOTATION MARK (")</dt>
2306: <dt>U+0027 APOSTROPHE (')</dt>
2307: <dt>U+003C LESS-THAN SIGN (<)</dt>
2308: <dd><a href="#parse-error">Parse error</a>. Treat it as per the "anything else"
2309: entry below.</dd>
2310:
2311: <dt>EOF</dt>
2312: <dd><a href="#parse-error">Parse error</a>. Reconsume the EOF character in the
2313: <a href="#data-state">data state</a>.</dd>
2314:
2315: <dt>Anything else</dt>
2316: <dd>Start a new attribute in the current tag token. Set that
2317: attribute's name to the <a href="#current-input-character">current input character</a>, and its value to
2318: the empty string. Switch to the <a href="#attribute-name-state">attribute name
2319: state</a>.</dd>
2320:
2321: </dl><h5 id="before-attribute-value-state"><span class="secno">9.2.4.9 </span><dfn>Before attribute value state</dfn></h5>
2322:
2323: <p>Consume the <a href="#next-input-character">next input character</a>:</p>
2324:
2325: <dl class="switch"><dt>U+0009 CHARACTER TABULATION</dt>
2326: <dt>U+000A LINE FEED (LF)</dt>
2327: <dt>U+000C FORM FEED (FF)</dt>
2328: <!--<dt>U+000D CARRIAGE RETURN (CR)</dt>-->
2329: <dt>U+0020 SPACE</dt>
2330: <dd>Stay in the <a href="#before-attribute-value-state">before attribute value state</a>.</dd>
2331:
2332: <dt>U+0022 QUOTATION MARK (")</dt>
2333: <dd>Switch to the <a href="#attribute-value-double-quoted-state">attribute value (double-quoted) state</a>.</dd>
2334:
2335: <dt>U+0026 AMPERSAND (&)</dt>
2336: <dd>Switch to the <a href="#attribute-value-unquoted-state">attribute value (unquoted) state</a>
2337: and reconsume this input character.</dd>
2338:
2339: <dt>U+0027 APOSTROPHE (')</dt>
2340: <dd>Switch to the <a href="#attribute-value-single-quoted-state">attribute value (single-quoted) state</a>.</dd>
2341:
2342: <dt>U+003E GREATER-THAN SIGN (>)</dt>
2343: <dd><a href="#parse-error">Parse error</a>. Emit the current tag token. Switch to
2344: the <a href="#data-state">data state</a>.</dd>
2345:
2346: <dt>U+003C LESS-THAN SIGN (<)</dt>
2347: <dt>U+003D EQUALS SIGN (=)</dt>
2348: <dd><a href="#parse-error">Parse error</a>. Treat it as per the "anything else"
2349: entry below.</dd>
2350:
2351: <dt>EOF</dt>
2352: <dd><a href="#parse-error">Parse error</a>. Reconsume the EOF character in the
2353: <a href="#data-state">data state</a>.</dd>
2354:
2355: <dt>Anything else</dt>
2356: <dd>Append the <a href="#current-input-character">current input character</a> to the current
2357: attribute's value. Switch to the <a href="#attribute-value-unquoted-state">attribute value (unquoted)
2358: state</a>.</dd>
2359:
2360: </dl><h5 id="attribute-value-double-quoted-state"><span class="secno">9.2.4.10 </span><dfn>Attribute value (double-quoted) state</dfn></h5>
2361:
2362: <p>Consume the <a href="#next-input-character">next input character</a>:</p>
2363:
2364: <dl class="switch"><dt>U+0022 QUOTATION MARK (")</dt>
2365: <dd>Switch to the <a href="#after-attribute-value-quoted-state">after attribute value (quoted)
2366: state</a>.</dd>
2367:
2368: <dt>U+0026 AMPERSAND (&)</dt>
2369: <dd>Switch to the <a href="#character-reference-in-attribute-value-state">character reference in attribute value
2370: state</a>, with the <a href="#additional-allowed-character">additional allowed character</a>
2371: being U+0022 QUOTATION MARK (").</dd>
2372:
2373: <dt>EOF</dt>
2374: <dd><a href="#parse-error">Parse error</a>. Reconsume the EOF character in the
2375: <a href="#data-state">data state</a>.</dd>
2376:
2377: <dt>Anything else</dt>
2378: <dd>Append the <a href="#current-input-character">current input character</a> to the current attribute's
2379: value. Stay in the <a href="#attribute-value-double-quoted-state">attribute value (double-quoted)
2380: state</a>.</dd>
2381:
2382: </dl><h5 id="attribute-value-single-quoted-state"><span class="secno">9.2.4.11 </span><dfn>Attribute value (single-quoted) state</dfn></h5>
2383:
2384: <p>Consume the <a href="#next-input-character">next input character</a>:</p>
2385:
2386: <dl class="switch"><dt>U+0027 APOSTROPHE (')</dt>
2387: <dd>Switch to the <a href="#after-attribute-value-quoted-state">after attribute value (quoted)
2388: state</a>.</dd>
2389:
2390: <dt>U+0026 AMPERSAND (&)</dt>
2391: <dd>Switch to the <a href="#character-reference-in-attribute-value-state">character reference in attribute value
2392: state</a>, with the <a href="#additional-allowed-character">additional allowed character</a>
2393: being U+0027 APOSTROPHE (').</dd>
2394:
2395: <dt>EOF</dt>
2396: <dd><a href="#parse-error">Parse error</a>. Reconsume the EOF character in the
2397: <a href="#data-state">data state</a>.</dd>
2398:
2399: <dt>Anything else</dt>
2400: <dd>Append the <a href="#current-input-character">current input character</a> to the current attribute's
2401: value. Stay in the <a href="#attribute-value-single-quoted-state">attribute value (single-quoted)
2402: state</a>.</dd>
2403:
2404: </dl><h5 id="attribute-value-unquoted-state"><span class="secno">9.2.4.12 </span><dfn>Attribute value (unquoted) state</dfn></h5>
2405:
2406: <p>Consume the <a href="#next-input-character">next input character</a>:</p>
2407:
2408: <dl class="switch"><dt>U+0009 CHARACTER TABULATION</dt>
2409: <dt>U+000A LINE FEED (LF)</dt>
2410: <dt>U+000C FORM FEED (FF)</dt>
2411: <!--<dt>U+000D CARRIAGE RETURN (CR)</dt>-->
2412: <dt>U+0020 SPACE</dt>
2413: <dd>Switch to the <a href="#before-attribute-name-state">before attribute name state</a>.</dd>
2414:
2415: <dt>U+0026 AMPERSAND (&)</dt>
2416: <dd>Switch to the <a href="#character-reference-in-attribute-value-state">character reference in attribute value
2417: state</a>, with no <a href="#additional-allowed-character">additional allowed
2418: character</a>.</dd>
2419:
2420: <dt>U+003E GREATER-THAN SIGN (>)</dt>
2421: <dd>Emit the current tag token. Switch to the <a href="#data-state">data
2422: state</a>.</dd>
2423:
2424: <dt>U+0022 QUOTATION MARK (")</dt>
2425: <dt>U+0027 APOSTROPHE (')</dt>
2426: <dt>U+003C LESS-THAN SIGN (<)</dt>
2427: <dt>U+003D EQUALS SIGN (=)</dt>
2428: <dd><a href="#parse-error">Parse error</a>. Treat it as per the "anything else"
2429: entry below.</dd>
2430:
2431: <dt>EOF</dt>
2432: <dd><a href="#parse-error">Parse error</a>. Reconsume the EOF character in the
2433: <a href="#data-state">data state</a>.</dd>
2434:
2435: <dt>Anything else</dt>
2436: <dd>Append the <a href="#current-input-character">current input character</a> to the current attribute's
2437: value. Stay in the <a href="#attribute-value-unquoted-state">attribute value (unquoted)
2438: state</a>.</dd>
2439:
2440: </dl><h5 id="character-reference-in-attribute-value-state"><span class="secno">9.2.4.13 </span><dfn>Character reference in attribute value state</dfn></h5>
2441:
2442: <p>Attempt to <a href="#consume-a-character-reference">consume a character reference</a>.</p>
2443:
2444: <p>If nothing is returned, append a U+0026 AMPERSAND character to
2445: the current attribute's value.</p>
2446:
2447: <p>Otherwise, append the returned character token to the current
2448: attribute's value.</p>
2449:
2450: <p>Finally, switch back to the attribute value state that you were
2451: in when were switched into this state.</p>
2452:
2453:
2454: <h5 id="after-attribute-value-quoted-state"><span class="secno">9.2.4.14 </span><dfn>After attribute value (quoted) state</dfn></h5>
2455:
2456: <p>Consume the <a href="#next-input-character">next input character</a>:</p>
2457:
2458: <dl class="switch"><dt>U+0009 CHARACTER TABULATION</dt>
2459: <dt>U+000A LINE FEED (LF)</dt>
2460: <dt>U+000C FORM FEED (FF)</dt>
2461: <!--<dt>U+000D CARRIAGE RETURN (CR)</dt>-->
2462: <dt>U+0020 SPACE</dt>
2463: <dd>Switch to the <a href="#before-attribute-name-state">before attribute name state</a>.</dd>
2464:
2465: <dt>U+002F SOLIDUS (/)</dt>
2466: <dd>Switch to the <a href="#self-closing-start-tag-state">self-closing start tag state</a>.</dd>
2467:
2468: <dt>U+003E GREATER-THAN SIGN (>)</dt>
2469: <dd>Emit the current tag token. Switch to the <a href="#data-state">data
2470: state</a>.</dd>
2471:
2472: <dt>EOF</dt>
2473: <dd><a href="#parse-error">Parse error</a>. Reconsume the EOF character in the
2474: <a href="#data-state">data state</a>.</dd>
2475:
2476: <dt>Anything else</dt>
2477: <dd><a href="#parse-error">Parse error</a>. Reconsume the character in
2478: the <a href="#before-attribute-name-state">before attribute name state</a>.</dd>
2479:
2480: </dl><h5 id="self-closing-start-tag-state"><span class="secno">9.2.4.15 </span><dfn>Self-closing start tag state</dfn></h5>
2481:
2482: <p>Consume the <a href="#next-input-character">next input character</a>:</p>
2483:
2484: <dl class="switch"><dt>U+003E GREATER-THAN SIGN (>)</dt>
2485: <dd>Set the <i>self-closing flag</i> of the current tag
2486: token. Emit the current tag token. Switch to the <a href="#data-state">data
2487: state</a>.</dd>
2488:
2489: <dt>EOF</dt>
2490: <dd><a href="#parse-error">Parse error</a>. Reconsume the EOF character in the
2491: <a href="#data-state">data state</a>.</dd>
2492:
2493: <dt>Anything else</dt>
2494: <dd><a href="#parse-error">Parse error</a>. Reconsume the character in
2495: the <a href="#before-attribute-name-state">before attribute name state</a>.</dd>
2496:
2497: </dl><h5 id="bogus-comment-state"><span class="secno">9.2.4.16 </span><dfn>Bogus comment state</dfn></h5>
2498:
2499: <p><i>(This can only happen if the <a href="#content-model-flag">content model
2500: flag</a> is set to the PCDATA state.)</i></p>
2501:
2502: <p>Consume every character up to and including the first U+003E
2503: GREATER-THAN SIGN character (>) or the end of the file (EOF),
2504: whichever comes first. Emit a comment token whose data is the
2505: concatenation of all the characters starting from and including
2506: the character that caused the state machine to switch into the
2507: bogus comment state, up to and including the character immediately
2508: before the last consumed character (i.e. up to the character just
2509: before the U+003E or EOF character). (If the comment was started
2510: by the end of the file (EOF), the token is empty.)</p>
2511:
2512: <p>Switch to the <a href="#data-state">data state</a>.</p>
2513:
2514: <p>If the end of the file was reached, reconsume the EOF
2515: character.</p>
2516:
2517:
2518: <h5 id="markup-declaration-open-state"><span class="secno">9.2.4.17 </span><dfn>Markup declaration open state</dfn></h5>
2519:
2520: <p><i>(This can only happen if the <a href="#content-model-flag">content model
2521: flag</a> is set to the PCDATA state.)</i></p>
2522:
2523: <p>If the next two characters are both U+002D HYPHEN-MINUS (-)
2524: characters, consume those two characters, create a comment token
2525: whose data is the empty string, and switch to the <a href="#comment-start-state">comment
2526: start state</a>.</p>
2527:
2528: <p>Otherwise, if the next seven characters are an <a href="infrastructure.html#ascii-case-insensitive">ASCII
2529: case-insensitive</a> match for the word "DOCTYPE", then consume
2530: those characters and switch to the <a href="#doctype-state">DOCTYPE state</a>.</p>
2531:
2532: <p>Otherwise, if the <a href="#insertion-mode">insertion mode</a> is "<a href="#parsing-main-inforeign" title="insertion mode: in foreign content">in foreign
2533: content</a>" and the <a href="#current-node">current node</a> is not an element
2534: in the <a href="#html-namespace-0">HTML namespace</a> and the next seven characters are
2535: an <span>ASCII case-sensitive</span> match for the string "[CDATA["
2536: (the five uppercase letters "CDATA" with a U+005B LEFT SQUARE
2537: BRACKET character before and after), then consume those characters
1.91 mike 2538: and switch to the <a href="#cdata-section-state">CDATA section state</a>.</p>
1.1 mike 2539:
2540: <p>Otherwise, this is a <a href="#parse-error">parse error</a>. Switch to the
2541: <a href="#bogus-comment-state">bogus comment state</a>. The next character that is
2542: consumed, if any, is the first character that will be in the
2543: comment.</p>
2544:
2545:
2546: <h5 id="comment-start-state"><span class="secno">9.2.4.18 </span><dfn>Comment start state</dfn></h5>
2547:
2548: <p>Consume the <a href="#next-input-character">next input character</a>:</p>
2549:
2550: <dl class="switch"><dt>U+002D HYPHEN-MINUS (-)</dt>
2551: <dd>Switch to the <a href="#comment-start-dash-state">comment start dash state</a>.</dd>
2552:
2553: <dt>U+003E GREATER-THAN SIGN (>)</dt>
2554: <dd><a href="#parse-error">Parse error</a>. Emit the comment token. Switch to the
2555: <a href="#data-state">data state</a>.</dd> <!-- see comment in comment end state
2556: -->
2557:
2558: <dt>EOF</dt>
2559: <dd><a href="#parse-error">Parse error</a>. Emit the comment token. Reconsume
2560: the EOF character in the <a href="#data-state">data state</a>.</dd>
2561:
2562: <dt>Anything else</dt>
2563: <dd>Append the input character to the comment token's
2564: data. Switch to the <a href="#comment-state">comment state</a>.</dd>
2565:
2566: </dl><h5 id="comment-start-dash-state"><span class="secno">9.2.4.19 </span><dfn>Comment start dash state</dfn></h5>
2567:
2568: <p>Consume the <a href="#next-input-character">next input character</a>:</p>
2569:
2570: <dl class="switch"><dt>U+002D HYPHEN-MINUS (-)</dt>
2571: <dd>Switch to the <a href="#comment-end-state">comment end state</a></dd>
2572:
2573: <dt>U+003E GREATER-THAN SIGN (>)</dt>
2574: <dd><a href="#parse-error">Parse error</a>. Emit the comment token. Switch to
2575: the <a href="#data-state">data state</a>.</dd>
2576:
2577: <dt>EOF</dt>
2578: <dd><a href="#parse-error">Parse error</a>. Emit the comment token. Reconsume the
2579: EOF character in the <a href="#data-state">data state</a>.</dd> <!-- see comment
2580: in comment end state -->
2581:
2582: <dt>Anything else</dt>
2583: <dd>Append a U+002D HYPHEN-MINUS (-) character and the input
2584: character to the comment token's data. Switch to the
2585: <a href="#comment-state">comment state</a>.</dd>
2586:
2587: </dl><h5 id="comment-state"><span class="secno">9.2.4.20 </span><dfn id="comment">Comment state</dfn></h5>
2588:
2589: <p>Consume the <a href="#next-input-character">next input character</a>:</p>
2590:
2591: <dl class="switch"><dt>U+002D HYPHEN-MINUS (-)</dt>
2592: <dd>Switch to the <a href="#comment-end-dash-state">comment end dash state</a></dd>
2593:
2594: <dt>EOF</dt>
2595: <dd><a href="#parse-error">Parse error</a>. Emit the comment token. Reconsume the
2596: EOF character in the <a href="#data-state">data state</a>.</dd> <!-- see comment
2597: in comment end state -->
2598:
2599: <dt>Anything else</dt>
2600: <dd>Append the input character to the comment token's data. Stay
2601: in the <a href="#comment-state">comment state</a>.</dd>
2602:
2603: </dl><h5 id="comment-end-dash-state"><span class="secno">9.2.4.21 </span><dfn>Comment end dash state</dfn></h5>
2604:
2605: <p>Consume the <a href="#next-input-character">next input character</a>:</p>
2606:
2607: <dl class="switch"><dt>U+002D HYPHEN-MINUS (-)</dt>
2608: <dd>Switch to the <a href="#comment-end-state">comment end state</a></dd>
2609:
2610: <dt>EOF</dt>
2611: <dd><a href="#parse-error">Parse error</a>. Emit the comment token. Reconsume the
2612: EOF character in the <a href="#data-state">data state</a>.</dd> <!-- see comment
2613: in comment end state -->
2614:
2615: <dt>Anything else</dt>
2616: <dd>Append a U+002D HYPHEN-MINUS (-) character and the input
2617: character to the comment token's data. Switch to the
2618: <a href="#comment-state">comment state</a>.</dd>
2619:
2620: </dl><h5 id="comment-end-state"><span class="secno">9.2.4.22 </span><dfn>Comment end state</dfn></h5>
2621:
2622: <p>Consume the <a href="#next-input-character">next input character</a>:</p>
2623:
2624: <dl class="switch"><dt>U+003E GREATER-THAN SIGN (>)</dt>
2625: <dd>Emit the comment token. Switch to the <a href="#data-state">data
2626: state</a>.</dd>
2627:
2628: <dt>U+002D HYPHEN-MINUS (-)</dt>
2629: <dd><a href="#parse-error">Parse error</a>. Append a U+002D HYPHEN-MINUS
2630: (-) character to the comment token's data. Stay in the
2631: <a href="#comment-end-state">comment end state</a>.</dd>
2632:
2633: <dt>U+0009 CHARACTER TABULATION</dt>
2634: <dt>U+000A LINE FEED (LF)</dt>
2635: <dt>U+000C FORM FEED (FF)</dt>
2636: <!--<dt>U+000D CARRIAGE RETURN (CR)</dt>-->
2637: <dt>U+0020 SPACE</dt>
2638: <dd><a href="#parse-error">Parse error</a>. Append two U+002D HYPHEN-MINUS (-)
2639: characters and the input character to the comment token's
2640: data. Switch to the <a href="#comment-end-space-state">comment end space state</a>.</dd>
2641:
2642: <dt>U+0021 EXCLAMATION MARK (!)</dt>
2643: <dd><a href="#parse-error">Parse error</a>. Switch to the <a href="#comment-end-bang-state">comment end bang
2644: state</a>.</dd>
2645:
2646: <dt>EOF</dt>
2647: <dd><a href="#parse-error">Parse error</a>. Emit the comment token. Reconsume
2648: the EOF character in the <a href="#data-state">data state</a>.</dd> <!-- For
2649: security reasons: otherwise, hostile user could put a <script> in
2650: a comment e.g. in a blog comment and then DOS the server so that
2651: the end tag isn't read, and then the commented <script> tag would
2652: be treated as live code -->
2653:
2654: <dt>Anything else</dt>
2655: <dd><a href="#parse-error">Parse error</a>. Append two U+002D HYPHEN-MINUS (-)
2656: characters and the input character to the comment token's
2657: data. Switch to the <a href="#comment-state">comment state</a>.</dd>
2658:
2659: </dl><h5 id="comment-end-bang-state"><span class="secno">9.2.4.23 </span><dfn>Comment end bang state</dfn></h5>
2660:
2661: <p>Consume the <a href="#next-input-character">next input character</a>:</p>
2662:
2663: <dl class="switch"><dt>U+003E GREATER-THAN SIGN (>)</dt>
2664: <dd>Emit the comment token. Switch to the <a href="#data-state">data
2665: state</a>.</dd>
2666:
2667: <dt>U+002D HYPHEN-MINUS (-)</dt>
2668: <dd>Append two U+002D HYPHEN-MINUS (-) characters and a U+0021
2669: EXCLAMATION MARK (!) character to the comment token's data. Switch
2670: to the <a href="#comment-end-dash-state">comment end dash state</a>.</dd>
2671:
2672: <dt>EOF</dt>
2673: <dd><a href="#parse-error">Parse error</a>. Emit the comment token. Reconsume
2674: the EOF character in the <a href="#data-state">data state</a>.</dd> <!-- see
2675: comment in comment end state -->
2676:
2677: <dt>Anything else</dt>
2678: <dd>Append two U+002D HYPHEN-MINUS (-) characters, a U+0021
2679: EXCLAMATION MARK (!) character, and the input character to the
2680: comment token's data. Switch to the <a href="#comment-state">comment
2681: state</a>.</dd>
2682:
2683: </dl><h5 id="comment-end-space-state"><span class="secno">9.2.4.24 </span><dfn>Comment end space state</dfn></h5>
2684:
2685: <p>Consume the <a href="#next-input-character">next input character</a>:</p>
2686:
2687: <dl class="switch"><dt>U+003E GREATER-THAN SIGN (>)</dt>
2688: <dd>Emit the comment token. Switch to the <a href="#data-state">data
2689: state</a>.</dd>
2690:
2691: <dt>U+002D HYPHEN-MINUS (-)</dt>
2692: <dd>Switch to the <a href="#comment-end-dash-state">comment end dash state</a>.</dd>
2693:
2694: <dt>U+0009 CHARACTER TABULATION</dt>
2695: <dt>U+000A LINE FEED (LF)</dt>
2696: <dt>U+000C FORM FEED (FF)</dt>
2697: <!--<dt>U+000D CARRIAGE RETURN (CR)</dt>-->
2698: <dt>U+0020 SPACE</dt>
2699: <dd>Append the input character to the comment token's data. Stay in
2700: the <a href="#comment-end-space-state">comment end space state</a>.</dd>
2701:
2702: <dt>EOF</dt>
2703: <dd><a href="#parse-error">Parse error</a>. Emit the comment token. Reconsume
2704: the EOF character in the <a href="#data-state">data state</a>.</dd> <!-- see
2705: comment in comment end state -->
2706:
2707: <dt>Anything else</dt>
2708: <dd>Append the input character to the comment token's data. Switch
2709: to the <a href="#comment-state">comment state</a>.</dd>
2710:
2711: </dl><h5 id="doctype-state"><span class="secno">9.2.4.25 </span><dfn>DOCTYPE state</dfn></h5>
2712:
2713: <p>Consume the <a href="#next-input-character">next input character</a>:</p>
2714:
2715: <dl class="switch"><dt>U+0009 CHARACTER TABULATION</dt>
2716: <dt>U+000A LINE FEED (LF)</dt>
2717: <dt>U+000C FORM FEED (FF)</dt>
2718: <!--<dt>U+000D CARRIAGE RETURN (CR)</dt>-->
2719: <dt>U+0020 SPACE</dt>
2720: <dd>Switch to the <a href="#before-doctype-name-state">before DOCTYPE name state</a>.</dd>
2721:
2722: <dt>EOF</dt>
2723: <dd><a href="#parse-error">Parse error</a>. Create a new DOCTYPE token. Set its
2724: <i>force-quirks flag</i> to <i>on</i>. Emit the token. Reconsume
2725: the EOF character in the <a href="#data-state">data state</a>.</dd>
2726:
2727: <dt>Anything else</dt>
2728: <dd><a href="#parse-error">Parse error</a>. Reconsume the current
2729: character in the <a href="#before-doctype-name-state">before DOCTYPE name state</a>.</dd>
2730:
2731: </dl><h5 id="before-doctype-name-state"><span class="secno">9.2.4.26 </span><dfn>Before DOCTYPE name state</dfn></h5>
2732:
2733: <p>Consume the <a href="#next-input-character">next input character</a>:</p>
2734:
2735: <dl class="switch"><dt>U+0009 CHARACTER TABULATION</dt>
2736: <dt>U+000A LINE FEED (LF)</dt>
2737: <dt>U+000C FORM FEED (FF)</dt>
2738: <!--<dt>U+000D CARRIAGE RETURN (CR)</dt>-->
2739: <dt>U+0020 SPACE</dt>
2740: <dd>Stay in the <a href="#before-doctype-name-state">before DOCTYPE name state</a>.</dd>
2741:
2742: <dt>U+003E GREATER-THAN SIGN (>)</dt>
2743: <dd><a href="#parse-error">Parse error</a>. Create a new DOCTYPE token. Set its
2744: <i>force-quirks flag</i> to <i>on</i>. Emit the token. Switch to
2745: the <a href="#data-state">data state</a>.</dd>
2746:
2747: <dt>U+0041 LATIN CAPITAL LETTER A through to U+005A LATIN CAPITAL LETTER Z</dt>
2748: <dd>Create a new DOCTYPE token. Set the token's name to the
2749: lowercase version of the input character (add 0x0020 to the
2750: character's code point). Switch to the <a href="#doctype-name-state">DOCTYPE name
2751: state</a>.</dd>
2752:
2753: <dt>EOF</dt>
2754: <dd><a href="#parse-error">Parse error</a>. Create a new DOCTYPE token. Set its
2755: <i>force-quirks flag</i> to <i>on</i>. Emit the token. Reconsume
2756: the EOF character in the <a href="#data-state">data state</a>.</dd>
2757:
2758: <dt>Anything else</dt>
2759: <dd>Create a new DOCTYPE token. Set the token's name to the
2760: <a href="#current-input-character">current input character</a>. Switch to the <a href="#doctype-name-state">DOCTYPE name
2761: state</a>.</dd>
2762:
2763: </dl><h5 id="doctype-name-state"><span class="secno">9.2.4.27 </span><dfn>DOCTYPE name state</dfn></h5>
2764:
2765: <p>Consume the <a href="#next-input-character">next input character</a>:</p>
2766:
2767: <dl class="switch"><dt>U+0009 CHARACTER TABULATION</dt>
2768: <dt>U+000A LINE FEED (LF)</dt>
2769: <dt>U+000C FORM FEED (FF)</dt>
2770: <!--<dt>U+000D CARRIAGE RETURN (CR)</dt>-->
2771: <dt>U+0020 SPACE</dt>
2772: <dd>Switch to the <a href="#after-doctype-name-state">after DOCTYPE name state</a>.</dd>
2773:
2774: <dt>U+003E GREATER-THAN SIGN (>)</dt>
2775: <dd>Emit the current DOCTYPE token. Switch to the <a href="#data-state">data
2776: state</a>.</dd>
2777:
2778: <dt>U+0041 LATIN CAPITAL LETTER A through to U+005A LATIN CAPITAL LETTER Z</dt>
2779: <dd>Append the lowercase version of the input character (add 0x0020
2780: to the character's code point) to the current DOCTYPE token's
2781: name. Stay in the <a href="#doctype-name-state">DOCTYPE name state</a>.</dd>
2782:
2783: <dt>EOF</dt>
2784: <dd><a href="#parse-error">Parse error</a>. Set the DOCTYPE token's
2785: <i>force-quirks flag</i> to <i>on</i>. Emit that DOCTYPE token.
2786: Reconsume the EOF character in the <a href="#data-state">data state</a>.</dd>
2787:
2788: <dt>Anything else</dt>
2789: <dd>Append the <a href="#current-input-character">current input character</a> to the current DOCTYPE
2790: token's name. Stay in the <a href="#doctype-name-state">DOCTYPE name state</a>.</dd>
2791:
2792: </dl><h5 id="after-doctype-name-state"><span class="secno">9.2.4.28 </span><dfn>After DOCTYPE name state</dfn></h5>
2793:
2794: <p>Consume the <a href="#next-input-character">next input character</a>:</p>
2795:
2796: <dl class="switch"><dt>U+0009 CHARACTER TABULATION</dt>
2797: <dt>U+000A LINE FEED (LF)</dt>
2798: <dt>U+000C FORM FEED (FF)</dt>
2799: <!--<dt>U+000D CARRIAGE RETURN (CR)</dt>-->
2800: <dt>U+0020 SPACE</dt>
2801: <dd>Stay in the <a href="#after-doctype-name-state">after DOCTYPE name state</a>.</dd>
2802:
2803: <dt>U+003E GREATER-THAN SIGN (>)</dt>
2804: <dd>Emit the current DOCTYPE token. Switch to the <a href="#data-state">data
2805: state</a>.</dd>
2806:
2807: <dt>EOF</dt>
2808: <dd><a href="#parse-error">Parse error</a>. Set the DOCTYPE token's
2809: <i>force-quirks flag</i> to <i>on</i>. Emit that DOCTYPE token.
2810: Reconsume the EOF character in the <a href="#data-state">data state</a>.</dd>
2811:
2812: <dt>Anything else</dt>
2813: <dd>
2814:
2815: <p>If the six characters starting from the <a href="#current-input-character">current input
2816: character</a> are an <a href="infrastructure.html#ascii-case-insensitive">ASCII case-insensitive</a> match
2817: for the word "PUBLIC", then consume those characters and switch to
2818: the <a href="#before-doctype-public-identifier-state">before DOCTYPE public identifier state</a>.</p>
2819:
2820: <p>Otherwise, if the six characters starting from the
2821: <a href="#current-input-character">current input character</a> are an <a href="infrastructure.html#ascii-case-insensitive">ASCII
2822: case-insensitive</a> match for the word "SYSTEM", then consume
2823: those characters and switch to the <a href="#before-doctype-system-identifier-state">before DOCTYPE system
2824: identifier state</a>.</p>
2825:
2826: <p>Otherwise, this is the <a href="#parse-error">parse error</a>. Set the
2827: DOCTYPE token's <i>force-quirks flag</i> to <i>on</i>. Switch to
2828: the <a href="#bogus-doctype-state">bogus DOCTYPE state</a>.</p>
2829:
2830: </dd>
2831:
2832: </dl><h5 id="before-doctype-public-identifier-state"><span class="secno">9.2.4.29 </span><dfn>Before DOCTYPE public identifier state</dfn></h5>
2833:
2834: <p>Consume the <a href="#next-input-character">next input character</a>:</p>
2835:
2836: <dl class="switch"><dt>U+0009 CHARACTER TABULATION</dt>
2837: <dt>U+000A LINE FEED (LF)</dt>
2838: <dt>U+000C FORM FEED (FF)</dt>
2839: <!--<dt>U+000D CARRIAGE RETURN (CR)</dt>-->
2840: <dt>U+0020 SPACE</dt>
2841: <dd>Stay in the <a href="#before-doctype-public-identifier-state">before DOCTYPE public identifier state</a>.</dd>
2842:
2843: <dt>U+0022 QUOTATION MARK (")</dt>
2844: <dd>Set the DOCTYPE token's public identifier to the empty string
2845: (not missing), then switch to the <a href="#doctype-public-identifier-double-quoted-state">DOCTYPE public identifier
2846: (double-quoted) state</a>.</dd>
2847:
2848: <dt>U+0027 APOSTROPHE (')</dt>
2849: <dd>Set the DOCTYPE token's public identifier to the empty string
2850: (not missing), then switch to the <a href="#doctype-public-identifier-single-quoted-state">DOCTYPE public identifier
2851: (single-quoted) state</a>.</dd>
2852:
2853: <dt>U+003E GREATER-THAN SIGN (>)</dt>
2854: <dd><a href="#parse-error">Parse error</a>. Set the DOCTYPE token's
2855: <i>force-quirks flag</i> to <i>on</i>. Emit that DOCTYPE
2856: token. Switch to the <a href="#data-state">data state</a>.</dd>
2857:
2858: <dt>EOF</dt>
2859: <dd><a href="#parse-error">Parse error</a>. Set the DOCTYPE token's
2860: <i>force-quirks flag</i> to <i>on</i>. Emit that DOCTYPE token.
2861: Reconsume the EOF character in the <a href="#data-state">data state</a>.</dd>
2862:
2863: <dt>Anything else</dt>
2864: <dd><a href="#parse-error">Parse error</a>. Set the DOCTYPE token's
2865: <i>force-quirks flag</i> to <i>on</i>. Switch to the <a href="#bogus-doctype-state">bogus
2866: DOCTYPE state</a>.</dd>
2867:
2868: </dl><h5 id="doctype-public-identifier-double-quoted-state"><span class="secno">9.2.4.30 </span><dfn>DOCTYPE public identifier (double-quoted) state</dfn></h5>
2869:
2870: <p>Consume the <a href="#next-input-character">next input character</a>:</p>
2871:
2872: <dl class="switch"><dt>U+0022 QUOTATION MARK (")</dt>
2873: <dd>Switch to the <a href="#after-doctype-public-identifier-state">after DOCTYPE public identifier state</a>.</dd>
2874:
2875: <dt>U+003E GREATER-THAN SIGN (>)</dt>
2876: <dd><a href="#parse-error">Parse error</a>. Set the DOCTYPE token's
2877: <i>force-quirks flag</i> to <i>on</i>. Emit that DOCTYPE
2878: token. Switch to the <a href="#data-state">data state</a>.</dd>
2879:
2880: <dt>EOF</dt>
2881: <dd><a href="#parse-error">Parse error</a>. Set the DOCTYPE token's
2882: <i>force-quirks flag</i> to <i>on</i>. Emit that DOCTYPE token.
2883: Reconsume the EOF character in the <a href="#data-state">data state</a>.</dd>
2884:
2885: <dt>Anything else</dt>
2886: <dd>Append the <a href="#current-input-character">current input character</a> to the current DOCTYPE
2887: token's public identifier. Stay in the <a href="#doctype-public-identifier-double-quoted-state">DOCTYPE public
2888: identifier (double-quoted) state</a>.</dd>
2889:
2890: </dl><h5 id="doctype-public-identifier-single-quoted-state"><span class="secno">9.2.4.31 </span><dfn>DOCTYPE public identifier (single-quoted) state</dfn></h5>
2891:
2892: <p>Consume the <a href="#next-input-character">next input character</a>:</p>
2893:
2894: <dl class="switch"><dt>U+0027 APOSTROPHE (')</dt>
2895: <dd>Switch to the <a href="#after-doctype-public-identifier-state">after DOCTYPE public identifier state</a>.</dd>
2896:
2897: <dt>U+003E GREATER-THAN SIGN (>)</dt>
2898: <dd><a href="#parse-error">Parse error</a>. Set the DOCTYPE token's
2899: <i>force-quirks flag</i> to <i>on</i>. Emit that DOCTYPE
2900: token. Switch to the <a href="#data-state">data state</a>.</dd>
2901:
2902: <dt>EOF</dt>
2903: <dd><a href="#parse-error">Parse error</a>. Set the DOCTYPE token's
2904: <i>force-quirks flag</i> to <i>on</i>. Emit that DOCTYPE token.
2905: Reconsume the EOF character in the <a href="#data-state">data state</a>.</dd>
2906:
2907: <dt>Anything else</dt>
2908: <dd>Append the <a href="#current-input-character">current input character</a> to the current DOCTYPE
2909: token's public identifier. Stay in the <a href="#doctype-public-identifier-single-quoted-state">DOCTYPE public
2910: identifier (single-quoted) state</a>.</dd>
2911:
2912: </dl><h5 id="after-doctype-public-identifier-state"><span class="secno">9.2.4.32 </span><dfn>After DOCTYPE public identifier state</dfn></h5>
2913:
2914: <p>Consume the <a href="#next-input-character">next input character</a>:</p>
2915:
2916: <dl class="switch"><dt>U+0009 CHARACTER TABULATION</dt>
2917: <dt>U+000A LINE FEED (LF)</dt>
2918: <dt>U+000C FORM FEED (FF)</dt>
2919: <!--<dt>U+000D CARRIAGE RETURN (CR)</dt>-->
2920: <dt>U+0020 SPACE</dt>
2921: <dd>Stay in the <a href="#after-doctype-public-identifier-state">after DOCTYPE public identifier state</a>.</dd>
2922:
2923: <dt>U+0022 QUOTATION MARK (")</dt>
2924: <dd>Set the DOCTYPE token's system identifier to the empty string
2925: (not missing), then switch to the <a href="#doctype-system-identifier-double-quoted-state">DOCTYPE system identifier
2926: (double-quoted) state</a>.</dd>
2927:
2928: <dt>U+0027 APOSTROPHE (')</dt>
2929: <dd>Set the DOCTYPE token's system identifier to the empty string
2930: (not missing), then switch to the <a href="#doctype-system-identifier-single-quoted-state">DOCTYPE system identifier
2931: (single-quoted) state</a>.</dd>
2932:
2933: <dt>U+003E GREATER-THAN SIGN (>)</dt>
2934: <dd>Emit the current DOCTYPE token. Switch to the <a href="#data-state">data
2935: state</a>.</dd>
2936:
2937: <dt>EOF</dt>
2938: <dd><a href="#parse-error">Parse error</a>. Set the DOCTYPE token's
2939: <i>force-quirks flag</i> to <i>on</i>. Emit that DOCTYPE token.
2940: Reconsume the EOF character in the <a href="#data-state">data state</a>.</dd>
2941:
2942: <dt>Anything else</dt>
2943: <dd><a href="#parse-error">Parse error</a>. Set the DOCTYPE token's
2944: <i>force-quirks flag</i> to <i>on</i>. Switch to the <a href="#bogus-doctype-state">bogus
2945: DOCTYPE state</a>.</dd>
2946:
2947: </dl><h5 id="before-doctype-system-identifier-state"><span class="secno">9.2.4.33 </span><dfn>Before DOCTYPE system identifier state</dfn></h5>
2948:
2949: <p>Consume the <a href="#next-input-character">next input character</a>:</p>
2950:
2951: <dl class="switch"><dt>U+0009 CHARACTER TABULATION</dt>
2952: <dt>U+000A LINE FEED (LF)</dt>
2953: <dt>U+000C FORM FEED (FF)</dt>
2954: <!--<dt>U+000D CARRIAGE RETURN (CR)</dt>-->
2955: <dt>U+0020 SPACE</dt>
2956: <dd>Stay in the <a href="#before-doctype-system-identifier-state">before DOCTYPE system identifier state</a>.</dd>
2957:
2958: <dt>U+0022 QUOTATION MARK (")</dt>
2959: <dd>Set the DOCTYPE token's system identifier to the empty string
2960: (not missing), then switch to the <a href="#doctype-system-identifier-double-quoted-state">DOCTYPE system identifier
2961: (double-quoted) state</a>.</dd>
2962:
2963: <dt>U+0027 APOSTROPHE (')</dt>
2964: <dd>Set the DOCTYPE token's system identifier to the empty string
2965: (not missing), then switch to the <a href="#doctype-system-identifier-single-quoted-state">DOCTYPE system identifier
2966: (single-quoted) state</a>.</dd>
2967:
2968: <dt>U+003E GREATER-THAN SIGN (>)</dt>
2969: <dd><a href="#parse-error">Parse error</a>. Set the DOCTYPE token's
2970: <i>force-quirks flag</i> to <i>on</i>. Emit that DOCTYPE
2971: token. Switch to the <a href="#data-state">data state</a>.</dd>
2972:
2973: <dt>EOF</dt>
2974: <dd><a href="#parse-error">Parse error</a>. Set the DOCTYPE token's
2975: <i>force-quirks flag</i> to <i>on</i>. Emit that DOCTYPE token.
2976: Reconsume the EOF character in the <a href="#data-state">data state</a>.</dd>
2977:
2978: <dt>Anything else</dt>
2979: <dd><a href="#parse-error">Parse error</a>. Set the DOCTYPE token's
2980: <i>force-quirks flag</i> to <i>on</i>. Switch to the <a href="#bogus-doctype-state">bogus
2981: DOCTYPE state</a>.</dd>
2982:
2983: </dl><h5 id="doctype-system-identifier-double-quoted-state"><span class="secno">9.2.4.34 </span><dfn>DOCTYPE system identifier (double-quoted) state</dfn></h5>
2984:
2985: <p>Consume the <a href="#next-input-character">next input character</a>:</p>
2986:
2987: <dl class="switch"><dt>U+0022 QUOTATION MARK (")</dt>
2988: <dd>Switch to the <a href="#after-doctype-system-identifier-state">after DOCTYPE system identifier state</a>.</dd>
2989:
2990: <dt>U+003E GREATER-THAN SIGN (>)</dt>
2991: <dd><a href="#parse-error">Parse error</a>. Set the DOCTYPE token's
2992: <i>force-quirks flag</i> to <i>on</i>. Emit that DOCTYPE
2993: token. Switch to the <a href="#data-state">data state</a>.</dd>
2994:
2995: <dt>EOF</dt>
2996: <dd><a href="#parse-error">Parse error</a>. Set the DOCTYPE token's
2997: <i>force-quirks flag</i> to <i>on</i>. Emit that DOCTYPE token.
2998: Reconsume the EOF character in the <a href="#data-state">data state</a>.</dd>
2999:
3000: <dt>Anything else</dt>
3001: <dd>Append the <a href="#current-input-character">current input character</a> to the current DOCTYPE
3002: token's system identifier. Stay in the <a href="#doctype-system-identifier-double-quoted-state">DOCTYPE system
3003: identifier (double-quoted) state</a>.</dd>
3004:
3005: </dl><h5 id="doctype-system-identifier-single-quoted-state"><span class="secno">9.2.4.35 </span><dfn>DOCTYPE system identifier (single-quoted) state</dfn></h5>
3006:
3007: <p>Consume the <a href="#next-input-character">next input character</a>:</p>
3008:
3009: <dl class="switch"><dt>U+0027 APOSTROPHE (')</dt>
3010: <dd>Switch to the <a href="#after-doctype-system-identifier-state">after DOCTYPE system identifier state</a>.</dd>
3011:
3012: <dt>U+003E GREATER-THAN SIGN (>)</dt>
3013: <dd><a href="#parse-error">Parse error</a>. Set the DOCTYPE token's
3014: <i>force-quirks flag</i> to <i>on</i>. Emit that DOCTYPE
3015: token. Switch to the <a href="#data-state">data state</a>.</dd>
3016:
3017: <dt>EOF</dt>
3018: <dd><a href="#parse-error">Parse error</a>. Set the DOCTYPE token's
3019: <i>force-quirks flag</i> to <i>on</i>. Emit that DOCTYPE token.
3020: Reconsume the EOF character in the <a href="#data-state">data state</a>.</dd>
3021:
3022: <dt>Anything else</dt>
3023: <dd>Append the <a href="#current-input-character">current input character</a> to the current DOCTYPE
3024: token's system identifier. Stay in the <a href="#doctype-system-identifier-single-quoted-state">DOCTYPE system
3025: identifier (single-quoted) state</a>.</dd>
3026:
3027: </dl><h5 id="after-doctype-system-identifier-state"><span class="secno">9.2.4.36 </span><dfn>After DOCTYPE system identifier state</dfn></h5>
3028:
3029: <p>Consume the <a href="#next-input-character">next input character</a>:</p>
3030:
3031: <dl class="switch"><dt>U+0009 CHARACTER TABULATION</dt>
3032: <dt>U+000A LINE FEED (LF)</dt>
3033: <dt>U+000C FORM FEED (FF)</dt>
3034: <!--<dt>U+000D CARRIAGE RETURN (CR)</dt>-->
3035: <dt>U+0020 SPACE</dt>
3036: <dd>Stay in the <a href="#after-doctype-system-identifier-state">after DOCTYPE system identifier state</a>.</dd>
3037:
3038: <dt>U+003E GREATER-THAN SIGN (>)</dt>
3039: <dd>Emit the current DOCTYPE token. Switch to the <a href="#data-state">data
3040: state</a>.</dd>
3041:
3042: <dt>EOF</dt>
3043: <dd><a href="#parse-error">Parse error</a>. Set the DOCTYPE token's
3044: <i>force-quirks flag</i> to <i>on</i>. Emit that DOCTYPE token.
3045: Reconsume the EOF character in the <a href="#data-state">data state</a>.</dd>
3046:
3047: <dt>Anything else</dt>
3048: <dd><a href="#parse-error">Parse error</a>. Switch to the <a href="#bogus-doctype-state">bogus DOCTYPE
3049: state</a>. (This does <em>not</em> set the DOCTYPE token's
3050: <i>force-quirks flag</i> to <i>on</i>.)</dd>
3051:
3052: </dl><h5 id="bogus-doctype-state"><span class="secno">9.2.4.37 </span><dfn>Bogus DOCTYPE state</dfn></h5>
3053:
3054: <p>Consume the <a href="#next-input-character">next input character</a>:</p>
3055:
3056: <dl class="switch"><dt>U+003E GREATER-THAN SIGN (>)</dt>
3057: <dd>Emit the DOCTYPE token. Switch to the <a href="#data-state">data
3058: state</a>.</dd>
3059:
3060: <dt>EOF</dt>
3061: <dd>Emit the DOCTYPE token. Reconsume the EOF character in the
3062: <a href="#data-state">data state</a>.</dd>
3063:
3064: <dt>Anything else</dt>
3065: <dd>Stay in the <a href="#bogus-doctype-state">bogus DOCTYPE state</a>.</dd>
3066:
3067: </dl><h5 id="cdata-section-state"><span class="secno">9.2.4.38 </span><dfn>CDATA section state</dfn></h5>
3068:
3069: <p><i>(This can only happen if the <a href="#content-model-flag">content model
1.91 mike 3070: flag</a> is set to the PCDATA state.)</i></p>
1.1 mike 3071:
3072: <p>Consume every character up to the next occurrence of the three
3073: character sequence U+005D RIGHT SQUARE BRACKET U+005D RIGHT SQUARE
3074: BRACKET U+003E GREATER-THAN SIGN (<code title="">]]></code>), or the
3075: end of the file (EOF), whichever comes first. Emit a series of
3076: character tokens consisting of all the characters consumed except
3077: the matching three character sequence at the end (if one was found
3078: before the end of the file).</p>
3079:
3080: <p>Switch to the <a href="#data-state">data state</a>.</p>
3081:
3082: <p>If the end of the file was reached, reconsume the EOF
3083: character.</p>
3084:
3085:
3086:
3087: <h5 id="tokenizing-character-references"><span class="secno">9.2.4.39 </span>Tokenizing character references</h5>
3088:
3089: <p>This section defines how to <dfn id="consume-a-character-reference">consume a character
3090: reference</dfn>. This definition is used when parsing character
3091: references <a href="#character-reference-data-state" title="character reference data state">in
3092: text</a> and <a href="#character-reference-in-attribute-value-state" title="character reference in attribute value
3093: state">in attributes</a>.</p>
3094:
3095: <p>The behavior depends on the identity of the next character (the
3096: one immediately after the U+0026 AMPERSAND character):</p>
3097:
3098: <dl class="switch"><dt>U+0009 CHARACTER TABULATION</dt>
3099: <dt>U+000A LINE FEED (LF)</dt>
3100: <dt>U+000C FORM FEED (FF)</dt>
3101: <!--<dt>U+000D CARRIAGE RETURN (CR)</dt>-->
3102: <dt>U+0020 SPACE</dt>
3103: <dt>U+003C LESS-THAN SIGN</dt>
3104: <dt>U+0026 AMPERSAND</dt>
3105: <dt>EOF</dt>
3106: <dt>The <dfn id="additional-allowed-character">additional allowed character</dfn>, if there is one</dt>
3107:
3108: <dd>Not a character reference. No characters are consumed, and
3109: nothing is returned. (This is not an error, either.)</dd>
3110:
3111:
3112: <dt>U+0023 NUMBER SIGN (#)</dt>
3113:
3114: <dd>
3115:
3116: <p>Consume the U+0023 NUMBER SIGN.</p>
3117:
3118: <p>The behavior further depends on the character after the U+0023
3119: NUMBER SIGN:</p>
3120:
3121: <dl class="switch"><dt>U+0078 LATIN SMALL LETTER X</dt>
3122: <dt>U+0058 LATIN CAPITAL LETTER X</dt>
3123:
3124: <dd>
3125:
3126: <p>Consume the X.</p>
3127:
3128: <p>Follow the steps below, but using the range of characters
3129: U+0030 DIGIT ZERO through to U+0039 DIGIT NINE, U+0061 LATIN
3130: SMALL LETTER A through to U+0066 LATIN SMALL LETTER F, and
3131: U+0041 LATIN CAPITAL LETTER A, through to U+0046 LATIN CAPITAL
3132: LETTER F (in other words, 0-9, A-F, a-f).</p>
3133:
3134: <p>When it comes to interpreting the number, interpret it as a
3135: hexadecimal number.</p>
3136:
3137: </dd>
3138:
3139:
3140: <dt>Anything else</dt>
3141:
3142: <dd>
3143:
3144: <p>Follow the steps below, but using the range of characters
3145: U+0030 DIGIT ZERO through to U+0039 DIGIT NINE (i.e. just
3146: 0-9).</p>
3147:
3148: <p>When it comes to interpreting the number, interpret it as a
3149: decimal number.</p>
3150:
3151: </dd>
3152:
3153: </dl><p>Consume as many characters as match the range of characters
3154: given above.</p>
3155:
3156: <p>If no characters match the range, then don't consume any
3157: characters (and unconsume the U+0023 NUMBER SIGN character and, if
3158: appropriate, the X character). This is a <a href="#parse-error">parse
3159: error</a>; nothing is returned.</p>
3160:
3161: <p>Otherwise, if the next character is a U+003B SEMICOLON, consume
3162: that too. If it isn't, there is a <a href="#parse-error">parse
3163: error</a>.</p>
3164:
3165: <p>If one or more characters match the range, then take them all
3166: and interpret the string of characters as a number (either
3167: hexadecimal or decimal as appropriate).</p>
3168:
3169: <p>If that number is one of the numbers in the first column of the
3170: following table, then this is a <a href="#parse-error">parse error</a>. Find the
3171: row with that number in the first column, and return a character
3172: token for the Unicode character given in the second column of that
3173: row.</p>
3174:
3175: <table><thead><tr><th>Number </th><th colspan="2">Unicode character
3176: </th></tr></thead><tbody><tr><td>0x00 </td><td>U+FFFD </td><td>REPLACEMENT CHARACTER
3177: </td></tr><tr><td>0x0D </td><td>U+000A </td><td>LINE FEED (LF)
3178: </td></tr><tr><td>0x80 </td><td>U+20AC </td><td>EURO SIGN ('€')
3179: </td></tr><tr><td>0x81 </td><td>U+0081 </td><td><control>
3180: </td></tr><tr><td>0x82 </td><td>U+201A </td><td>SINGLE LOW-9 QUOTATION MARK ('‚')
3181: </td></tr><tr><td>0x83 </td><td>U+0192 </td><td>LATIN SMALL LETTER F WITH HOOK ('ƒ')
3182: </td></tr><tr><td>0x84 </td><td>U+201E </td><td>DOUBLE LOW-9 QUOTATION MARK ('„')
3183: </td></tr><tr><td>0x85 </td><td>U+2026 </td><td>HORIZONTAL ELLIPSIS ('…')
3184: </td></tr><tr><td>0x86 </td><td>U+2020 </td><td>DAGGER ('†')
3185: </td></tr><tr><td>0x87 </td><td>U+2021 </td><td>DOUBLE DAGGER ('‡')
3186: </td></tr><tr><td>0x88 </td><td>U+02C6 </td><td>MODIFIER LETTER CIRCUMFLEX ACCENT ('ˆ')
3187: </td></tr><tr><td>0x89 </td><td>U+2030 </td><td>PER MILLE SIGN ('‰')
3188: </td></tr><tr><td>0x8A </td><td>U+0160 </td><td>LATIN CAPITAL LETTER S WITH CARON ('Š')
3189: </td></tr><tr><td>0x8B </td><td>U+2039 </td><td>SINGLE LEFT-POINTING ANGLE QUOTATION MARK ('‹')
3190: </td></tr><tr><td>0x8C </td><td>U+0152 </td><td>LATIN CAPITAL LIGATURE OE ('Œ')
3191: </td></tr><tr><td>0x8D </td><td>U+008D </td><td><control>
3192: </td></tr><tr><td>0x8E </td><td>U+017D </td><td>LATIN CAPITAL LETTER Z WITH CARON ('Ž')
3193: </td></tr><tr><td>0x8F </td><td>U+008F </td><td><control>
3194: </td></tr><tr><td>0x90 </td><td>U+0090 </td><td><control>
3195: </td></tr><tr><td>0x91 </td><td>U+2018 </td><td>LEFT SINGLE QUOTATION MARK ('‘')
3196: </td></tr><tr><td>0x92 </td><td>U+2019 </td><td>RIGHT SINGLE QUOTATION MARK ('’')
3197: </td></tr><tr><td>0x93 </td><td>U+201C </td><td>LEFT DOUBLE QUOTATION MARK ('“')
3198: </td></tr><tr><td>0x94 </td><td>U+201D </td><td>RIGHT DOUBLE QUOTATION MARK ('”')
3199: </td></tr><tr><td>0x95 </td><td>U+2022 </td><td>BULLET ('•')
3200: </td></tr><tr><td>0x96 </td><td>U+2013 </td><td>EN DASH ('–')
3201: </td></tr><tr><td>0x97 </td><td>U+2014 </td><td>EM DASH ('—')
3202: </td></tr><tr><td>0x98 </td><td>U+02DC </td><td>SMALL TILDE ('˜')
3203: </td></tr><tr><td>0x99 </td><td>U+2122 </td><td>TRADE MARK SIGN ('™')
3204: </td></tr><tr><td>0x9A </td><td>U+0161 </td><td>LATIN SMALL LETTER S WITH CARON ('š')
3205: </td></tr><tr><td>0x9B </td><td>U+203A </td><td>SINGLE RIGHT-POINTING ANGLE QUOTATION MARK ('›')
3206: </td></tr><tr><td>0x9C </td><td>U+0153 </td><td>LATIN SMALL LIGATURE OE ('œ')
3207: </td></tr><tr><td>0x9D </td><td>U+009D </td><td><control>
3208: </td></tr><tr><td>0x9E </td><td>U+017E </td><td>LATIN SMALL LETTER Z WITH CARON ('ž')
3209: </td></tr><tr><td>0x9F </td><td>U+0178 </td><td>LATIN CAPITAL LETTER Y WITH DIAERESIS ('Ÿ')
1.18 mike 3210: </td></tr></tbody></table><p>Otherwise, if the number is greater than 0x10FFFF, then this is
3211: a <a href="#parse-error">parse error</a>. Return a U+FFFD REPLACEMENT
3212: CHARACTER.</p>
3213:
3214: <p>Otherwise, return a character token for the Unicode character
1.1 mike 3215: whose code point is that number.
3216:
3217: <!-- this is the same as the equivalent list in the input stream
3218: section -->
3219: If the number is in the range 0x0001 to 0x0008, <!-- HT, LF
3220: allowed --> <!-- U+000B is in the next list --> <!-- FF, CR
3221: allowed --> 0x000E to 0x001F, <!-- ASCII allowed --> 0x007F <!--to
3222: 0x0084, (0x0085 NEL not allowed), 0x0086--> to 0x009F, 0xD800 to
3223: 0xDFFF<!-- surrogates not allowed -->, 0xFDD0 to 0xFDEF, or is one
3224: of 0x000B, 0xFFFE, 0xFFFF, 0x1FFFE, 0x1FFFF, 0x2FFFE, 0x2FFFF,
3225: 0x3FFFE, 0x3FFFF, 0x4FFFE, 0x4FFFF, 0x5FFFE, 0x5FFFF, 0x6FFFE,
3226: 0x6FFFF, 0x7FFFE, 0x7FFFF, 0x8FFFE, 0x8FFFF, 0x9FFFE, 0x9FFFF,
3227: 0xAFFFE, 0xAFFFF, 0xBFFFE, 0xBFFFF, 0xCFFFE, 0xCFFFF, 0xDFFFE,
3228: 0xDFFFF, 0xEFFFE, 0xEFFFF, 0xFFFFE, 0xFFFFF, 0x10FFFE, or
1.18 mike 3229: 0x10FFFF, then this is a <a href="#parse-error">parse error</a>.</p>
1.1 mike 3230:
3231: </dd>
3232:
3233:
3234: <dt>Anything else</dt>
3235:
3236: <dd>
3237:
3238: <p>Consume the maximum number of characters possible, with the
3239: consumed characters matching one of the identifiers in the first
3240: column of the <a href="named-character-references.html#named-character-references">named character references</a> table (in a
3241: <a href="infrastructure.html#case-sensitive">case-sensitive</a> manner).</p>
3242:
3243: <p>If no match can be made, then this is a <a href="#parse-error">parse
3244: error</a>. No characters are consumed, and nothing is
3245: returned.</p>
3246:
3247: <p>If the last character matched is not a U+003B SEMICOLON (<code title="">;</code>), there is a <a href="#parse-error">parse error</a>.</p>
3248:
3249: <p>If the character reference is being consumed <a href="#character-reference-in-attribute-value-state" title="character reference in attribute value state">as part of an
3250: attribute</a>, and the last character matched is not a U+003B
3251: SEMICOLON (<code title="">;</code>), and the next character is in
3252: the range U+0030 DIGIT ZERO to U+0039 DIGIT NINE, U+0041 LATIN
3253: CAPITAL LETTER A to U+005A LATIN CAPITAL LETTER Z, or U+0061 LATIN
3254: SMALL LETTER A to U+007A LATIN SMALL LETTER Z, then, for
3255: historical reasons, all the characters that were matched after the
3256: U+0026 AMPERSAND (&) must be unconsumed, and nothing is
3257: returned.</p>
3258:
3259: <p>Otherwise, return a character token for the character
3260: corresponding to the character reference name (as given by the
3261: second column of the <a href="named-character-references.html#named-character-references">named character references</a>
3262: table).</p>
3263:
3264: <div class="example">
3265:
3266: <p>If the markup contains <code title="">I'm &notit; I tell
3267: you</code>, the character reference is parsed as "not", as in,
3268: <code title="">I'm ¬it; I tell you</code>. But if the markup
3269: was <code title="">I'm &notin; I tell you</code>, the
3270: character reference would be parsed as "notin;", resulting in
3271: <code title="">I'm ∉ I tell you</code>.</p>
3272:
3273: </div>
3274:
3275: </dd>
3276:
3277: </dl><h4 id="tree-construction"><span class="secno">9.2.5 </span><dfn>Tree construction</dfn></h4>
3278:
3279: <p>The input to the tree construction stage is a sequence of tokens
3280: from the <a href="#tokenization">tokenization</a> stage. The tree construction
3281: stage is associated with a DOM <code>Document</code> object when a
3282: parser is created. The "output" of this stage consists of
3283: dynamically modifying or extending that document's DOM tree.</p>
3284:
3285: <p>This specification does not define when an interactive user agent
3286: has to render the <code>Document</code> so that it is available to
3287: the user, or when it has to begin accepting user input.</p>
3288:
3289: <p>As each token is emitted from the tokenizer, the user agent must
3290: process the token according to the rules given in the section
3291: corresponding to the current <a href="#insertion-mode">insertion mode</a>.</p>
3292:
3293: <p>When the steps below require the UA to <dfn id="insert-a-character">insert a
3294: character</dfn> into a node, if that node has a child immediately
3295: before where the character is to be inserted, and that child is a
3296: <code>Text</code> node, and that <code>Text</code> node was the last
3297: node that the parser inserted into the document, then the character
3298: must be appended to that <code>Text</code> node; otherwise, a new
3299: <code>Text</code> node whose data is just that character must be
3300: inserted in the appropriate place.</p>
3301:
3302: <div class="example">
3303:
3304: <p>Here are some sample inputs to the parser and the corresponding
3305: number of text nodes that they result in, assuming a user agent
3306: that executes scripts.</p>
3307:
3308: <table><thead><tr><th>Input </th><th>Number of text nodes
3309: </th></tr></thead><tbody><tr><td><pre>A<script>
3310: var script = document.getElementsByTagName('script')[0];
3311: document.body.removeChild(script);
3312: </script>B</pre>
3313: </td><td>Two adjacent text nodes in the document, containing "A" and "B".
3314: </td></tr><tr><td><pre>A<script>
3315: var text = document.createTextNode('B');
3316: document.body.appendChild(text);
3317: </script>C</pre>
3318: </td><td>Four text nodes; "A" before the script, the script's contents, "B" after the script, and then, immediately after that, "C".
3319: </td></tr><tr><td><pre>A<script>
3320: var text = document.getElementsByTagName('script')[0].firstChild;
3321: text.data = 'B';
3322: document.body.appendChild(text);
3323: </script>B</pre>
3324: </td><td>Two adjacent text nodes in the document, containing "A" and "BB".
3325: </td></tr><tr><td><pre>A<table>B<tr>C</tr>C</table></pre>
3326: </td><td>Three adjacent text nodes before the table, containing "A", "B", and "CC" respectively. (This is caused by <a href="#foster-parent" title="foster parent">foster parenting</a>.)
3327: </td></tr><tr><td><pre>A<table><tr> B</tr> B</table></pre>
3328: </td><td>Two adjacent text nodes before the table, containing "A" and " B B" (space-B-space-B) respectively. (This is caused by <a href="#foster-parent" title="foster parent">foster parenting</a>.)
3329: </td></tr><tr><td><pre>A<table><tr> B</tr> </em>C</table></pre>
3330: </td><td>Three adjacent text nodes before the table, containing "A", " B" (space-B), and "C" respectively, and one text node inside the table (as a child of a <code><a href="tabular-data.html#the-tbody-element">tbody</a></code>) with a single space character. (Space characters separated from non-space characters by non-character tokens are not affected by <a href="#foster-parent" title="foster parent">foster parenting</a>, even if those other tokens then get ignored.)
3331: </td></tr></tbody></table></div>
3332:
3333: <p id="mutation-during-parsing">DOM mutation events must not fire
3334: for changes caused by the UA parsing the document. (Conceptually,
3335: the parser is not mutating the DOM, it is constructing it.) This
1.159 mike 3336: includes the parsing of any content inserted using <code title="dom-document-write"><a href="embedded-content-0.html#dom-document-write">document.write()</a></code> and <code title="dom-document-writeln"><a href="embedded-content-0.html#dom-document-writeln">document.writeln()</a></code> calls. <a href="references.html#refsDOMEVENTS">[DOMEVENTS]</a></p>
1.1 mike 3337:
3338: <p class="note">Not all of the tag names mentioned below are
3339: conformant tag names in this specification; many are included to
3340: handle legacy content. They still form part of the algorithm that
3341: implementations are required to implement to claim conformance.</p>
3342:
3343: <p class="note">The algorithm described below places no limit on the
3344: depth of the DOM tree generated, or on the length of tag names,
3345: attribute names, attribute values, text nodes, etc. While
3346: implementors are encouraged to avoid arbitrary limits, it is
3347: recognized that <a href="infrastructure.html#hardwareLimitations">practical
3348: concerns</a> will likely force user agents to impose nesting
3349: depths.</p>
3350:
3351:
3352: <h5 id="creating-and-inserting-elements"><span class="secno">9.2.5.1 </span>Creating and inserting elements</h5>
3353:
3354: <p>When the steps below require the UA to <dfn id="create-an-element-for-the-token" title="create an
3355: element for the token">create an element for a token</dfn> in a
3356: particular namespace, the UA must create a node implementing the
3357: interface appropriate for the element type corresponding to the tag
3358: name of the token in the given namespace (as given in the
3359: specification that defines that element, e.g. for an <code><a href="text-level-semantics.html#the-a-element">a</a></code>
3360: element in the <a href="#html-namespace-0">HTML namespace</a>, this specification
3361: defines it to be the <code><a href="text-level-semantics.html#htmlanchorelement">HTMLAnchorElement</a></code> interface), with
3362: the tag name being the name of that element, with the node being in
3363: the given namespace, and with the attributes on the node being those
3364: given in the given token.</p>
3365:
3366: <p>The interface appropriate for an element in the <a href="#html-namespace-0">HTML
3367: namespace</a> that is not defined in this specification is
1.44 mike 3368: <code><a href="dom.html#htmlunknownelement">HTMLUnknownElement</a></code>. Element in other namespaces whose
1.1 mike 3369: interface is not defined by that namespace's specification must use
3370: the interface <code>Element</code>.</p>
3371:
3372: <p>When a <a href="forms.html#category-reset" title="category-reset">resettable</a> element is
3373: created in this manner, its <a href="forms.html#concept-form-reset-control" title="concept-form-reset-control">reset algorithm</a> must be
3374: invoked once the attributes are set. (This initializes the element's
3375: <a href="forms.html#concept-fe-value" title="concept-fe-value">value</a> and <a href="forms.html#concept-fe-checked" title="concept-fe-checked">checkedness</a> based on the element's
3376: attributes.)</p>
3377:
3378: <hr><p>When the steps below require the UA to <dfn id="insert-an-html-element">insert an HTML
3379: element</dfn> for a token, the UA must first <a href="#create-an-element-for-the-token">create an element
3380: for the token</a> in the <a href="#html-namespace-0">HTML namespace</a>, and then
3381: append this node to the <a href="#current-node">current node</a>, and push it onto
3382: the <a href="#stack-of-open-elements">stack of open elements</a> so that it is the new
3383: <a href="#current-node">current node</a>.</p>
3384:
3385: <p>The steps below may also require that the UA insert an HTML
3386: element in a particular place, in which case the UA must follow the
3387: same steps except that it must insert or append the new node in the
3388: location specified instead of appending it to the <a href="#current-node">current
3389: node</a>. (This happens in particular during the parsing of
3390: tables with invalid content.)</p>
3391:
3392: <p>If an element created by the <a href="#insert-an-html-element">insert an HTML element</a>
3393: algorithm is a <a href="forms.html#form-associated-element">form-associated element</a>, and the
3394: <a href="#form-element-pointer"><code title="">form</code> element pointer</a> is not null,
3395: and the newly created element doesn't have a <code title="attr-fae-form"><a href="forms.html#attr-fae-form">form</a></code> attribute, the user agent must
3396: <a href="forms.html#concept-form-association" title="concept-form-association">associate</a> the newly
3397: created element with the <code><a href="forms.html#the-form-element">form</a></code> element pointed to by the
3398: <a href="#form-element-pointer"><code title="">form</code> element pointer</a> before
3399: inserting it wherever it is to be inserted.</p>
3400:
3401: <hr><p>When the steps below require the UA to <dfn id="insert-a-foreign-element">insert a foreign
3402: element</dfn> for a token, the UA must first <a href="#create-an-element-for-the-token">create an element
3403: for the token</a> in the given namespace, and then append this
3404: node to the <a href="#current-node">current node</a>, and push it onto the
3405: <a href="#stack-of-open-elements">stack of open elements</a> so that it is the new
3406: <a href="#current-node">current node</a>. If the newly created element has an <code title="">xmlns</code> attribute in the <a href="#xmlns-namespace">XMLNS namespace</a>
3407: whose value is not exactly the same as the element's namespace, that
3408: is a <a href="#parse-error">parse error</a>. Similarly, if the newly created
3409: element has an <code title="">xmlns:xlink</code> attribute in the
3410: <a href="#xmlns-namespace">XMLNS namespace</a> whose value is not the <a href="#xlink-namespace">XLink
3411: Namespace</a>, that is a <a href="#parse-error">parse error</a>.</p>
3412:
3413: <p>When the steps below require the user agent to <dfn id="adjust-mathml-attributes">adjust MathML
3414: attributes</dfn> for a token, then, if the token has an attribute
3415: named <code title="">definitionurl</code>, change its name to <code title="">definitionURL</code> (note the case difference).</p>
3416:
3417: <p>When the steps below require the user agent to <dfn id="adjust-svg-attributes">adjust SVG
3418: attributes</dfn> for a token, then, for each attribute on the token
3419: whose attribute name is one of the ones in the first column of the
3420: following table, change the attribute's name to the name given in
3421: the corresponding cell in the second column. (This fixes the case of
3422: SVG attributes that are not all lowercase.)</p>
3423:
3424: <table><thead><tr><th> Attribute name on token </th><th> Attribute name on element
3425: </th></tr></thead><tbody><tr><td> <code title="">attributename</code> </td><td> <code title="">attributeName</code>
3426: </td></tr><tr><td> <code title="">attributetype</code> </td><td> <code title="">attributeType</code>
3427: </td></tr><tr><td> <code title="">basefrequency</code> </td><td> <code title="">baseFrequency</code>
3428: </td></tr><tr><td> <code title="">baseprofile</code> </td><td> <code title="">baseProfile</code>
3429: </td></tr><tr><td> <code title="">calcmode</code> </td><td> <code title="">calcMode</code>
3430: </td></tr><tr><td> <code title="">clippathunits</code> </td><td> <code title="">clipPathUnits</code>
3431: </td></tr><tr><td> <code title="">contentscripttype</code> </td><td> <code title="">contentScriptType</code>
3432: </td></tr><tr><td> <code title="">contentstyletype</code> </td><td> <code title="">contentStyleType</code>
3433: </td></tr><tr><td> <code title="">diffuseconstant</code> </td><td> <code title="">diffuseConstant</code>
3434: </td></tr><tr><td> <code title="">edgemode</code> </td><td> <code title="">edgeMode</code>
3435: </td></tr><tr><td> <code title="">externalresourcesrequired</code> </td><td> <code title="">externalResourcesRequired</code>
3436: </td></tr><tr><td> <code title="">filterres</code> </td><td> <code title="">filterRes</code>
3437: </td></tr><tr><td> <code title="">filterunits</code> </td><td> <code title="">filterUnits</code>
3438: </td></tr><tr><td> <code title="">glyphref</code> </td><td> <code title="">glyphRef</code>
3439: </td></tr><tr><td> <code title="">gradienttransform</code> </td><td> <code title="">gradientTransform</code>
3440: </td></tr><tr><td> <code title="">gradientunits</code> </td><td> <code title="">gradientUnits</code>
3441: </td></tr><tr><td> <code title="">kernelmatrix</code> </td><td> <code title="">kernelMatrix</code>
3442: </td></tr><tr><td> <code title="">kernelunitlength</code> </td><td> <code title="">kernelUnitLength</code>
3443: </td></tr><tr><td> <code title="">keypoints</code> </td><td> <code title="">keyPoints</code>
3444: </td></tr><tr><td> <code title="">keysplines</code> </td><td> <code title="">keySplines</code>
3445: </td></tr><tr><td> <code title="">keytimes</code> </td><td> <code title="">keyTimes</code>
3446: </td></tr><tr><td> <code title="">lengthadjust</code> </td><td> <code title="">lengthAdjust</code>
3447: </td></tr><tr><td> <code title="">limitingconeangle</code> </td><td> <code title="">limitingConeAngle</code>
3448: </td></tr><tr><td> <code title="">markerheight</code> </td><td> <code title="">markerHeight</code>
3449: </td></tr><tr><td> <code title="">markerunits</code> </td><td> <code title="">markerUnits</code>
3450: </td></tr><tr><td> <code title="">markerwidth</code> </td><td> <code title="">markerWidth</code>
3451: </td></tr><tr><td> <code title="">maskcontentunits</code> </td><td> <code title="">maskContentUnits</code>
3452: </td></tr><tr><td> <code title="">maskunits</code> </td><td> <code title="">maskUnits</code>
3453: </td></tr><tr><td> <code title="">numoctaves</code> </td><td> <code title="">numOctaves</code>
3454: </td></tr><tr><td> <code title="">pathlength</code> </td><td> <code title="">pathLength</code>
3455: </td></tr><tr><td> <code title="">patterncontentunits</code> </td><td> <code title="">patternContentUnits</code>
3456: </td></tr><tr><td> <code title="">patterntransform</code> </td><td> <code title="">patternTransform</code>
3457: </td></tr><tr><td> <code title="">patternunits</code> </td><td> <code title="">patternUnits</code>
3458: </td></tr><tr><td> <code title="">pointsatx</code> </td><td> <code title="">pointsAtX</code>
3459: </td></tr><tr><td> <code title="">pointsaty</code> </td><td> <code title="">pointsAtY</code>
3460: </td></tr><tr><td> <code title="">pointsatz</code> </td><td> <code title="">pointsAtZ</code>
3461: </td></tr><tr><td> <code title="">preservealpha</code> </td><td> <code title="">preserveAlpha</code>
3462: </td></tr><tr><td> <code title="">preserveaspectratio</code> </td><td> <code title="">preserveAspectRatio</code>
3463: </td></tr><tr><td> <code title="">primitiveunits</code> </td><td> <code title="">primitiveUnits</code>
3464: </td></tr><tr><td> <code title="">refx</code> </td><td> <code title="">refX</code>
3465: </td></tr><tr><td> <code title="">refy</code> </td><td> <code title="">refY</code>
3466: </td></tr><tr><td> <code title="">repeatcount</code> </td><td> <code title="">repeatCount</code>
3467: </td></tr><tr><td> <code title="">repeatdur</code> </td><td> <code title="">repeatDur</code>
3468: </td></tr><tr><td> <code title="">requiredextensions</code> </td><td> <code title="">requiredExtensions</code>
3469: </td></tr><tr><td> <code title="">requiredfeatures</code> </td><td> <code title="">requiredFeatures</code>
3470: </td></tr><tr><td> <code title="">specularconstant</code> </td><td> <code title="">specularConstant</code>
3471: </td></tr><tr><td> <code title="">specularexponent</code> </td><td> <code title="">specularExponent</code>
3472: </td></tr><tr><td> <code title="">spreadmethod</code> </td><td> <code title="">spreadMethod</code>
3473: </td></tr><tr><td> <code title="">startoffset</code> </td><td> <code title="">startOffset</code>
3474: </td></tr><tr><td> <code title="">stddeviation</code> </td><td> <code title="">stdDeviation</code>
3475: </td></tr><tr><td> <code title="">stitchtiles</code> </td><td> <code title="">stitchTiles</code>
3476: </td></tr><tr><td> <code title="">surfacescale</code> </td><td> <code title="">surfaceScale</code>
3477: </td></tr><tr><td> <code title="">systemlanguage</code> </td><td> <code title="">systemLanguage</code>
3478: </td></tr><tr><td> <code title="">tablevalues</code> </td><td> <code title="">tableValues</code>
3479: </td></tr><tr><td> <code title="">targetx</code> </td><td> <code title="">targetX</code>
3480: </td></tr><tr><td> <code title="">targety</code> </td><td> <code title="">targetY</code>
3481: </td></tr><tr><td> <code title="">textlength</code> </td><td> <code title="">textLength</code>
3482: </td></tr><tr><td> <code title="">viewbox</code> </td><td> <code title="">viewBox</code>
3483: </td></tr><tr><td> <code title="">viewtarget</code> </td><td> <code title="">viewTarget</code>
3484: </td></tr><tr><td> <code title="">xchannelselector</code> </td><td> <code title="">xChannelSelector</code>
3485: </td></tr><tr><td> <code title="">ychannelselector</code> </td><td> <code title="">yChannelSelector</code>
3486: </td></tr><tr><td> <code title="">zoomandpan</code> </td><td> <code title="">zoomAndPan</code>
3487: </td></tr></tbody></table><p>When the steps below require the user agent to <dfn id="adjust-foreign-attributes">adjust
3488: foreign attributes</dfn> for a token, then, if any of the attributes
3489: on the token match the strings given in the first column of the
3490: following table, let the attribute be a namespaced attribute, with
3491: the prefix being the string given in the corresponding cell in the
3492: second column, the local name being the string given in the
3493: corresponding cell in the third column, and the namespace being the
3494: namespace given in the corresponding cell in the fourth
3495: column. (This fixes the use of namespaced attributes, in particular
3496: <a href="dom.html#attr-xml-lang" title="attr-xml-lang"><code title="">lang</code> attributes in
3497: the <span>XML namespace</span></a>.)</p>
3498:
3499: <table><thead><tr><th> Attribute name </th><th> Prefix </th><th> Local name </th><th> Namespace
3500: </th></tr></thead><tbody><tr><td> <code title="">xlink:actuate</code> </td><td> <code title="">xlink</code> </td><td> <code title="">actuate</code> </td><td> <a href="#xlink-namespace">XLink namespace</a>
3501: </td></tr><tr><td> <code title="">xlink:arcrole</code> </td><td> <code title="">xlink</code> </td><td> <code title="">arcrole</code> </td><td> <a href="#xlink-namespace">XLink namespace</a>
3502: </td></tr><tr><td> <code title="">xlink:href</code> </td><td> <code title="">xlink</code> </td><td> <code title="">href</code> </td><td> <a href="#xlink-namespace">XLink namespace</a>
3503: </td></tr><tr><td> <code title="">xlink:role</code> </td><td> <code title="">xlink</code> </td><td> <code title="">role</code> </td><td> <a href="#xlink-namespace">XLink namespace</a>
3504: </td></tr><tr><td> <code title="">xlink:show</code> </td><td> <code title="">xlink</code> </td><td> <code title="">show</code> </td><td> <a href="#xlink-namespace">XLink namespace</a>
3505: </td></tr><tr><td> <code title="">xlink:title</code> </td><td> <code title="">xlink</code> </td><td> <code title="">title</code> </td><td> <a href="#xlink-namespace">XLink namespace</a>
3506: </td></tr><tr><td> <code title="">xlink:type</code> </td><td> <code title="">xlink</code> </td><td> <code title="">type</code> </td><td> <a href="#xlink-namespace">XLink namespace</a>
3507: </td></tr><tr><td> <code title="">xml:base</code> </td><td> <code title="">xml</code> </td><td> <code title="">base</code> </td><td> <a href="#xml-namespace">XML namespace</a> <!-- attr-xml-base -->
3508: </td></tr><tr><td> <code title="">xml:lang</code> </td><td> <code title="">xml</code> </td><td> <code title="">lang</code> </td><td> <a href="#xml-namespace">XML namespace</a>
3509: </td></tr><tr><td> <code title="">xml:space</code> </td><td> <code title="">xml</code> </td><td> <code title="">space</code> </td><td> <a href="#xml-namespace">XML namespace</a>
3510: </td></tr><tr><td> <code title="">xmlns</code> </td><td> (none) </td><td> <code title="">xmlns</code> </td><td> <a href="#xmlns-namespace">XMLNS namespace</a>
3511: </td></tr><tr><td> <code title="">xmlns:xlink</code> </td><td> <code title="">xmlns</code> </td><td> <code title="">xlink</code> </td><td> <a href="#xmlns-namespace">XMLNS namespace</a>
1.91 mike 3512: </td></tr></tbody></table><hr><p>The <dfn id="generic-raw-text-element-parsing-algorithm">generic raw text element parsing algorithm</dfn> and the
1.1 mike 3513: <dfn id="generic-rcdata-element-parsing-algorithm">generic RCDATA element parsing algorithm</dfn> consist of the
3514: following steps. These algorithms are always invoked in response to
3515: a start tag token.</p>
3516:
3517: <ol><li><p><a href="#insert-an-html-element">Insert an HTML element</a> for the token.</p></li>
3518:
1.91 mike 3519: <li><p>If the algorithm that was invoked is the <a href="#generic-raw-text-element-parsing-algorithm">generic raw
3520: text element parsing algorithm</a>, switch the tokenizer's
3521: <a href="#content-model-flag">content model flag</a> to the RAWTEXT state; otherwise the
1.1 mike 3522: algorithm invoked was the <a href="#generic-rcdata-element-parsing-algorithm">generic RCDATA element parsing
3523: algorithm</a>, switch the tokenizer's <a href="#content-model-flag">content model
3524: flag</a> to the RCDATA state.</p></li>
3525:
3526: <li><p>Let the <a href="#original-insertion-mode">original insertion mode</a> be the current
3527: <a href="#insertion-mode">insertion mode</a>.</p>
3528:
1.91 mike 3529: </li><li><p>Then, switch the <a href="#insertion-mode">insertion mode</a> to "<a href="#parsing-main-incdata" title="insertion mode: in RAWTEXT/RCDATA">in
3530: RAWTEXT/RCDATA</a>".</p></li>
1.1 mike 3531:
3532: </ol><h5 id="closing-elements-that-have-implied-end-tags"><span class="secno">9.2.5.2 </span>Closing elements that have implied end tags</h5>
3533:
3534: <p>When the steps below require the UA to <dfn id="generate-implied-end-tags">generate implied end
3535: tags</dfn>, then, while the <a href="#current-node">current node</a> is a
3536: <code><a href="semantics.html#the-dd-element">dd</a></code> element, a <code><a href="semantics.html#the-dt-element">dt</a></code> element, an
3537: <code><a href="semantics.html#the-li-element">li</a></code> element, an <code><a href="forms.html#the-option-element">option</a></code> element, an
3538: <code><a href="forms.html#the-optgroup-element">optgroup</a></code> element, a <code><a href="semantics.html#the-p-element">p</a></code> element, an
3539: <code><a href="text-level-semantics.html#the-rp-element">rp</a></code> element, or an <code><a href="text-level-semantics.html#the-rt-element">rt</a></code> element, the UA must
3540: pop the <a href="#current-node">current node</a> off the <a href="#stack-of-open-elements">stack of open
3541: elements</a>.</p>
3542:
3543: <p>If a step requires the UA to generate implied end tags but lists
3544: an element to exclude from the process, then the UA must perform the
3545: above steps as if that element was not in the above list.</p>
3546:
3547:
3548: <h5 id="foster-parenting"><span class="secno">9.2.5.3 </span>Foster parenting</h5>
3549:
3550: <p>Foster parenting happens when content is misnested in tables.</p>
3551:
3552: <p>When a node <var title="">node</var> is to be <dfn id="foster-parent" title="foster
3553: parent">foster parented</dfn>, the node <var title="">node</var>
3554: must be inserted into the <i><a href="#foster-parent-element">foster parent element</a></i>.</p>
3555:
3556: <p>The <dfn id="foster-parent-element">foster parent element</dfn> is the parent element of the
3557: last <code><a href="tabular-data.html#the-table-element">table</a></code> element in the <a href="#stack-of-open-elements">stack of open
3558: elements</a>, if there is a <code><a href="tabular-data.html#the-table-element">table</a></code> element and it has
3559: such a parent element. If there is no <code><a href="tabular-data.html#the-table-element">table</a></code> element in
3560: the <a href="#stack-of-open-elements">stack of open elements</a> (<a href="#fragment-case">fragment
3561: case</a>), then the <i><a href="#foster-parent-element">foster parent element</a></i> is the first
3562: element in the <a href="#stack-of-open-elements">stack of open elements</a> (the
1.159 mike 3563: <code><a href="semantics.html#the-html-element-0">html</a></code> element). Otherwise, if there is a
1.1 mike 3564: <code><a href="tabular-data.html#the-table-element">table</a></code> element in the <a href="#stack-of-open-elements">stack of open
3565: elements</a>, but the last <code><a href="tabular-data.html#the-table-element">table</a></code> element in the
3566: <a href="#stack-of-open-elements">stack of open elements</a> has no parent, or its parent
3567: node is not an element, then the <i><a href="#foster-parent-element">foster parent element</a></i> is
3568: the element before the last <code><a href="tabular-data.html#the-table-element">table</a></code> element in the
3569: <a href="#stack-of-open-elements">stack of open elements</a>.</p>
3570:
3571: <p>If the <i><a href="#foster-parent-element">foster parent element</a></i> is the parent element of the
3572: last <code><a href="tabular-data.html#the-table-element">table</a></code> element in the <a href="#stack-of-open-elements">stack of open
3573: elements</a>, then <var title="">node</var> must be inserted
3574: immediately <em>before</em> the last <code><a href="tabular-data.html#the-table-element">table</a></code> element in
3575: the <a href="#stack-of-open-elements">stack of open elements</a> in the <i><a href="#foster-parent-element">foster parent
3576: element</a></i>; otherwise, <var title="">node</var> must be
3577: <em>appended</em> to the <i><a href="#foster-parent-element">foster parent element</a></i>.</p>
3578:
3579:
3580:
3581: <h5 id="the-initial-insertion-mode"><span class="secno">9.2.5.4 </span>The "<dfn title="insertion mode: initial">initial</dfn>" insertion mode</h5>
3582:
3583: <p>When the <a href="#insertion-mode">insertion mode</a> is "<a href="#the-initial-insertion-mode" title="insertion
3584: mode: initial">initial</a>", tokens must be handled as follows:</p>
3585:
3586: <dl class="switch"><dt>A character token that is one of U+0009 CHARACTER
3587: TABULATION, U+000A LINE FEED (LF), U+000C FORM FEED (FF),
3588: <!--U+000D CARRIAGE RETURN (CR),--> or U+0020 SPACE</dt>
3589: <dd>
3590: <p>Ignore the token.</p>
3591: </dd>
3592:
3593: <dt>A comment token</dt>
3594: <dd>
3595: <p>Append a <code>Comment</code> node to the <code>Document</code>
3596: object with the <code title="">data</code> attribute set to the
3597: data given in the comment token.</p>
3598: </dd>
3599:
3600: <dt>A DOCTYPE token</dt>
3601: <dd>
3602:
3603: <p>If the DOCTYPE token's name is not a
3604: <a href="infrastructure.html#case-sensitive">case-sensitive</a> match for the string "<code title="">html</code>", or the token's public identifier is not
3605: missing, or the token's system identifier is neither missing nor a
3606: <a href="infrastructure.html#case-sensitive">case-sensitive</a> match for the string
3607: "<code>about:legacy-compat</code>", and none of the sets of
3608: conditions in the following list are matched, then there is a
3609: <a href="#parse-error">parse error</a>. If one of the sets of conditions in the
3610: following list is matched, then there is an <dfn id="obsolete-permitted-doctype">obsolete
3611: permitted DOCTYPE</dfn>.</p>
3612:
3613: <ul><!-- only things that trigger no-quirks mode and were valid in
3614: some other spec are allowed in this list --><li>The DOCTYPE token's name is an <a href="infrastructure.html#ascii-case-insensitive">ASCII
1.98 mike 3615: case-insensitive</a> match for the string "<code title="">html</code>", the token's public identifier is the
1.1 mike 3616: <a href="infrastructure.html#case-sensitive">case-sensitive</a> string "<code title="">-//W3C//DTD HTML 4.0//EN</code>", and the
3617: token's system identifier is either missing or the
3618: <a href="infrastructure.html#case-sensitive">case-sensitive</a> string "<code title="">http://www.w3.org/TR/REC-html40/strict.dtd</code>".</li>
3619:
3620: <li>The DOCTYPE token's name is an <a href="infrastructure.html#ascii-case-insensitive">ASCII
1.98 mike 3621: case-insensitive</a> match for the string "<code title="">html</code>", the token's public identifier is the
1.1 mike 3622: <a href="infrastructure.html#case-sensitive">case-sensitive</a> string "<code title="">-//W3C//DTD HTML 4.01//EN</code>", and the
3623: token's system identifier is either missing or the
3624: <a href="infrastructure.html#case-sensitive">case-sensitive</a> string "<code title="">http://www.w3.org/TR/html4/strict.dtd</code>".</li>
3625:
3626: <li>The DOCTYPE token's name is an <a href="infrastructure.html#ascii-case-insensitive">ASCII
1.98 mike 3627: case-insensitive</a> match for the string "<code title="">html</code>", the token's public identifier is the
1.1 mike 3628: <a href="infrastructure.html#case-sensitive">case-sensitive</a> string "<code title="">-//W3C//DTD XHTML 1.0 Strict//EN</code>",
3629: and the token's system identifier is the
3630: <a href="infrastructure.html#case-sensitive">case-sensitive</a> string "<code title="">http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd</code>".</li>
3631:
3632: <li>The DOCTYPE token's name is an <a href="infrastructure.html#ascii-case-insensitive">ASCII
1.98 mike 3633: case-insensitive</a> match for the string "<code title="">html</code>", the token's public identifier is the
1.1 mike 3634: <a href="infrastructure.html#case-sensitive">case-sensitive</a> string "<code title="">-//W3C//DTD XHTML 1.1//EN</code>", and the
3635: token's system identifier is the <a href="infrastructure.html#case-sensitive">case-sensitive</a>
3636: string "<code title="">http://www.w3.org/TR/xhtml11/DTD/xhtml11.dtd</code>".</li>
3637:
3638: </ul><p>Conformance checkers may, based on the values (including
3639: presence or lack thereof) of the DOCTYPE token's name, public
3640: identifier, or system identifier, switch to a conformance checking
3641: mode for another language (e.g. based on the DOCTYPE token a
3642: conformance checker could recognize that the document is an
1.183 mike 3643: HTML4-era document, and defer to an HTML4 conformance
1.1 mike 3644: checker.)</p>
3645:
3646: <p>Append a <code>DocumentType</code> node to the
3647: <code>Document</code> node, with the <code title="">name</code>
3648: attribute set to the name given in the DOCTYPE token, or the empty
3649: string if the name was missing; the <code title="">publicId</code>
3650: attribute set to the public identifier given in the DOCTYPE token,
3651: or the empty string if the public identifier was missing; the
3652: <code title="">systemId</code> attribute set to the system
3653: identifier given in the DOCTYPE token, or the empty string if the
3654: system identifier was missing; and the other attributes specific
3655: to <code>DocumentType</code> objects set to null and empty lists
3656: as appropriate. Associate the <code>DocumentType</code> node with
3657: the <code>Document</code> object so that it is returned as the
3658: value of the <code title="">doctype</code> attribute of the
3659: <code>Document</code> object.</p>
3660:
3661: <p id="quirks-mode-doctypes">Then, if the DOCTYPE token matches
3662: one of the conditions in the following list, then set the
3663: <code>Document</code> to <a href="dom.html#quirks-mode">quirks mode</a>:</p>
3664:
3665: <ul class="brief"><li> The <i>force-quirks flag</i> is set to <i>on</i>. </li>
3666: <li> The name is set to anything other than "<code title="">HTML</code>". </li>
3667: <li> The public identifier starts with: "<code title="">+//Silmaril//dtd html Pro v0r11 19970101//<!--EN--></code>" </li>
3668: <li> The public identifier starts with: "<code title="">-//AdvaSoft Ltd//DTD HTML 3.0 asWedit + extensions//<!--EN--></code>" </li>
3669: <li> The public identifier starts with: "<code title="">-//AS//DTD HTML 3.0 asWedit + extensions//<!--EN--></code>" </li>
3670: <li> The public identifier starts with: "<code title="">-//IETF//DTD HTML 2.0 Level 1//<!--EN--></code>" </li>
3671: <li> The public identifier starts with: "<code title="">-//IETF//DTD HTML 2.0 Level 2//<!--EN--></code>" </li>
3672: <li> The public identifier starts with: "<code title="">-//IETF//DTD HTML 2.0 Strict Level 1//<!--EN--></code>" </li>
3673: <li> The public identifier starts with: "<code title="">-//IETF//DTD HTML 2.0 Strict Level 2//<!--EN--></code>" </li>
3674: <li> The public identifier starts with: "<code title="">-//IETF//DTD HTML 2.0 Strict//<!--EN--></code>" </li>
3675: <li> The public identifier starts with: "<code title="">-//IETF//DTD HTML 2.0//<!--EN--></code>" </li>
3676: <li> The public identifier starts with: "<code title="">-//IETF//DTD HTML 2.1E//<!--EN--></code>" </li>
3677: <li> The public identifier starts with: "<code title="">-//IETF//DTD HTML 3.0//<!--EN--></code>" </li>
3678: <!--<li> The public identifier is set to: "<code title="">-//IETF//DTD HTML 3.0//EN//</code>" </li>-->
3679: <li> The public identifier starts with: "<code title="">-//IETF//DTD HTML 3.2 Final//<!--EN--></code>" </li>
3680: <li> The public identifier starts with: "<code title="">-//IETF//DTD HTML 3.2//<!--EN--></code>" </li>
3681: <li> The public identifier starts with: "<code title="">-//IETF//DTD HTML 3//<!--EN--></code>" </li>
3682: <li> The public identifier starts with: "<code title="">-//IETF//DTD HTML Level 0//<!--EN--></code>" </li>
3683: <!--<li> The public identifier is set to: "<code title="">-//IETF//DTD HTML Level 0//EN//2.0</code>" </li>-->
3684: <li> The public identifier starts with: "<code title="">-//IETF//DTD HTML Level 1//<!--EN--></code>" </li>
3685: <!--<li> The public identifier is set to: "<code title="">-//IETF//DTD HTML Level 1//EN//2.0</code>" </li>-->
3686: <li> The public identifier starts with: "<code title="">-//IETF//DTD HTML Level 2//<!--EN--></code>" </li>
3687: <!--<li> The public identifier is set to: "<code title="">-//IETF//DTD HTML Level 2//EN//2.0</code>" </li>-->
3688: <li> The public identifier starts with: "<code title="">-//IETF//DTD HTML Level 3//<!--EN--></code>" </li>
3689: <!--<li> The public identifier is set to: "<code title="">-//IETF//DTD HTML Level 3//EN//3.0</code>" </li>-->
3690: <li> The public identifier starts with: "<code title="">-//IETF//DTD HTML Strict Level 0//<!--EN--></code>" </li>
3691: <!--<li> The public identifier is set to: "<code title="">-//IETF//DTD HTML Strict Level 0//EN//2.0</code>" </li>-->
3692: <li> The public identifier starts with: "<code title="">-//IETF//DTD HTML Strict Level 1//<!--EN--></code>" </li>
3693: <!--<li> The public identifier is set to: "<code title="">-//IETF//DTD HTML Strict Level 1//EN//2.0</code>" </li>-->
3694: <li> The public identifier starts with: "<code title="">-//IETF//DTD HTML Strict Level 2//<!--EN--></code>" </li>
3695: <!--<li> The public identifier is set to: "<code title="">-//IETF//DTD HTML Strict Level 2//EN//2.0</code>" </li>-->
3696: <li> The public identifier starts with: "<code title="">-//IETF//DTD HTML Strict Level 3//<!--EN--></code>" </li>
3697: <!--<li> The public identifier is set to: "<code title="">-//IETF//DTD HTML Strict Level 3//EN//3.0</code>" </li>-->
3698: <li> The public identifier starts with: "<code title="">-//IETF//DTD HTML Strict//<!--EN--></code>" </li>
3699: <!--<li> The public identifier is set to: "<code title="">-//IETF//DTD HTML Strict//EN//2.0</code>" </li>-->
3700: <!--<li> The public identifier is set to: "<code title="">-//IETF//DTD HTML Strict//EN//3.0</code>" </li>-->
3701: <li> The public identifier starts with: "<code title="">-//IETF//DTD HTML//<!--EN--></code>" </li>
3702: <!--<li> The public identifier is set to: "<code title="">-//IETF//DTD HTML//EN//2.0</code>" </li>-->
3703: <!--<li> The public identifier is set to: "<code title="">-//IETF//DTD HTML//EN//3.0</code>" </li>-->
3704: <li> The public identifier starts with: "<code title="">-//Metrius//DTD Metrius Presentational//<!--EN--></code>" </li>
3705: <li> The public identifier starts with: "<code title="">-//Microsoft//DTD Internet Explorer 2.0 HTML Strict//<!--EN--></code>" </li>
3706: <li> The public identifier starts with: "<code title="">-//Microsoft//DTD Internet Explorer 2.0 HTML//<!--EN--></code>" </li>
3707: <li> The public identifier starts with: "<code title="">-//Microsoft//DTD Internet Explorer 2.0 Tables//<!--EN--></code>" </li>
3708: <li> The public identifier starts with: "<code title="">-//Microsoft//DTD Internet Explorer 3.0 HTML Strict//<!--EN--></code>" </li>
3709: <li> The public identifier starts with: "<code title="">-//Microsoft//DTD Internet Explorer 3.0 HTML//<!--EN--></code>" </li>
3710: <li> The public identifier starts with: "<code title="">-//Microsoft//DTD Internet Explorer 3.0 Tables//<!--EN--></code>" </li>
3711: <li> The public identifier starts with: "<code title="">-//Netscape Comm. Corp.//DTD HTML//<!--EN--></code>" </li>
3712: <li> The public identifier starts with: "<code title="">-//Netscape Comm. Corp.//DTD Strict HTML//<!--EN--></code>" </li>
3713: <li> The public identifier starts with: "<code title="">-//O'Reilly and Associates//DTD HTML 2.0//<!--EN--></code>" </li>
3714: <li> The public identifier starts with: "<code title="">-//O'Reilly and Associates//DTD HTML Extended 1.0//<!--EN--></code>" </li>
3715: <li> The public identifier starts with: "<code title="">-//O'Reilly and Associates//DTD HTML Extended Relaxed 1.0//<!--EN--></code>" </li>
3716: <li> The public identifier starts with: "<code title="">-//SoftQuad Software//DTD HoTMetaL PRO 6.0::19990601::extensions to HTML 4.0//<!--EN--></code>" </li>
3717: <li> The public identifier starts with: "<code title="">-//SoftQuad//DTD HoTMetaL PRO 4.0::19971010::extensions to HTML 4.0//<!--EN--></code>" </li>
3718: <li> The public identifier starts with: "<code title="">-//Spyglass//DTD HTML 2.0 Extended//<!--EN--></code>" </li>
3719: <li> The public identifier starts with: "<code title="">-//SQ//DTD HTML 2.0 HoTMetaL + extensions//<!--EN--></code>" </li>
3720: <li> The public identifier starts with: "<code title="">-//Sun Microsystems Corp.//DTD HotJava HTML//<!--EN--></code>" </li>
3721: <li> The public identifier starts with: "<code title="">-//Sun Microsystems Corp.//DTD HotJava Strict HTML//<!--EN--></code>" </li>
3722: <li> The public identifier starts with: "<code title="">-//W3C//DTD HTML 3 1995-03-24//<!--EN--></code>" </li>
3723: <li> The public identifier starts with: "<code title="">-//W3C//DTD HTML 3.2 Draft//<!--EN--></code>" </li>
3724: <li> The public identifier starts with: "<code title="">-//W3C//DTD HTML 3.2 Final//<!--EN--></code>" </li>
3725: <li> The public identifier starts with: "<code title="">-//W3C//DTD HTML 3.2//<!--EN--></code>" </li>
3726: <li> The public identifier starts with: "<code title="">-//W3C//DTD HTML 3.2S Draft//<!--EN--></code>" </li>
3727: <li> The public identifier starts with: "<code title="">-//W3C//DTD HTML 4.0 Frameset//<!--EN--></code>" </li>
3728: <li> The public identifier starts with: "<code title="">-//W3C//DTD HTML 4.0 Transitional//<!--EN--></code>" </li>
3729: <li> The public identifier starts with: "<code title="">-//W3C//DTD HTML Experimental 19960712//<!--EN--></code>" </li>
3730: <li> The public identifier starts with: "<code title="">-//W3C//DTD HTML Experimental 970421//<!--EN--></code>" </li>
3731: <li> The public identifier starts with: "<code title="">-//W3C//DTD W3 HTML//<!--EN--></code>" </li>
3732: <li> The public identifier starts with: "<code title="">-//W3O//DTD W3 HTML 3.0//<!--EN--></code>" </li>
3733: <!--<li> The public identifier is set to: "<code title="">-//W3O//DTD W3 HTML 3.0//EN//</code>" </li>-->
3734: <li> The public identifier is set to: "<code title="">-//W3O//DTD W3 HTML Strict 3.0//EN//</code>" </li>
3735: <li> The public identifier starts with: "<code title="">-//WebTechs//DTD Mozilla HTML 2.0//<!--EN--></code>" </li>
3736: <li> The public identifier starts with: "<code title="">-//WebTechs//DTD Mozilla HTML//<!--EN--></code>" </li>
3737: <li> The public identifier is set to: "<code title="">-/W3C/DTD HTML 4.0 Transitional/EN</code>" </li>
3738: <li> The public identifier is set to: "<code title="">HTML</code>" </li>
3739: <li> The system identifier is set to: "<code title="">http://www.ibm.com/data/dtd/v11/ibmxhtml1-transitional.dtd</code>" </li>
3740: <li> The system identifier is missing and the public identifier starts with: "<code title="">-//W3C//DTD HTML 4.01 Frameset//<!--EN--></code>" </li>
3741: <li> The system identifier is missing and the public identifier starts with: "<code title="">-//W3C//DTD HTML 4.01 Transitional//<!--EN--></code>" </li>
3742: </ul><p>Otherwise, if the DOCTYPE token matches one of the conditions
3743: in the following list, then set the <code>Document</code> to
3744: <a href="dom.html#limited-quirks-mode">limited quirks mode</a>:</p>
3745:
3746: <ul class="brief"><li> The public identifier starts with: "<code title="">-//W3C//DTD XHTML 1.0 Frameset//<!--EN--></code>" </li>
3747: <li> The public identifier starts with: "<code title="">-//W3C//DTD XHTML 1.0 Transitional//<!--EN--></code>" </li>
3748: <li> The system identifier is not missing and the public identifier starts with: "<code title="">-//W3C//DTD HTML 4.01 Frameset//<!--EN--></code>" </li>
3749: <li> The system identifier is not missing and the public identifier starts with: "<code title="">-//W3C//DTD HTML 4.01 Transitional//<!--EN--></code>" </li>
3750: </ul><p>The name, system identifier, and public identifier strings must
3751: be compared to the values given in the lists above in an
3752: <a href="infrastructure.html#ascii-case-insensitive">ASCII case-insensitive</a> manner. A system identifier
3753: whose value is the empty string is not considered missing for the
3754: purposes of the conditions above.</p>
3755:
3756: <p>Then, switch the <a href="#insertion-mode">insertion mode</a> to "<a href="#the-before-html-insertion-mode" title="insertion mode: before html">before html</a>".</p>
3757:
3758: </dd>
3759:
3760: <dt>Anything else</dt>
3761: <dd>
3762:
3763: <p><a href="#parse-error">Parse error</a>.</p>
3764:
3765: <p>Set the <code>Document</code> to <a href="dom.html#quirks-mode">quirks mode</a>.</p>
3766:
3767: <p>Switch the <a href="#insertion-mode">insertion mode</a> to "<a href="#the-before-html-insertion-mode" title="insertion mode: before html">before html</a>", then
3768: reprocess the current token.</p>
3769:
3770: </dd>
3771:
3772: </dl><h5 id="the-before-html-insertion-mode"><span class="secno">9.2.5.5 </span>The "<dfn title="insertion mode: before html">before html</dfn>" insertion mode</h5>
3773:
3774: <p>When the <a href="#insertion-mode">insertion mode</a> is "<a href="#the-before-html-insertion-mode" title="insertion
3775: mode: before html">before html</a>", tokens must be handled as follows:</p>
3776:
3777: <dl class="switch"><dt>A DOCTYPE token</dt>
3778: <dd>
3779: <p><a href="#parse-error">Parse error</a>. Ignore the token.</p>
3780: </dd>
3781:
3782: <dt>A comment token</dt>
3783: <dd>
3784: <p>Append a <code>Comment</code> node to the <code>Document</code>
3785: object with the <code title="">data</code> attribute set to the
3786: data given in the comment token.</p>
3787: </dd>
3788:
3789: <dt>A character token that is one of U+0009 CHARACTER
3790: TABULATION, U+000A LINE FEED (LF), U+000C FORM FEED (FF),
3791: <!--U+000D CARRIAGE RETURN (CR),--> or U+0020 SPACE</dt>
3792: <dd>
3793: <p>Ignore the token.</p>
3794: </dd>
3795:
3796: <dt>A start tag whose tag name is "html"</dt>
3797: <dd>
3798:
3799: <p><a href="#create-an-element-for-the-token">Create an element for the token</a> in the <a href="#html-namespace-0">HTML
3800: namespace</a>. Append it to the <code>Document</code>
3801: object. Put this element in the <a href="#stack-of-open-elements">stack of open
3802: elements</a>.</p>
3803:
3804: <p id="parser-appcache">If the <code>Document</code> is being
3805: loaded as part of <a href="history.html#navigate" title="navigate">navigation</a> of a
3806: <a href="browsers.html#browsing-context">browsing context</a>, then: if the newly created element
3807: has a <code title="attr-html-manifest"><a href="semantics.html#attr-html-manifest">manifest</a></code> attribute,
3808: then <a href="infrastructure.html#resolve-a-url" title="resolve a url">resolve</a> the value of that
3809: attribute to an <a href="infrastructure.html#absolute-url">absolute URL</a>, relative to the newly
3810: created element, and if that is successful, run the <a href="offline.html#concept-appcache-init" title="concept-appcache-init">application cache selection
1.29 mike 3811: algorithm</a> with the resulting <a href="infrastructure.html#absolute-url">absolute URL</a> with
3812: any <a href="infrastructure.html#url-fragment" title="url-fragment"><fragment></a> component
3813: removed; otherwise, if there is no such attribute or resolving it
3814: fails, run the <a href="offline.html#concept-appcache-init" title="concept-appcache-init">application
3815: cache selection algorithm</a> with no manifest. The algorithm
3816: must be passed the <code>Document</code> object.</p>
1.1 mike 3817:
3818: <p>Switch the <a href="#insertion-mode">insertion mode</a> to "<a href="#the-before-head-insertion-mode" title="insertion mode: before head">before head</a>".</p>
3819:
3820: </dd>
3821:
3822: <dt>Anything else</dt>
3823: <dd>
3824:
1.159 mike 3825: <p>Create an <code><a href="semantics.html#the-html-element-0">html</a></code> element. Append it to the
1.1 mike 3826: <code>Document</code> object. Put this element in the <a href="#stack-of-open-elements">stack
3827: of open elements</a>.</p>
3828:
3829: <p>If the <code>Document</code> is being loaded as part of <a href="history.html#navigate" title="navigate">navigation</a> of a <a href="browsers.html#browsing-context">browsing
3830: context</a>, then: run the <a href="offline.html#concept-appcache-init" title="concept-appcache-init">application cache selection
3831: algorithm</a> with no manifest, passing it the
3832: <code>Document</code> object.</p>
3833:
3834: <p>Switch the <a href="#insertion-mode">insertion mode</a> to "<a href="#the-before-head-insertion-mode" title="insertion mode: before head">before head</a>", then
3835: reprocess the current token.</p>
3836:
3837: <p class="XXX">Should probably make end tags be ignored, so
3838: that "</head><!-- --><html>" puts the comment before the
3839: root node (or should we?)</p>
3840:
3841: </dd>
3842:
3843: </dl><p>The root element can end up being removed from the
3844: <code>Document</code> object, e.g. by scripts; nothing in particular
3845: happens in such cases, content continues being appended to the nodes
3846: as described in the next section.</p>
3847:
3848:
3849: <h5 id="the-before-head-insertion-mode"><span class="secno">9.2.5.6 </span>The "<dfn title="insertion mode: before head">before head</dfn>" insertion mode</h5>
3850:
3851: <p>When the <a href="#insertion-mode">insertion mode</a> is "<a href="#the-before-head-insertion-mode" title="insertion
3852: mode: before head">before head</a>", tokens must be handled as follows:</p>
3853:
3854: <dl class="switch"><dt>A character token that is one of U+0009 CHARACTER
3855: TABULATION, U+000A LINE FEED (LF), U+000C FORM FEED (FF),
3856: <!--U+000D CARRIAGE RETURN (CR),--> or U+0020 SPACE</dt>
3857: <dd>
3858: <p>Ignore the token.</p> <!-- :-( -->
3859: </dd>
3860:
3861: <dt>A comment token</dt>
3862: <dd>
3863: <p>Append a <code>Comment</code> node to the <a href="#current-node">current
3864: node</a> with the <code title="">data</code> attribute set to
3865: the data given in the comment token.</p>
3866: </dd>
3867:
3868: <dt>A DOCTYPE token</dt>
3869: <dd>
3870: <p><a href="#parse-error">Parse error</a>. Ignore the token.</p>
3871: </dd>
3872:
3873: <dt>A start tag whose tag name is "html"</dt>
3874: <dd>
3875: <p>Process the token <a href="#using-the-rules-for">using the rules for</a> the "<a href="#parsing-main-inbody" title="insertion mode: in body">in body</a>" <a href="#insertion-mode">insertion
3876: mode</a>.</p>
3877: </dd>
3878:
3879: <dt>A start tag whose tag name is "head"</dt>
3880: <dd>
3881:
3882: <p><a href="#insert-an-html-element">Insert an HTML element</a> for the token.</p>
3883:
3884: <p>Set the <a href="#head-element-pointer"><code title="">head</code> element pointer</a>
1.159 mike 3885: to the newly created <code><a href="semantics.html#the-head-element-0">head</a></code> element.</p>
1.1 mike 3886:
3887: <p>Switch the <a href="#insertion-mode">insertion mode</a> to "<a href="#parsing-main-inhead" title="insertion mode: in head">in head</a>".</p>
3888:
3889: </dd>
3890:
3891: <dt>An end tag whose tag name is one of: "head", "body", "html", "br"</dt>
3892: <dd>
3893:
3894: <p>Act as if a start tag token with the tag name "head" and no
3895: attributes had been seen, then reprocess the current token.</p>
3896:
3897: </dd>
3898:
3899: <dt>Any other end tag</dt>
3900: <dd>
3901:
3902: <p><a href="#parse-error">Parse error</a>. Ignore the token.</p>
3903:
3904: </dd>
3905:
3906: <dt>Anything else</dt>
3907: <dd>
3908:
3909: <p>Act as if a start tag token with the tag name "head" and no
3910: attributes had been seen, then reprocess the current
3911: token.</p>
3912:
3913: </dd>
3914:
3915: </dl><h5 id="parsing-main-inhead"><span class="secno">9.2.5.7 </span>The "<dfn title="insertion mode: in head">in head</dfn>" insertion mode</h5>
3916:
3917: <p>When the <a href="#insertion-mode">insertion mode</a> is "<a href="#parsing-main-inhead" title="insertion
3918: mode: in head">in head</a>", tokens must be handled as follows:</p>
3919:
3920: <dl class="switch"><dt>A character token that is one of U+0009 CHARACTER
3921: TABULATION, U+000A LINE FEED (LF), U+000C FORM FEED (FF),
3922: <!--U+000D CARRIAGE RETURN (CR),--> or U+0020 SPACE</dt>
3923: <dd>
3924: <p><a href="#insert-a-character" title="insert a character">Insert the character</a> into
3925: the <a href="#current-node">current node</a>.</p>
3926: </dd>
3927:
3928: <dt>A comment token</dt>
3929: <dd>
3930: <p>Append a <code>Comment</code> node to the <a href="#current-node">current
3931: node</a> with the <code title="">data</code> attribute set to
3932: the data given in the comment token.</p>
3933: </dd>
3934:
3935: <dt>A DOCTYPE token</dt>
3936: <dd>
3937: <p><a href="#parse-error">Parse error</a>. Ignore the token.</p>
3938: </dd>
3939:
3940: <dt>A start tag whose tag name is "html"</dt>
3941: <dd>
3942: <p>Process the token <a href="#using-the-rules-for">using the rules for</a> the "<a href="#parsing-main-inbody" title="insertion mode: in body">in body</a>" <a href="#insertion-mode">insertion
3943: mode</a>.</p>
3944: </dd>
3945:
3946: <dt>A start tag whose tag name is one of: "base", "command",
3947: "link"</dt>
3948: <dd>
3949:
3950: <p><a href="#insert-an-html-element">Insert an HTML element</a> for the token. Immediately
3951: pop the <a href="#current-node">current node</a> off the <a href="#stack-of-open-elements">stack of open
3952: elements</a>.</p>
3953:
3954: <p><a href="#acknowledge-self-closing-flag" title="acknowledge self-closing flag">Acknowledge the
3955: token's <i>self-closing flag</i></a>, if it is set.</p>
3956:
3957: </dd>
3958:
3959: <dt>A start tag whose tag name is "meta"</dt>
3960: <dd>
3961:
3962: <p><a href="#insert-an-html-element">Insert an HTML element</a> for the token. Immediately
3963: pop the <a href="#current-node">current node</a> off the <a href="#stack-of-open-elements">stack of open
3964: elements</a>.</p>
3965:
3966: <p><a href="#acknowledge-self-closing-flag" title="acknowledge self-closing flag">Acknowledge the
3967: token's <i>self-closing flag</i></a>, if it is set.</p>
3968:
3969: <p id="meta-charset-during-parse">If the element has a <code title="attr-meta-charset"><a href="semantics.html#attr-meta-charset">charset</a></code> attribute, and its
3970: value is a supported encoding, and the <a href="#concept-encoding-confidence" title="concept-encoding-confidence">confidence</a> is
3971: currently <i>tentative</i>, then <a href="#change-the-encoding">change the
3972: encoding</a> to the encoding given by the value of the
3973: <code title="attr-meta-charset"><a href="semantics.html#attr-meta-charset">charset</a></code> attribute.</p>
3974:
3975: <p>Otherwise, if the element has a <code title="attr-meta-charset"><a href="semantics.html#attr-meta-charset">content</a></code> attribute, and
3976: applying the <a href="infrastructure.html#algorithm-for-extracting-an-encoding-from-a-content-type">algorithm for extracting an encoding from a
3977: Content-Type</a> to its value returns a supported encoding
3978: <var title="">encoding</var>, and the <a href="#concept-encoding-confidence" title="concept-encoding-confidence">confidence</a> is
3979: currently <i>tentative</i>, then <a href="#change-the-encoding">change the
3980: encoding</a> to the encoding <var title="">encoding</var>.</p>
3981:
3982: </dd>
3983:
3984: <dt>A start tag whose tag name is "title"</dt>
3985: <dd>
3986: <p>Follow the <a href="#generic-rcdata-element-parsing-algorithm">generic RCDATA element parsing algorithm</a>.</p>
3987: </dd>
3988:
3989: <dt>A start tag whose tag name is "noscript", if the <a href="#scripting-flag">scripting flag</a> is enabled</dt>
3990: <dt>A start tag whose tag name is one of: "noframes", "style"</dt>
3991: <dd>
1.91 mike 3992: <p>Follow the <a href="#generic-raw-text-element-parsing-algorithm">generic raw text element parsing algorithm</a>.</p>
1.1 mike 3993: </dd>
3994:
3995: <dt>A start tag whose tag name is "noscript", if the <a href="#scripting-flag">scripting flag</a> is disabled</dt>
3996: <dd>
3997:
3998: <p><a href="#insert-an-html-element">Insert an HTML element</a> for the token.</p>
3999:
4000: <p>Switch the <a href="#insertion-mode">insertion mode</a> to "<a href="#parsing-main-inheadnoscript" title="insertion mode: in head noscript">in head
4001: noscript</a>".</p>
4002:
4003: </dd>
4004:
4005: <dt id="scriptTag">A start tag whose tag name is "script"</dt>
4006: <dd>
4007:
4008: <ol><li><p><a href="#create-an-element-for-the-token">Create an element for the token</a> in the
4009: <a href="#html-namespace-0">HTML namespace</a>.</p></li>
4010:
4011: <li>
4012:
4013: <p>Mark the element as being <a href="semantics.html#parser-inserted">"parser-inserted"</a>.</p>
4014:
4015: <p class="note">This ensures that, if the script is external,
1.159 mike 4016: any <code title="dom-document-write"><a href="embedded-content-0.html#dom-document-write">document.write()</a></code>
1.1 mike 4017: calls in the script will execute in-line, instead of blowing the
4018: document away, as would happen in most other cases. It also
4019: prevents the script from executing until the end tag is
4020: seen.</p>
4021:
4022: </li>
4023:
4024: <li><p>If the parser was originally created for the <a href="#html-fragment-parsing-algorithm">HTML
4025: fragment parsing algorithm</a>, then mark the
4026: <code><a href="semantics.html#script">script</a></code> element as <a href="semantics.html#already-executed">"already
4027: executed"</a>. (<a href="#fragment-case">fragment case</a>)</p></li>
4028:
4029: <li><p>Append the new element to the <a href="#current-node">current node</a>
4030: and push it onto the <a href="#stack-of-open-elements">stack of open
4031: elements</a>.</p></li>
4032:
4033: <li><p>Switch the tokenizer's <a href="#content-model-flag">content model flag</a> to
1.91 mike 4034: the RAWTEXT state.</p></li>
1.1 mike 4035:
4036: <li><p>Let the <a href="#original-insertion-mode">original insertion mode</a> be the current
4037: <a href="#insertion-mode">insertion mode</a>.</p>
4038:
1.91 mike 4039: </li><li><p>Switch the <a href="#insertion-mode">insertion mode</a> to "<a href="#parsing-main-incdata" title="insertion mode: in RAWTEXT/RCDATA">in
4040: RAWTEXT/RCDATA</a>".</p></li>
1.1 mike 4041:
4042: </ol></dd>
4043:
4044: <dt>An end tag whose tag name is "head"</dt>
4045: <dd>
4046:
4047: <p>Pop the <a href="#current-node">current node</a> (which will be the
1.159 mike 4048: <code><a href="semantics.html#the-head-element-0">head</a></code> element) off the <a href="#stack-of-open-elements">stack of open
1.1 mike 4049: elements</a>.</p>
4050:
4051: <p>Switch the <a href="#insertion-mode">insertion mode</a> to "<a href="#the-after-head-insertion-mode" title="insertion mode: after head">after head</a>".</p>
4052:
4053: </dd>
4054:
4055: <dt>An end tag whose tag name is one of: "body", "html", "br"</dt>
4056: <dd>
4057: <p>Act as described in the "anything else" entry below.</p>
4058: </dd>
4059:
4060: <dt>A start tag whose tag name is "head"</dt>
4061: <dt>Any other end tag</dt>
4062: <dd>
4063: <p><a href="#parse-error">Parse error</a>. Ignore the token.</p>
4064: </dd>
4065:
4066: <dt>Anything else</dt>
4067: <dd>
4068:
4069: <!-- can't get here with an EOF and a fragment case -->
4070:
4071: <p>Act as if an end tag token with the tag name "head" had
4072: been seen, and reprocess the current token.</p>
4073:
4074: <p class="XXX">In certain UAs, <a href="https://bugzilla.mozilla.org/attachment.cgi?id=180157&action=view">some
4075: elements</a> don't trigger the "in body" mode straight away,
4076: but instead get put into the head. Do we want to copy
4077: that?</p>
4078:
4079: </dd>
4080:
4081: </dl><h5 id="parsing-main-inheadnoscript"><span class="secno">9.2.5.8 </span>The "<dfn title="insertion mode: in head noscript">in head noscript</dfn>" insertion mode</h5>
4082:
4083: <p>When the <a href="#insertion-mode">insertion mode</a> is "<a href="#parsing-main-inheadnoscript" title="insertion
4084: mode: in head noscript">in head noscript</a>", tokens must be handled as follows:</p>
4085:
4086: <dl class="switch"><dt>A DOCTYPE token</dt>
4087: <dd>
4088: <p><a href="#parse-error">Parse error</a>. Ignore the token.</p>
4089: </dd>
4090:
4091: <dt>A start tag whose tag name is "html"</dt>
4092: <dd>
4093: <p>Process the token <a href="#using-the-rules-for">using the rules for</a> the "<a href="#parsing-main-inbody" title="insertion mode: in body">in body</a>" <a href="#insertion-mode">insertion
4094: mode</a>.</p>
4095: </dd>
4096:
4097: <dt>An end tag whose tag name is "noscript"</dt>
4098: <dd>
4099:
4100: <p>Pop the <a href="#current-node">current node</a> (which will be a
4101: <code><a href="semantics.html#the-noscript-element">noscript</a></code> element) from the <a href="#stack-of-open-elements">stack of open
4102: elements</a>; the new <a href="#current-node">current node</a> will be a
1.159 mike 4103: <code><a href="semantics.html#the-head-element-0">head</a></code> element.</p>
1.1 mike 4104:
4105: <p>Switch the <a href="#insertion-mode">insertion mode</a> to "<a href="#parsing-main-inhead" title="insertion mode: in head">in head</a>".</p>
4106:
4107: </dd>
4108:
4109: <dt>A character token that is one of U+0009 CHARACTER
4110: TABULATION, U+000A LINE FEED (LF), U+000C FORM FEED (FF),
4111: <!--U+000D CARRIAGE RETURN (CR),--> or U+0020 SPACE</dt>
4112: <dt>A comment token</dt>
4113: <dt>A start tag whose tag name is one of: "link", "meta", "noframes", "style"</dt>
4114: <dd>
4115: <p>Process the token <a href="#using-the-rules-for">using the rules for</a> the "<a href="#parsing-main-inhead" title="insertion mode: in head">in head</a>" <a href="#insertion-mode">insertion
4116: mode</a>.</p>
4117: </dd>
4118:
4119: <dt>An end tag whose tag name is "br"</dt>
4120: <dd>
4121: <p>Act as described in the "anything else" entry below.</p>
4122: </dd>
4123:
4124: <dt>A start tag whose tag name is one of: "head", "noscript"</dt>
4125: <dt>Any other end tag</dt>
4126: <dd>
4127: <p><a href="#parse-error">Parse error</a>. Ignore the token.</p>
4128: </dd>
4129:
4130: <dt>Anything else</dt>
4131: <dd>
4132:
4133: <!-- can't get here with an EOF and a fragment case -->
4134:
4135: <p><a href="#parse-error">Parse error</a>. Act as if an end tag with the tag
4136: name "noscript" had been seen and reprocess the current
4137: token.</p>
4138:
4139: </dd>
4140:
4141: </dl><h5 id="the-after-head-insertion-mode"><span class="secno">9.2.5.9 </span>The "<dfn title="insertion mode: after head">after head</dfn>" insertion mode</h5>
4142:
4143: <p>When the <a href="#insertion-mode">insertion mode</a> is "<a href="#the-after-head-insertion-mode" title="insertion
4144: mode: after head">after head</a>", tokens must be handled as follows:</p>
4145:
4146: <dl class="switch"><dt>A character token that is one of U+0009 CHARACTER
4147: TABULATION, U+000A LINE FEED (LF), U+000C FORM FEED (FF),
4148: <!--U+000D CARRIAGE RETURN (CR),--> or U+0020 SPACE</dt>
4149: <dd>
4150: <p><a href="#insert-a-character" title="insert a character">Insert the character</a> into
4151: the <a href="#current-node">current node</a>.</p>
4152: </dd>
4153:
4154: <dt>A comment token</dt>
4155: <dd>
4156: <p>Append a <code>Comment</code> node to the <a href="#current-node">current
4157: node</a> with the <code title="">data</code> attribute set to
4158: the data given in the comment token.</p>
4159: </dd>
4160:
4161: <dt>A DOCTYPE token</dt>
4162: <dd>
4163: <p><a href="#parse-error">Parse error</a>. Ignore the token.</p>
4164: </dd>
4165:
4166: <dt>A start tag whose tag name is "html"</dt>
4167: <dd>
4168: <p>Process the token <a href="#using-the-rules-for">using the rules for</a> the "<a href="#parsing-main-inbody" title="insertion mode: in body">in body</a>" <a href="#insertion-mode">insertion
4169: mode</a>.</p>
4170: </dd>
4171:
4172: <dt>A start tag whose tag name is "body"</dt>
4173: <dd>
4174:
4175: <p><a href="#insert-an-html-element">Insert an HTML element</a> for the token.</p>
4176:
4177: <p>Set the <a href="#frameset-ok-flag">frameset-ok flag</a> to "not ok".</p>
4178:
4179: <p>Switch the <a href="#insertion-mode">insertion mode</a> to "<a href="#parsing-main-inbody" title="insertion mode: in body">in body</a>".</p>
4180:
4181: </dd>
4182:
4183: <dt>A start tag whose tag name is "frameset"</dt>
4184: <dd>
4185:
4186: <p><a href="#insert-an-html-element">Insert an HTML element</a> for the token.</p>
4187:
4188: <p>Switch the <a href="#insertion-mode">insertion mode</a> to "<a href="#parsing-main-inframeset" title="insertion mode: in frameset">in frameset</a>".</p>
4189:
4190: </dd>
4191:
4192: <dt>A start tag token whose tag name is one of: "base", "link",
4193: "meta", "noframes", "script", "style", "title"</dt>
4194: <dd>
4195:
4196: <p><a href="#parse-error">Parse error</a>.</p>
4197:
4198: <p>Push the node pointed to by the <a href="#head-element-pointer"><code title="">head</code> element pointer</a> onto the
4199: <a href="#stack-of-open-elements">stack of open elements</a>.</p>
4200:
4201: <p>Process the token <a href="#using-the-rules-for">using the rules for</a> the "<a href="#parsing-main-inhead" title="insertion mode: in head">in head</a>" <a href="#insertion-mode">insertion
4202: mode</a>.</p>
4203:
4204: <p>Remove the node pointed to by the <a href="#head-element-pointer"><code title="">head</code> element pointer</a> from the <a href="#stack-of-open-elements">stack
4205: of open elements</a>.</p>
4206:
4207: <p class="note">The <a href="#head-element-pointer"><code title="">head</code> element
4208: pointer</a> cannot be null at this point.</p>
4209:
4210: </dd>
4211:
4212: <dt>An end tag whose tag name is one of: "body", "html", "br"</dt>
4213: <dd>
4214: <p>Act as described in the "anything else" entry below.</p>
4215: </dd>
4216:
4217: <dt>A start tag whose tag name is "head"</dt>
4218: <dt>Any other end tag</dt>
4219: <dd>
4220: <p><a href="#parse-error">Parse error</a>. Ignore the token.</p>
4221: </dd>
4222:
4223: <dt>Anything else</dt>
4224: <dd>
4225: <p>Act as if a start tag token with the tag name "body" and no
4226: attributes had been seen, then set the <a href="#frameset-ok-flag">frameset-ok
4227: flag</a> back to "ok", and then reprocess the current
4228: token.</p>
4229: </dd>
4230:
4231: </dl><h5 id="parsing-main-inbody"><span class="secno">9.2.5.10 </span>The "<dfn title="insertion mode: in body">in body</dfn>" insertion mode</h5>
4232:
4233: <p>When the <a href="#insertion-mode">insertion mode</a> is "<a href="#parsing-main-inbody" title="insertion
4234: mode: in body">in body</a>", tokens must be handled as follows:</p>
4235:
4236: <dl class="switch"><dt>A character token</dt>
4237: <dd>
4238:
4239: <p><a href="#reconstruct-the-active-formatting-elements">Reconstruct the active formatting elements</a>, if
4240: any.</p>
4241:
4242: <p><a href="#insert-a-character" title="insert a character">Insert the token's
4243: character</a> into the <a href="#current-node">current node</a>.</p>
4244:
4245: <p>If the token is not one of U+0009 CHARACTER TABULATION, U+000A
4246: LINE FEED (LF), U+000C FORM FEED (FF), <!--U+000D CARRIAGE RETURN
4247: (CR),--> or U+0020 SPACE, then set the <a href="#frameset-ok-flag">frameset-ok
4248: flag</a> to "not ok".</p>
4249:
4250: </dd>
4251:
4252: <dt>A comment token</dt>
4253: <dd>
4254: <p>Append a <code>Comment</code> node to the <a href="#current-node">current
4255: node</a> with the <code title="">data</code> attribute set to
4256: the data given in the comment token.</p>
4257: </dd>
4258:
4259: <dt>A DOCTYPE token</dt>
4260: <dd>
4261: <p><a href="#parse-error">Parse error</a>. Ignore the token.</p>
4262: </dd>
4263:
4264: <dt>A start tag whose tag name is "html"</dt>
4265: <dd>
4266: <p><a href="#parse-error">Parse error</a>. For each attribute on the token,
4267: check to see if the attribute is already present on the top
4268: element of the <a href="#stack-of-open-elements">stack of open elements</a>. If it is not,
4269: add the attribute and its corresponding value to that element.</p>
4270: </dd>
4271:
4272: <dt>A start tag token whose tag name is one of: "base", "command",
4273: "link", "meta", "noframes", "script", "style", "title"</dt>
4274: <dd>
4275: <p>Process the token <a href="#using-the-rules-for">using the rules for</a> the "<a href="#parsing-main-inhead" title="insertion mode: in head">in head</a>" <a href="#insertion-mode">insertion
4276: mode</a>.</p>
4277: </dd>
4278:
4279: <dt>A start tag whose tag name is "body"</dt>
4280: <dd>
4281:
4282: <p><a href="#parse-error">Parse error</a>.</p>
4283:
4284: <p>If the second element on the <a href="#stack-of-open-elements">stack of open
1.159 mike 4285: elements</a> is not a <code><a href="semantics.html#the-body-element-0">body</a></code> element, or, if the
1.1 mike 4286: <a href="#stack-of-open-elements">stack of open elements</a> has only one node on it,
4287: then ignore the token. (<a href="#fragment-case">fragment case</a>)</p>
4288:
4289: <p>Otherwise, for each attribute on the token, check to see if
1.159 mike 4290: the attribute is already present on the <code><a href="semantics.html#the-body-element-0">body</a></code>
1.1 mike 4291: element (the second element) on the <a href="#stack-of-open-elements">stack of open
4292: elements</a>. If it is not, add the attribute and its
4293: corresponding value to that element.</p>
4294:
4295: </dd>
4296:
4297: <dt>A start tag whose tag name is "frameset"</dt>
4298: <dd>
4299:
4300: <p><a href="#parse-error">Parse error</a>.</p>
4301:
4302: <p>If the second element on the <a href="#stack-of-open-elements">stack of open
1.159 mike 4303: elements</a> is not a <code><a href="semantics.html#the-body-element-0">body</a></code> element, or, if the
1.1 mike 4304: <a href="#stack-of-open-elements">stack of open elements</a> has only one node on it,
4305: then ignore the token. (<a href="#fragment-case">fragment case</a>)</p>
4306:
4307: <p>If the <a href="#frameset-ok-flag">frameset-ok flag</a> is set to "not ok", ignore
4308: the token.</p>
4309:
4310: <p>Otherwise, run the following steps:</p>
4311:
4312: <ol><li><p>Remove the second element on the <a href="#stack-of-open-elements">stack of open
4313: elements</a> from its parent node, if it has one.</p></li>
4314:
4315: <li><p>Pop all the nodes from the bottom of the <a href="#stack-of-open-elements">stack of
4316: open elements</a>, from the <a href="#current-node">current node</a> up to,
1.159 mike 4317: but not including, the root <code><a href="semantics.html#the-html-element-0">html</a></code> element.</p>
1.1 mike 4318:
4319: </li><li><p><a href="#insert-an-html-element">Insert an HTML element</a> for the
4320: token.</p></li>
4321:
4322: <li><p>Switch the <a href="#insertion-mode">insertion mode</a> to "<a href="#parsing-main-inframeset" title="insertion mode: in frameset">in frameset</a>".</p>
4323:
4324: </li></ol></dd>
4325:
4326: <dt>An end-of-file token</dt>
4327: <dd>
4328:
4329: <p>If there is a node in the <a href="#stack-of-open-elements">stack of open elements</a>
4330: that is not either a <code><a href="semantics.html#the-dd-element">dd</a></code> element, a <code><a href="semantics.html#the-dt-element">dt</a></code>
4331: element, an <code><a href="semantics.html#the-li-element">li</a></code> element, a <code><a href="semantics.html#the-p-element">p</a></code> element, a
4332: <code><a href="tabular-data.html#the-tbody-element">tbody</a></code> element, a <code><a href="tabular-data.html#the-td-element">td</a></code> element, a
4333: <code><a href="tabular-data.html#the-tfoot-element">tfoot</a></code> element, a <code><a href="tabular-data.html#the-th-element">th</a></code> element, a
4334: <code><a href="tabular-data.html#the-thead-element">thead</a></code> element, a <code><a href="tabular-data.html#the-tr-element">tr</a></code> element, the
1.159 mike 4335: <code><a href="semantics.html#the-body-element-0">body</a></code> element, or the <code><a href="semantics.html#the-html-element-0">html</a></code> element, then
1.1 mike 4336: this is a <a href="#parse-error">parse error</a>.</p> <!-- (some of those are
4337: fragment cases) -->
4338:
4339: <p><a href="#stop-parsing">Stop parsing</a>.</p>
4340:
4341: </dd>
4342:
4343: <dt>An end tag whose tag name is "body"</dt>
4344: <dd>
4345:
4346: <p>If the <a href="#stack-of-open-elements">stack of open elements</a> does not <a href="#has-an-element-in-scope" title="has an element in scope">have a <code>body</code> element
4347: in scope</a>, this is a <a href="#parse-error">parse error</a>; ignore the
4348: token.</p>
4349:
4350: <!-- if we get here, the insertion mode here is forcibly "in
4351: body". -->
4352:
4353: <p>Otherwise, if there is a node in the <a href="#stack-of-open-elements">stack of open
4354: elements</a> that is not either a <code><a href="semantics.html#the-dd-element">dd</a></code> element, a
4355: <code><a href="semantics.html#the-dt-element">dt</a></code> element, an <code><a href="semantics.html#the-li-element">li</a></code> element, an
4356: <code><a href="forms.html#the-optgroup-element">optgroup</a></code> element, an <code><a href="forms.html#the-option-element">option</a></code> element, a
4357: <code><a href="semantics.html#the-p-element">p</a></code> element, an <code><a href="text-level-semantics.html#the-rp-element">rp</a></code> element, an
4358: <code><a href="text-level-semantics.html#the-rt-element">rt</a></code> element, a <code><a href="tabular-data.html#the-tbody-element">tbody</a></code> element, a
4359: <code><a href="tabular-data.html#the-td-element">td</a></code> element, a <code><a href="tabular-data.html#the-tfoot-element">tfoot</a></code> element, a
4360: <code><a href="tabular-data.html#the-th-element">th</a></code> element, a <code><a href="tabular-data.html#the-thead-element">thead</a></code> element, a
1.159 mike 4361: <code><a href="tabular-data.html#the-tr-element">tr</a></code> element, the <code><a href="semantics.html#the-body-element-0">body</a></code> element, or the
4362: <code><a href="semantics.html#the-html-element-0">html</a></code> element, then this is a <a href="#parse-error">parse
1.1 mike 4363: error</a>.</p> <!-- (some of those are fragment cases, e.g. for
4364: <tbody> you'd have hit the first paragraph since the <body>
4365: wouldn't be in scope, unless it was a fragment case) -->
4366:
4367: <!-- If we ever change the frameset-ok flag to an insertion mode,
4368: then we'd have to somehow keep track of its state when we switch
4369: to after-body. -->
4370:
4371: <p>Switch the <a href="#insertion-mode">insertion mode</a> to "<a href="#parsing-main-afterbody" title="insertion mode: after body">after body</a>".</p>
4372:
4373: </dd>
4374:
4375: <dt>An end tag whose tag name is "html"</dt>
4376: <dd>
4377:
4378: <p>Act as if an end tag with tag name "body" had been seen,
4379: then, if that token wasn't ignored, reprocess the current
4380: token.</p>
4381:
4382: <p class="note">The fake end tag token here can only be
4383: ignored in the <a href="#fragment-case">fragment case</a>.</p>
4384:
4385: </dd>
4386:
4387: <!-- start tags for non-phrasing flow content elements -->
4388:
4389: <!-- the normal ones -->
4390: <dt>A start tag whose tag name is one of: "address", "article",
1.86 mike 4391: "aside", "blockquote", "center", <!--v2DATAGRID-->"datagrid",
4392: "details", "dialog", "dir", "div", "dl", "fieldset", "figure",
4393: "footer", "header", "hgroup", "menu", "nav", "ol", "p", "section",
4394: "ul"</dt>
1.1 mike 4395: <dd>
4396:
4397: <!-- As of May 2008 this doesn't match any browser exactly, but is
4398: as close to what IE does as I can get without doing the non-tree
4399: DOM nonsense, and thus should actually afford better compatibility
4400: when implemented by the other browsers. -->
4401:
4402: <p>If the <a href="#stack-of-open-elements">stack of open elements</a> <a href="#has-an-element-in-scope" title="has
4403: an element in scope">has a <code>p</code> element in
4404: scope</a>, then act as if an end tag with the tag name
4405: "p" had been seen.</p>
4406:
4407: <p><a href="#insert-an-html-element">Insert an HTML element</a> for the token.</p>
4408:
4409: </dd>
4410:
4411: <!-- as normal, but close h1-h6 if it's the current node -->
4412: <dt>A start tag whose tag name is one of: "h1", "h2", "h3", "h4",
4413: "h5", "h6"</dt>
4414: <dd>
4415:
4416: <p>If the <a href="#stack-of-open-elements">stack of open elements</a> <a href="#has-an-element-in-scope" title="has
4417: an element in scope">has a <code>p</code> element in
4418: scope</a>, then act as if an end tag with the tag name
4419: "p" had been seen.</p>
4420:
4421: <p>If the <a href="#current-node">current node</a> is an element whose tag name
4422: is one of "h1", "h2", "h3", "h4", "h5", or "h6", then this is a
4423: <a href="#parse-error">parse error</a>; pop the <a href="#current-node">current node</a> off
4424: the <a href="#stack-of-open-elements">stack of open elements</a>.</p>
4425: <!-- See https://bugs.webkit.org/show_bug.cgi?id=12646 -->
4426:
4427: <p><a href="#insert-an-html-element">Insert an HTML element</a> for the token.</p>
4428:
4429: </dd>
4430:
4431: <!-- as normal, but drops leading newline -->
4432: <dt>A start tag whose tag name is one of: "pre", "listing"</dt>
4433: <dd>
4434:
4435: <p>If the <a href="#stack-of-open-elements">stack of open elements</a> <a href="#has-an-element-in-scope" title="has
4436: an element in scope">has a <code>p</code> element in
4437: scope</a>, then act as if an end tag with the tag name
4438: "p" had been seen.</p>
4439:
4440: <p><a href="#insert-an-html-element">Insert an HTML element</a> for the token.</p>
4441:
4442: <p>If the next token is a U+000A LINE FEED (LF) character
4443: token, then ignore that token and move on to the next
4444: one. (Newlines at the start of <code><a href="semantics.html#the-pre-element">pre</a></code> blocks are
4445: ignored as an authoring convenience.)</p>
4446:
4447: <p>Set the <a href="#frameset-ok-flag">frameset-ok flag</a> to "not ok".</p>
4448:
4449: </dd>
4450:
4451: <!-- as normal, but interacts with the form element pointer -->
4452: <dt>A start tag whose tag name is "form"</dt>
4453: <dd>
4454:
4455: <p>If the <a href="#form-element-pointer"><code title="form">form</code> element
4456: pointer</a> is not null, then this is a <a href="#parse-error">parse
4457: error</a>; ignore the token.</p>
4458:
4459: <p>Otherwise:</p>
4460:
4461: <p>If the <a href="#stack-of-open-elements">stack of open elements</a> <a href="#has-an-element-in-scope" title="has
4462: an element in scope">has a <code>p</code> element in
4463: scope</a>, then act as if an end tag with the tag name
4464: "p" had been seen.</p>
4465:
4466: <p><a href="#insert-an-html-element">Insert an HTML element</a> for the token, and set the
4467: <a href="#form-element-pointer"><code title="form">form</code> element pointer</a> to
4468: point to the element created.</p>
4469:
4470: </dd>
4471:
4472: <!-- as normal, but imply </li> when there's another <li> open in weird cases -->
4473: <dt>A start tag whose tag name is "li"</dt>
4474: <dd>
4475:
4476: <p>Run the following algorithm:</p>
4477:
4478: <ol><li><p>Set the <a href="#frameset-ok-flag">frameset-ok flag</a> to "not ok".</p></li>
4479:
4480: <li><p>Initialize <var title="">node</var> to be the <a href="#current-node">current
4481: node</a> (the bottommost node of the stack).</p></li>
4482:
4483: <li><p>If <var title="">node</var> is an <code><a href="semantics.html#the-li-element">li</a></code> element,
4484: then act as if an end tag with the tag name "li" had
4485: been seen, then jump to the last step.</p></li>
4486:
4487: <li><p><i>Loop</i>: If <var title="">node</var> is not in the
4488: <a href="#formatting">formatting</a> category, and is not in the
4489: <a href="#phrasing">phrasing</a> category, and is not an
4490: <code><a href="semantics.html#the-address-element">address</a></code>, <code><a href="interactive-elements.html#the-div-element">div</a></code>, or <code><a href="semantics.html#the-p-element">p</a></code>
4491: element, then jump to the last step.</p></li> <!-- an element
4492: <foo> is in this list if the following markup:
4493:
4494: <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd"><body><ol><li><foo><li>
4495:
4496: ...results in the second <li> not being (in any way) a descendant
4497: of the first <li>, or if <foo> is a formatting element that gets
4498: reopened later. -->
4499:
4500: <li><p>Otherwise, set <var title="">node</var> to the previous
4501: entry in the <a href="#stack-of-open-elements">stack of open elements</a> and return to
4502: the step labeled <i>loop</i>.</p></li>
4503:
4504: <li>
4505:
4506: <p>This is the last step.</p>
4507:
4508: <p>If the <a href="#stack-of-open-elements">stack of open elements</a> <a href="#has-an-element-in-scope" title="has
4509: an element in scope">has a <code>p</code> element in
4510: scope</a>, then act as if an end tag with the tag name
4511: "p" had been seen.</p>
4512:
4513: <p>Finally, <a href="#insert-an-html-element">insert an HTML element</a> for the
4514: token.</p>
4515:
4516: </li>
4517:
4518: </ol></dd>
4519:
4520: <!-- as normal, but imply </dt> or </dd> when there's another <dt> or <dd> open in weird cases -->
4521: <dt>A start tag whose tag name is one of: "dd", "dt"</dt>
4522: <dd>
4523:
4524: <p>Run the following algorithm:</p>
4525:
4526: <ol><li><p>Set the <a href="#frameset-ok-flag">frameset-ok flag</a> to "not ok".</p></li>
4527:
4528: <li><p>Initialize <var title="">node</var> to be the <a href="#current-node">current
4529: node</a> (the bottommost node of the stack).</p></li>
4530:
4531: <li><p><i>Loop</i>: If <var title="">node</var> is a
4532: <code><a href="semantics.html#the-dd-element">dd</a></code> or <code><a href="semantics.html#the-dt-element">dt</a></code> element, then act as if an end
4533: tag with the same tag name as <var title="">node</var> had been
4534: seen, then jump to the last step.</p></li>
4535:
4536: <li><p>If <var title="">node</var> is not in the
4537: <a href="#formatting">formatting</a> category, and is not in the
4538: <a href="#phrasing">phrasing</a> category, and is not an
4539: <code><a href="semantics.html#the-address-element">address</a></code>, <code><a href="interactive-elements.html#the-div-element">div</a></code>, or <code><a href="semantics.html#the-p-element">p</a></code>
4540: element, then jump to the last step.</p></li> <!-- an element
4541: <foo> is in this list if the following markup:
4542:
4543: <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd"><body><dl><dt><foo><dt>
4544:
4545: ...results in the second <dt> not being (in any way) a descendant
4546: of the first <dt>, or if <foo> is a formatting element that gets
4547: reopened later. -->
4548:
4549: <li><p>Otherwise, set <var title="">node</var> to the previous
4550: entry in the <a href="#stack-of-open-elements">stack of open elements</a> and return to
4551: the step labeled <i>loop</i>.</p></li>
4552:
4553: <li>
4554:
4555: <p>This is the last step.</p>
4556:
4557: <p>If the <a href="#stack-of-open-elements">stack of open elements</a> <a href="#has-an-element-in-scope" title="has
4558: an element in scope">has a <code>p</code> element in
4559: scope</a>, then act as if an end tag with the tag name
4560: "p" had been seen.</p>
4561:
4562: <p>Finally, <a href="#insert-an-html-element">insert an HTML element</a> for the
4563: token.</p>
4564:
4565: </li>
4566:
4567: </ol></dd>
4568:
4569: <!-- same as normal, but effectively ends parsing -->
4570: <dt>A start tag whose tag name is "plaintext"</dt>
4571: <dd>
4572:
4573: <p>If the <a href="#stack-of-open-elements">stack of open elements</a> <a href="#has-an-element-in-scope" title="has
4574: an element in scope">has a <code>p</code> element in
4575: scope</a>, then act as if an end tag with the tag name
4576: "p" had been seen.</p>
4577:
4578: <p><a href="#insert-an-html-element">Insert an HTML element</a> for the token.</p>
4579:
4580: <p>Switch the <a href="#content-model-flag">content model flag</a> to the PLAINTEXT
4581: state.</p>
4582:
4583: <p class="note">Once a start tag with the tag name "plaintext"
4584: has been seen, that will be the last token ever seen other
4585: than character tokens (and the end-of-file token), because
4586: there is no way to switch the <a href="#content-model-flag">content model flag</a>
4587: out of the PLAINTEXT state.</p>
4588:
4589: </dd>
4590:
4591: <!-- end tags for non-phrasing flow content elements -->
4592:
4593: <!-- the normal ones -->
4594: <dt>An end tag whose tag name is one of: "address", "article",
1.86 mike 4595: "aside", "blockquote", "center", <!--v2DATAGRID-->"datagrid",
4596: "details", "dialog", "dir", "div", "dl", "fieldset", "figure",
4597: "footer", "header", "hgroup", "listing", "menu", "nav", "ol",
4598: "pre", "section", "ul"</dt>
1.1 mike 4599: <dd>
4600:
4601: <p>If the <a href="#stack-of-open-elements">stack of open elements</a> does not <a href="#has-an-element-in-scope" title="has an element in scope">have an element in scope</a>
4602: with the same tag name as that of the token, then this is a
4603: <a href="#parse-error">parse error</a>; ignore the token.</p>
4604:
4605: <p>Otherwise, run these steps:</p>
4606:
4607: <ol><li><p><a href="#generate-implied-end-tags">Generate implied end tags</a>.</p></li>
4608:
4609: <li><p>If the <a href="#current-node">current node</a> is not an element with
4610: the same tag name as that of the token, then this is a
4611: <a href="#parse-error">parse error</a>.</p></li>
4612:
4613: <li><p>Pop elements from the <a href="#stack-of-open-elements">stack of open elements</a>
4614: until an element with the same tag name as the token has been
4615: popped from the stack.</p></li>
4616:
4617: </ol></dd>
4618:
4619: <!-- removes the form element pointer instead of the matching node -->
4620: <dt>An end tag whose tag name is "form"</dt>
4621: <dd>
4622:
4623: <p>Let <var title="">node</var> be the element that the
4624: <a href="#form-element-pointer"><code title="">form</code> element pointer</a> is set
4625: to.</p>
4626:
4627: <p>Set the <a href="#form-element-pointer"><code title="">form</code> element pointer</a>
4628: to null.</p>
4629:
4630: <p>If <var title="">node</var> is null or the <a href="#stack-of-open-elements">stack of open
4631: elements</a> does not <a href="#has-an-element-in-scope" title="has an element in
4632: scope">have <var title="">node</var> in scope</a>, then this is
4633: a <a href="#parse-error">parse error</a>; ignore the token.</p>
4634:
4635: <p>Otherwise, run these steps:</p>
4636:
4637: <ol><li><p><a href="#generate-implied-end-tags">Generate implied end tags</a>.</p></li>
4638:
4639: <li><p>If the <a href="#current-node">current node</a> is not <var title="">node</var>, then this is a <a href="#parse-error">parse
4640: error</a>.</p></li>
4641:
4642: <li><p>Remove <var title="">node</var> from the <a href="#stack-of-open-elements">stack of
4643: open elements</a>.</p></li>
4644:
4645: </ol></dd>
4646:
4647: <!-- as normal, except </p> implies <p> if there's no <p> in scope, and needs care as the elements have optional tags -->
4648: <dt>An end tag whose tag name is "p"</dt>
4649: <dd>
4650:
4651: <p>If the <a href="#stack-of-open-elements">stack of open elements</a> does not <a href="#has-an-element-in-scope" title="has an element in scope">have an element in scope</a>
4652: with the same tag name as that of the token, then this is a
4653: <a href="#parse-error">parse error</a>; act as if a start tag with the tag name
4654: "p" had been seen, then reprocess the current token.</p>
4655:
4656: <p>Otherwise, run these steps:</p>
4657:
4658: <ol><li><p><a href="#generate-implied-end-tags">Generate implied end tags</a>, except
4659: for elements with the same tag name as the token.</p></li>
4660:
4661: <li><p>If the <a href="#current-node">current node</a> is not an element with
4662: the same tag name as that of the token, then this is a
4663: <a href="#parse-error">parse error</a>.</p></li>
4664:
4665: <li><p>Pop elements from the <a href="#stack-of-open-elements">stack of open elements</a>
4666: until an element with the same tag name as the token has been
4667: popped from the stack.</p></li>
4668:
4669: </ol></dd>
4670:
1.234 mike 4671: <!-- as normal, but needs care as the elements have optional tags, and are further scoped by <ol>/<ul> -->
4672: <dt>An end tag whose tag name is "li"</dt>
4673: <dd>
4674:
4675: <p>If the <a href="#stack-of-open-elements">stack of open elements</a> does not <a href="#has-an-element-in-list-item-scope" title="has an element in list item scope">have an element in list
4676: item scope</a> with the same tag name as that of the token,
4677: then this is a <a href="#parse-error">parse error</a>; ignore the token.</p>
4678:
4679: <p>Otherwise, run these steps:</p>
4680:
4681: <ol><li><p><a href="#generate-implied-end-tags">Generate implied end tags</a>, except
4682: for elements with the same tag name as the token.</p></li>
4683:
4684: <li><p>If the <a href="#current-node">current node</a> is not an element with
4685: the same tag name as that of the token, then this is a
4686: <a href="#parse-error">parse error</a>.</p></li>
4687:
4688: <li><p>Pop elements from the <a href="#stack-of-open-elements">stack of open elements</a>
4689: until an element with the same tag name as the token has been
4690: popped from the stack.</p></li>
4691:
4692: </ol></dd>
4693:
1.1 mike 4694: <!-- as normal, but needs care as the elements have optional tags -->
1.234 mike 4695: <dt>An end tag whose tag name is one of: "dd", "dt"</dt>
1.1 mike 4696: <dd>
4697:
4698: <p>If the <a href="#stack-of-open-elements">stack of open elements</a> does not <a href="#has-an-element-in-scope" title="has an element in scope">have an element in scope</a>
4699: with the same tag name as that of the token, then this is a
4700: <a href="#parse-error">parse error</a>; ignore the token.</p>
4701:
4702: <p>Otherwise, run these steps:</p>
4703:
4704: <ol><li><p><a href="#generate-implied-end-tags">Generate implied end tags</a>, except
4705: for elements with the same tag name as the token.</p></li>
4706:
4707: <li><p>If the <a href="#current-node">current node</a> is not an element with
4708: the same tag name as that of the token, then this is a
4709: <a href="#parse-error">parse error</a>.</p></li>
4710:
4711: <li><p>Pop elements from the <a href="#stack-of-open-elements">stack of open elements</a>
4712: until an element with the same tag name as the token has been
4713: popped from the stack.</p></li>
4714:
4715: </ol></dd>
4716:
4717: <!-- as normal, except acts as a closer for any of the h1-h6 elements -->
4718: <dt>An end tag whose tag name is one of: "h1", "h2", "h3", "h4", "h5", "h6"</dt>
4719: <dd>
4720:
4721: <p>If the <a href="#stack-of-open-elements">stack of open elements</a> does not <a href="#has-an-element-in-scope" title="has an element in scope">have an element in scope</a>
4722: whose tag name is one of "h1", "h2", "h3", "h4", "h5", or "h6",
4723: then this is a <a href="#parse-error">parse error</a>; ignore the token.</p>
4724:
4725: <p>Otherwise, run these steps:</p>
4726:
4727: <ol><li><p><a href="#generate-implied-end-tags">Generate implied end tags</a>.</p></li>
4728:
4729: <li><p>If the <a href="#current-node">current node</a> is not an element with
4730: the same tag name as that of the token, then this is a
4731: <a href="#parse-error">parse error</a>.</p></li>
4732:
4733: <li><p>Pop elements from the <a href="#stack-of-open-elements">stack of open elements</a>
4734: until an element whose tag name is one of "h1", "h2", "h3", "h4",
4735: "h5", or "h6" has been popped from the stack.</p></li>
4736:
4737: </ol></dd>
4738:
4739: <dt>An end tag whose tag name is "sarcasm"</dt>
4740: <dd>
4741: <p>Take a deep breath, then act as described in the "any other end
4742: tag" entry below.</p>
4743: </dd>
4744:
4745: <!-- ADOPTION AGENCY ELEMENTS
4746: Mozilla-only: bdo blink del ins sub sup q
4747: Safari-only: code dfn kbd nobr samp var wbr
4748: Both: a b big em font i s small strike strong tt u -->
4749:
4750: <dt>A start tag whose tag name is "a"</dt>
4751: <dd>
4752:
4753: <p>If the <a href="#list-of-active-formatting-elements">list of active formatting elements</a>
4754: contains an element whose tag name is "a" between the end of
4755: the list and the last marker on the list (or the start of the
4756: list if there is no marker on the list), then this is a
4757: <a href="#parse-error">parse error</a>; act as if an end tag with the tag
4758: name "a" had been seen, then remove that element from the
4759: <a href="#list-of-active-formatting-elements">list of active formatting elements</a> and the
4760: <a href="#stack-of-open-elements">stack of open elements</a> if the end tag didn't
4761: already remove it (it might not have if the element is not
4762: <a href="#has-an-element-in-table-scope" title="has an element in table scope">in table
4763: scope</a>).</p>
4764:
4765: <p class="example">In the non-conforming stream
4766: <code><a href="a">a<table><a href="b">b</table>x</code>,
4767: the first <code><a href="text-level-semantics.html#the-a-element">a</a></code> element would be closed upon seeing
4768: the second one, and the "x" character would be inside a link
4769: to "b", not to "a". This is despite the fact that the outer
4770: <code><a href="text-level-semantics.html#the-a-element">a</a></code> element is not in table scope (meaning that a
4771: regular <code></a></code> end tag at the start of the table
4772: wouldn't close the outer <code><a href="text-level-semantics.html#the-a-element">a</a></code> element).</p>
4773:
4774: <p><a href="#reconstruct-the-active-formatting-elements">Reconstruct the active formatting elements</a>, if
4775: any.</p>
4776:
4777: <p><a href="#insert-an-html-element">Insert an HTML element</a> for the token. Add that
4778: element to the <a href="#list-of-active-formatting-elements">list of active formatting
4779: elements</a>.</p>
4780:
4781: </dd>
4782:
4783: <dt>A start tag whose tag name is one of: "b", "big", "code", "em",
4784: "font", "i", "s", "small", "strike", "strong", "tt", "u"</dt>
4785: <dd>
4786:
4787: <p><a href="#reconstruct-the-active-formatting-elements">Reconstruct the active formatting elements</a>, if
4788: any.</p>
4789:
4790: <p><a href="#insert-an-html-element">Insert an HTML element</a> for the token. Add that
4791: element to the <a href="#list-of-active-formatting-elements">list of active formatting
4792: elements</a>.</p>
4793:
4794: </dd>
4795:
4796: <dt>A start tag whose tag name is "nobr"</dt>
4797: <dd>
4798:
4799: <p><a href="#reconstruct-the-active-formatting-elements">Reconstruct the active formatting elements</a>, if
4800: any.</p>
4801:
4802: <p>If the <a href="#stack-of-open-elements">stack of open elements</a> <a href="#has-an-element-in-scope" title="has an
4803: element in scope">has a <code>nobr</code> element in scope</a>,
4804: then this is a <a href="#parse-error">parse error</a>; act as if an end tag with
4805: the tag name "nobr" had been seen, then once again
4806: <a href="#reconstruct-the-active-formatting-elements">reconstruct the active formatting elements</a>, if
4807: any.</p>
4808:
4809: <p><a href="#insert-an-html-element">Insert an HTML element</a> for the token. Add that
4810: element to the <a href="#list-of-active-formatting-elements">list of active formatting
4811: elements</a>.</p>
4812:
4813: </dd>
4814:
4815: <dt id="adoptionAgency">An end tag whose tag name is one of: "a",
4816: "b", "big", "code", "em", "font", "i", "nobr", "s", "small",
4817: "strike", "strong", "tt", "u"</dt>
4818: <dd>
4819:
4820: <p>Follow these steps:</p>
4821:
4822: <ol><li>
4823:
4824: <p>Let the <var title="">formatting element</var> be the
4825: last element in the <a href="#list-of-active-formatting-elements">list of active formatting
4826: elements</a> that:</p>
4827:
4828: <ul><li>is between the end of the list and the last scope
4829: marker in the list, if any, or the start of the list
4830: otherwise, and</li>
4831:
4832: <li>has the same tag name as the token.</li>
4833:
4834: </ul><p>If there is no such node, or, if that node is also in the
4835: <a href="#stack-of-open-elements">stack of open elements</a> but the element is not <a href="#has-an-element-in-scope" title="has an element in scope">in scope</a>, then this is a
4836: <a href="#parse-error">parse error</a>; ignore the token, and abort these
4837: steps.</p>
4838:
4839: <p>Otherwise, if there is such a node, but that node is not
4840: in the <a href="#stack-of-open-elements">stack of open elements</a>, then this is a
4841: <a href="#parse-error">parse error</a>; remove the element from the list,
4842: and abort these steps.</p>
4843:
4844: <p>Otherwise, there is a <var title="">formatting
4845: element</var> and that element is in <a href="#stack-of-open-elements" title="stack of
4846: open elements">the stack</a> and is <a href="#has-an-element-in-scope" title="has an
4847: element in scope">in scope</a>. If the element is not the
4848: <a href="#current-node">current node</a>, this is a <a href="#parse-error">parse
4849: error</a>. In any case, proceed with the algorithm as
4850: written in the following steps.</p>
4851:
4852: </li>
4853:
4854: <li><p>Let the <var title="">furthest block</var> be the
4855: topmost node in the <a href="#stack-of-open-elements">stack of open elements</a> that
4856: is lower in the stack than the <var title="">formatting
4857: element</var>, and is not an element in the
4858: <a href="#phrasing">phrasing</a> or <a href="#formatting">formatting</a>
4859: categories. There might not be one.</p></li>
4860:
4861: <li><p>If there is no <var title="">furthest block</var>,
4862: then the UA must skip the subsequent steps and instead just
4863: pop all the nodes from the bottom of the <a href="#stack-of-open-elements">stack of open
4864: elements</a>, from the <a href="#current-node">current node</a> up to and
4865: including the <var title="">formatting element</var>, and
4866: remove the <var title="">formatting element</var> from the
4867: <a href="#list-of-active-formatting-elements">list of active formatting elements</a>.</p></li>
4868:
4869: <li><p>Let the <var title="">common ancestor</var> be the element
4870: immediately above the <var title="">formatting element</var> in the
4871: <a href="#stack-of-open-elements">stack of open elements</a>.</p></li>
4872:
4873: <li><p>Let a bookmark note the position of the <var title="">formatting element</var> in the <a href="#list-of-active-formatting-elements">list of active
4874: formatting elements</a> relative to the elements on either
4875: side of it in the list.</p></li>
4876:
4877: <li>
4878:
4879: <p>Let <var title="">node</var> and <var title="">last node</var> be the
4880: <var title="">furthest block</var>. Follow these steps:</p>
4881:
4882: <ol><li>Let <var title="">node</var> be the element immediately
4883: above <var title="">node</var> in the <a href="#stack-of-open-elements">stack of open
4884: elements</a>.</li>
4885:
4886: <li>If <var title="">node</var> is not in the <a href="#list-of-active-formatting-elements">list of
4887: active formatting elements</a>, then remove <var title="">node</var> from the <a href="#stack-of-open-elements">stack of open
4888: elements</a> and then go back to step 1.</li>
4889:
4890: <li>Otherwise, if <var title="">node</var> is the <var title="">formatting element</var>, then go to the next step
4891: in the overall algorithm.</li>
4892:
4893: <li>Otherwise, if <var title="">last node</var> is the <var title="">furthest block</var>, then move the aforementioned
4894: bookmark to be immediately after the <var title="">node</var> in the <a href="#list-of-active-formatting-elements">list of active formatting
4895: elements</a>.</li>
4896:
4897: <li><a href="#create-an-element-for-the-token">Create an element for the token</a> for which the
4898: element <var title="">node</var> was created, replace the entry
4899: for <var title="">node</var> in the <a href="#list-of-active-formatting-elements">list of active
4900: formatting elements</a> with an entry for the new element,
4901: replace the entry for <var title="">node</var> in the
4902: <a href="#stack-of-open-elements">stack of open elements</a> with an entry for the new
4903: element, and let <var title="">node</var> be the new
4904: element.</li>
4905:
4906: <li>Insert <var title="">last node</var> into <var title="">node</var>, first removing it from its previous
4907: parent node if any.</li>
4908:
4909: <li>Let <var title="">last node</var> be <var title="">node</var>.</li>
4910:
4911: <li>Return to step 1 of this inner set of steps.</li>
4912:
4913: </ol></li>
4914:
4915: <li>
4916:
4917: <p>If the <var title="">common ancestor</var> node is a
4918: <code><a href="tabular-data.html#the-table-element">table</a></code>, <code><a href="tabular-data.html#the-tbody-element">tbody</a></code>, <code><a href="tabular-data.html#the-tfoot-element">tfoot</a></code>,
4919: <code><a href="tabular-data.html#the-thead-element">thead</a></code>, or <code><a href="tabular-data.html#the-tr-element">tr</a></code> element, then,
4920: <a href="#foster-parent">foster parent</a> whatever <var title="">last
4921: node</var> ended up being in the previous step, first removing
4922: it from its previous parent node if any.</p>
4923:
4924: <p>Otherwise, append whatever <var title="">last node</var>
4925: ended up being in the previous step to the <var title="">common
4926: ancestor</var> node, first removing it from its previous parent
4927: node if any.</p>
4928:
4929: </li>
4930:
4931: <li><p><a href="#create-an-element-for-the-token">Create an element for the token</a> for which the
4932: <var title="">formatting element</var> was created.</p></li>
4933:
4934: <li><p>Take all of the child nodes of the <var title="">furthest
4935: block</var> and append them to the element created in the last
4936: step.</p></li>
4937:
4938: <li><p>Append that new element to the <var title="">furthest
4939: block</var>.</p></li>
4940:
4941: <li><p>Remove the <var title="">formatting element</var> from the
4942: <a href="#list-of-active-formatting-elements">list of active formatting elements</a>, and insert the
4943: new element into the <a href="#list-of-active-formatting-elements">list of active formatting
4944: elements</a> at the position of the aforementioned
4945: bookmark.</p></li>
4946:
4947: <li><p>Remove the <var title="">formatting element</var> from the
4948: <a href="#stack-of-open-elements">stack of open elements</a>, and insert the new element
4949: into the <a href="#stack-of-open-elements">stack of open elements</a> immediately below
4950: the position of the <var title="">furthest block</var> in that
4951: stack.</p></li>
4952:
4953: <li><p>Jump back to step 1 in this series of steps.</p></li>
4954:
4955: </ol><!--XXX
4956: <div class="example">
4957: <p class="XXX">Need an example.</p>
4958: </div>
4959: --><p class="note">Because of the way this algorithm causes elements to
4960: change parents, it has been dubbed the "adoption agency algorithm"
4961: (in contrast with other possibly algorithms for dealing with
4962: misnested content, which included the "incest algorithm", the
4963: "secret affair algorithm", and the "Heisenberg algorithm").</p>
4964:
4965: </dd>
4966:
4967: <dt>A start tag whose tag name is "button"</dt>
4968: <dd>
4969:
4970: <p>If the <a href="#stack-of-open-elements">stack of open elements</a> <a href="#has-an-element-in-scope" title="has
4971: an element in scope">has a <code>button</code> element in
4972: scope</a>, then this is a <a href="#parse-error">parse error</a>;
4973: act as if an end tag with the tag name "button" had been seen,
4974: then reprocess the token.</p>
4975:
4976: <p>Otherwise:</p>
4977:
4978: <p><a href="#reconstruct-the-active-formatting-elements">Reconstruct the active formatting elements</a>, if
4979: any.</p>
4980:
4981: <p><a href="#insert-an-html-element">Insert an HTML element</a> for the token.</p>
4982:
4983: <p>Insert a marker at the end of the <a href="#list-of-active-formatting-elements">list of active
4984: formatting elements</a>.</p>
4985:
4986: <p>Set the <a href="#frameset-ok-flag">frameset-ok flag</a> to "not ok".</p>
4987:
4988: </dd>
4989:
4990: <dt>A start tag token whose tag name is one of: "applet",
4991: "marquee", "object"</dt>
4992: <dd>
4993:
4994: <p><a href="#reconstruct-the-active-formatting-elements">Reconstruct the active formatting elements</a>, if
4995: any.</p>
4996:
4997: <p><a href="#insert-an-html-element">Insert an HTML element</a> for the token.</p>
4998:
4999: <p>Insert a marker at the end of the <a href="#list-of-active-formatting-elements">list of active
5000: formatting elements</a>.</p>
5001:
5002: <p>Set the <a href="#frameset-ok-flag">frameset-ok flag</a> to "not ok".</p>
5003:
5004: </dd>
5005:
5006: <dt>An end tag token whose tag name is one of: "applet", "button",
5007: "marquee", "object"</dt>
5008: <dd>
5009:
5010: <p>If the <a href="#stack-of-open-elements">stack of open elements</a> does not <a href="#has-an-element-in-scope" title="has an element in scope">have an element in scope</a>
5011: with the same tag name as that of the token, then this is a
5012: <a href="#parse-error">parse error</a>; ignore the token.</p>
5013:
5014: <p>Otherwise, run these steps:</p>
5015:
5016: <ol><li><p><a href="#generate-implied-end-tags">Generate implied end tags</a>.</p></li>
5017:
5018: <li><p>If the <a href="#current-node">current node</a> is not an element with
5019: the same tag name as that of the token, then this is a
5020: <a href="#parse-error">parse error</a>.</p></li>
5021:
5022: <li><p>Pop elements from the <a href="#stack-of-open-elements">stack of open elements</a>
5023: until an element with the same tag name as the token has been
5024: popped from the stack.</p></li>
5025:
5026: <li><a href="#clear-the-list-of-active-formatting-elements-up-to-the-last-marker">Clear the list of active formatting elements up to the
5027: last marker</a>.</li>
5028:
5029: </ol></dd>
5030:
5031: <dt>A start tag whose tag name is "table"</dt>
5032: <dd>
5033:
5034: <p>If the <code>Document</code> is <em>not</em> set to
5035: <a href="dom.html#quirks-mode">quirks mode</a>, and the <a href="#stack-of-open-elements">stack of open
5036: elements</a> <a href="#has-an-element-in-scope" title="has an element in scope">has a
5037: <code>p</code> element in scope</a>, then act as if an end tag
5038: with the tag name "p" had been seen.</p> <!-- i hate myself (this
5039: quirk was basically caused by acid2; if i'd realised we could
5040: change the specs when i wrote acid2, we could have avoided having
5041: any parsing-mode quirks) -Hixie -->
5042:
5043: <p><a href="#insert-an-html-element">Insert an HTML element</a> for the token.</p>
5044:
5045: <p>Set the <a href="#frameset-ok-flag">frameset-ok flag</a> to "not ok".</p>
5046:
5047: <p>Switch the <a href="#insertion-mode">insertion mode</a> to "<a href="#parsing-main-intable" title="insertion mode: in table">in table</a>".</p>
5048:
5049: </dd>
5050:
5051: <dt>A start tag whose tag name is one of: "area", "basefont",
5052: "bgsound", "br", "embed", "img", "input", "keygen", "spacer",
5053: "wbr"</dt>
5054: <dd>
5055:
5056: <p><a href="#reconstruct-the-active-formatting-elements">Reconstruct the active formatting elements</a>, if
5057: any.</p>
5058:
5059: <p><a href="#insert-an-html-element">Insert an HTML element</a> for the token. Immediately
5060: pop the <a href="#current-node">current node</a> off the <a href="#stack-of-open-elements">stack of open
5061: elements</a>.</p>
5062:
5063: <p><a href="#acknowledge-self-closing-flag" title="acknowledge self-closing flag">Acknowledge the
5064: token's <i>self-closing flag</i></a>, if it is set.</p>
5065:
5066: <p>Set the <a href="#frameset-ok-flag">frameset-ok flag</a> to "not ok".</p> <!--
5067: shouldn't really do it for <area>, <basefont>, <bgsound>, and
5068: <spacer> -->
5069:
5070: </dd>
5071:
5072: <dt>A start tag whose tag name is one of: "param", "source"</dt>
5073: <dd>
5074:
5075: <p><a href="#insert-an-html-element">Insert an HTML element</a> for the token. Immediately
5076: pop the <a href="#current-node">current node</a> off the <a href="#stack-of-open-elements">stack of open
5077: elements</a>.</p>
5078:
5079: <p><a href="#acknowledge-self-closing-flag" title="acknowledge self-closing flag">Acknowledge the
5080: token's <i>self-closing flag</i></a>, if it is set.</p>
5081:
5082: </dd>
5083:
5084: <dt>A start tag whose tag name is "hr"</dt>
5085: <dd>
5086:
5087: <p>If the <a href="#stack-of-open-elements">stack of open elements</a> <a href="#has-an-element-in-scope" title="has
5088: an element in scope">has a <code>p</code> element in
5089: scope</a>, then act as if an end tag with the tag name
5090: "p" had been seen.</p>
5091:
5092: <p><a href="#insert-an-html-element">Insert an HTML element</a> for the token. Immediately
5093: pop the <a href="#current-node">current node</a> off the <a href="#stack-of-open-elements">stack of open
5094: elements</a>.</p>
5095:
5096: <p><a href="#acknowledge-self-closing-flag" title="acknowledge self-closing flag">Acknowledge the
5097: token's <i>self-closing flag</i></a>, if it is set.</p>
5098:
5099: <p>Set the <a href="#frameset-ok-flag">frameset-ok flag</a> to "not ok".</p>
5100:
5101: </dd>
5102:
5103: <dt>A start tag whose tag name is "image"</dt>
5104: <dd>
5105: <p><a href="#parse-error">Parse error</a>. Change the token's tag name
5106: to "img" and reprocess it. (Don't ask.)</p> <!-- As of
5107: 2005-12, studies showed that around 0.2% of pages used the
5108: <image> element. -->
5109: </dd>
5110:
5111: <dt id="isindex">A start tag whose tag name is "isindex"</dt>
5112: <dd>
5113:
5114: <p><a href="#parse-error">Parse error</a>.</p>
5115:
5116: <p>If the <a href="#form-element-pointer"><code title="">form</code> element
5117: pointer</a> is not null, then ignore the token.</p>
5118:
5119: <p>Otherwise:</p>
5120:
5121: <p><a href="#acknowledge-self-closing-flag" title="acknowledge self-closing flag">Acknowledge the
5122: token's <i>self-closing flag</i></a>, if it is set.</p> <!--
5123: purely to reduce the number of errors (we don't care if they
5124: included the /, they're not supposed to be including the tag at
5125: all! -->
5126:
5127: <p>Act as if a start tag token with the tag name "form" had been seen.</p>
5128:
5129: <p>If the token has an attribute called "action", set the
5130: <code title="attr-form-action">action</code> attribute on the
5131: resulting <code><a href="forms.html#the-form-element">form</a></code> element to the value of the
5132: "action" attribute of the token.</p>
5133:
5134: <p>Act as if a start tag token with the tag name "hr" had been
5135: seen.</p>
5136:
5137: <p>Act as if a start tag token with the tag name "label" had been
5138: seen.</p>
5139:
5140: <p>Act as if a stream of character tokens had been seen (see below
5141: for what they should say).</p>
5142:
5143: <p>Act as if a start tag token with the tag name "input" had been
5144: seen, with all the attributes from the "isindex" token except
5145: "name", "action", and "prompt". Set the <code title="attr-fe-name"><a href="forms.html#attr-fe-name">name</a></code> attribute of the resulting
5146: <code><a href="forms.html#the-input-element">input</a></code> element to the value "<code title="">isindex</code>".</p>
5147:
5148: <p>Act as if a stream of character tokens had been seen (see
5149: below for what they should say).</p>
5150:
5151: <p>Act as if an end tag token with the tag name "label" had been
5152: seen.</p>
5153:
5154: <p>Act as if a start tag token with the tag name "hr" had been
5155: seen.</p>
5156:
5157: <p>Act as if an end tag token with the tag name "form" had been
5158: seen.</p>
5159:
5160: <p>If the token has an attribute with the name "prompt", then the
5161: first stream of characters must be the same string as given in
5162: that attribute, and the second stream of characters must be
5163: empty. Otherwise, the two streams of character tokens together
5164: should, together with the <code><a href="forms.html#the-input-element">input</a></code> element, express the
5165: equivalent of "This is a searchable index. Insert your search
5166: keywords here: (input field)" in the user's preferred
5167: language.</p>
5168:
5169: </dd>
5170:
5171: <dt>A start tag whose tag name is "textarea"</dt>
5172: <dd>
5173:
5174: <ol><li><p><a href="#insert-an-html-element">Insert an HTML element</a> for the
5175: token.</p></li>
5176:
5177: <li><p>If the next token is a U+000A LINE FEED (LF) character
5178: token, then ignore that token and move on to the next
5179: one. (Newlines at the start of <code><a href="forms.html#the-textarea-element">textarea</a></code> elements are
5180: ignored as an authoring convenience.)</p></li>
5181:
5182: <li><p>Switch the tokenizer's <a href="#content-model-flag">content model flag</a> to
5183: the RCDATA state.</p></li>
5184:
5185: <li><p>Let the <a href="#original-insertion-mode">original insertion mode</a> be the
5186: current <a href="#insertion-mode">insertion mode</a>.</p>
5187:
5188: </li><li><p>Set the <a href="#frameset-ok-flag">frameset-ok flag</a> to "not
5189: ok".</p></li>
5190:
1.91 mike 5191: <li><p>Switch the <a href="#insertion-mode">insertion mode</a> to "<a href="#parsing-main-incdata" title="insertion mode: in RAWTEXT/RCDATA">in
5192: RAWTEXT/RCDATA</a>".</p></li>
1.1 mike 5193:
5194: </ol></dd>
5195:
5196: <dt>A start tag whose tag name is "xmp"</dt>
5197: <dd>
5198:
1.233 mike 5199: <p>If the <a href="#stack-of-open-elements">stack of open elements</a> <a href="#has-an-element-in-scope" title="has
5200: an element in scope">has a <code>p</code> element in
5201: scope</a>, then act as if an end tag with the tag name
5202: "p" had been seen.</p>
5203:
1.1 mike 5204: <p><a href="#reconstruct-the-active-formatting-elements">Reconstruct the active formatting elements</a>, if
5205: any.</p>
5206:
5207: <p>Set the <a href="#frameset-ok-flag">frameset-ok flag</a> to "not ok".</p>
5208:
1.91 mike 5209: <p>Follow the <a href="#generic-raw-text-element-parsing-algorithm">generic raw text element parsing algorithm</a>.</p>
1.1 mike 5210:
5211: </dd>
5212:
5213: <dt>A start tag whose tag name is "iframe"</dt>
5214: <dd>
5215:
5216: <p>Set the <a href="#frameset-ok-flag">frameset-ok flag</a> to "not ok".</p>
5217:
1.91 mike 5218: <p>Follow the <a href="#generic-raw-text-element-parsing-algorithm">generic raw text element parsing algorithm</a>.</p>
1.1 mike 5219:
5220: </dd>
5221:
5222: <dt>A start tag whose tag name is "noembed"</dt>
5223: <dt>A start tag whose tag name is "noscript", if the <a href="#scripting-flag">scripting flag</a> is enabled</dt>
5224: <dd>
5225:
1.91 mike 5226: <p>Follow the <a href="#generic-raw-text-element-parsing-algorithm">generic raw text element parsing algorithm</a>.</p>
1.1 mike 5227:
5228: </dd>
5229:
5230: <dt>A start tag whose tag name is "select"</dt>
5231: <dd>
5232:
5233: <p><a href="#reconstruct-the-active-formatting-elements">Reconstruct the active formatting elements</a>, if
5234: any.</p>
5235:
5236: <p><a href="#insert-an-html-element">Insert an HTML element</a> for the token.</p>
5237:
5238: <p>Set the <a href="#frameset-ok-flag">frameset-ok flag</a> to "not ok".</p>
5239:
5240: <p>If the <a href="#insertion-mode">insertion mode</a> is one of <a href="#parsing-main-intable" title="insertion mode: in table">in table</a>", "<a href="#parsing-main-incaption" title="insertion mode: in caption">in caption</a>", "<a href="#parsing-main-incolgroup" title="insertion mode: in column group">in column group</a>",
5241: "<a href="#parsing-main-intbody" title="insertion mode: in table body">in table
5242: body</a>", "<a href="#parsing-main-intr" title="insertion mode: in row">in
5243: row</a>", or "<a href="#parsing-main-intd" title="insertion mode: in cell">in
5244: cell</a>", then switch the <a href="#insertion-mode">insertion mode</a> to
5245: "<a href="#parsing-main-inselectintable" title="insertion mode: in select in table">in select in
5246: table</a>". Otherwise, switch the <a href="#insertion-mode">insertion mode</a>
5247: to "<a href="#parsing-main-inselect" title="insertion mode: in select">in select</a>".</p>
5248:
5249: </dd>
5250:
5251: <dt>A start tag whose tag name is one of: "optgroup", "option"</dt>
5252: <dd>
5253:
5254: <p>If the <a href="#stack-of-open-elements">stack of open elements</a> <a href="#has-an-element-in-scope" title="has an
5255: element in scope">has an <code>option</code> element in
5256: scope</a>, then act as if an end tag with the tag name "option"
5257: had been seen.</p>
5258:
5259: <p><a href="#reconstruct-the-active-formatting-elements">Reconstruct the active formatting elements</a>, if
5260: any.</p>
5261:
5262: <p><a href="#insert-an-html-element">Insert an HTML element</a> for the token.</p>
5263:
5264: </dd>
5265:
5266: <dt>A start tag whose tag name is one of: "rp", "rt"</dt>
5267: <dd>
5268:
5269: <!-- the parsing rules for ruby really don't match IE much at all,
5270: but in practice the markup used is very simple and so strict
5271: compatibility with IE isn't required. For example, as defined
5272: here we get very, very different behaviour than IE for
5273: pathological cases like:
5274:
5275: <ruby><ol><li><p>a<rt>b
5276: <ruby>a<rt>b<p>c
5277:
5278: But in practice most ruby markup falls into these cases:
5279:
5280: <ruby>a<rt>b</ruby>
5281: <ruby>a<rp>b<rt>c<rp>d</ruby>
5282: <ruby>a<rt>b</rt></ruby>
5283: <ruby>a<rp>b</rp><rt>c</rt><rp>d</rp></ruby>
5284:
5285: -->
5286:
5287: <p>If the <a href="#stack-of-open-elements">stack of open elements</a> <a href="#has-an-element-in-scope" title="has an
5288: element in scope">has a <code>ruby</code> element in scope</a>,
5289: then <a href="#generate-implied-end-tags">generate implied end tags</a>. If the <a href="#current-node">current
5290: node</a> is not then a <code><a href="text-level-semantics.html#the-ruby-element">ruby</a></code> element, this is a
5291: <a href="#parse-error">parse error</a>; pop all the nodes from the <a href="#current-node">current
5292: node</a> up to the node immediately before the bottommost
5293: <code><a href="text-level-semantics.html#the-ruby-element">ruby</a></code> element on the <a href="#stack-of-open-elements">stack of open
5294: elements</a>.</p>
5295:
5296: <p><a href="#insert-an-html-element">Insert an HTML element</a> for the token.</p>
5297:
5298: </dd>
5299:
5300: <dt>An end tag whose tag name is "br"</dt>
5301: <dd>
5302: <p><a href="#parse-error">Parse error</a>. Act as if a start tag token with
5303: the tag name "br" had been seen. Ignore the end tag token.</p>
5304: </dd>
5305:
5306: <dt>A start tag whose tag name is "math"</dt>
5307: <dd>
5308:
5309: <p><a href="#reconstruct-the-active-formatting-elements">Reconstruct the active formatting elements</a>, if
5310: any.</p>
5311:
5312: <p><a href="#adjust-mathml-attributes">Adjust MathML attributes</a> for the token. (This
5313: fixes the case of MathML attributes that are not all
5314: lowercase.)</p>
5315:
5316: <p><a href="#adjust-foreign-attributes">Adjust foreign attributes</a> for the token. (This
5317: fixes the use of namespaced attributes, in particular XLink.)</p>
5318:
5319: <p><a href="#insert-a-foreign-element">Insert a foreign element</a> for the token, in the
5320: <a href="#mathml-namespace">MathML namespace</a>.</p>
5321:
5322: <!-- If we ever change the frameset-ok flag to an insertion mode,
5323: the following change would be implied, except we'd have to do it
5324: even in the face of a self-closed tag:
5325: <p>Set the <span>frameset-ok flag</span> to "not ok".</p>
5326: -->
5327:
5328: <p>If the token has its <i>self-closing flag</i> set, pop the
5329: <a href="#current-node">current node</a> off the <a href="#stack-of-open-elements">stack of open
5330: elements</a> and <a href="#acknowledge-self-closing-flag" title="acknowledge self-closing
5331: flag">acknowledge the token's <i>self-closing flag</i></a>.</p>
5332:
5333: <p>Otherwise, if the <a href="#insertion-mode">insertion mode</a> is not already
5334: "<a href="#parsing-main-inforeign" title="insertion mode: in foreign content">in foreign
5335: content</a>", let the <a href="#secondary-insertion-mode">secondary insertion mode</a> be
5336: the current <a href="#insertion-mode">insertion mode</a>, and then switch the
5337: <a href="#insertion-mode">insertion mode</a> to "<a href="#parsing-main-inforeign" title="insertion mode: in
5338: foreign content">in foreign content</a>".</p>
5339:
5340: </dd>
5341:
5342: <dt>A start tag whose tag name is "svg"</dt>
5343: <dd>
5344:
5345: <p><a href="#reconstruct-the-active-formatting-elements">Reconstruct the active formatting elements</a>, if
5346: any.</p>
5347:
5348: <p><a href="#adjust-svg-attributes">Adjust SVG attributes</a> for the token. (This fixes
5349: the case of SVG attributes that are not all lowercase.)</p>
5350:
5351: <p><a href="#adjust-foreign-attributes">Adjust foreign attributes</a> for the token. (This
5352: fixes the use of namespaced attributes, in particular XLink in
5353: SVG.)</p>
5354:
5355: <p><a href="#insert-a-foreign-element">Insert a foreign element</a> for the token, in the
5356: <a href="#svg-namespace">SVG namespace</a>.</p>
5357:
5358: <!-- If we ever change the frameset-ok flag to an insertion mode,
5359: the following change would be implied, except we'd have to do it
5360: even in the face of a self-closed tag:
5361: <p>Set the <span>frameset-ok flag</span> to "not ok".</p>
5362: -->
5363:
5364: <p>If the token has its <i>self-closing flag</i> set, pop the
5365: <a href="#current-node">current node</a> off the <a href="#stack-of-open-elements">stack of open
5366: elements</a> and <a href="#acknowledge-self-closing-flag" title="acknowledge self-closing
5367: flag">acknowledge the token's <i>self-closing flag</i></a>.</p>
5368:
5369: <p>Otherwise, if the <a href="#insertion-mode">insertion mode</a> is not already
5370: "<a href="#parsing-main-inforeign" title="insertion mode: in foreign content">in foreign
5371: content</a>", let the <a href="#secondary-insertion-mode">secondary insertion mode</a> be
5372: the current <a href="#insertion-mode">insertion mode</a>, and then switch the
5373: <a href="#insertion-mode">insertion mode</a> to "<a href="#parsing-main-inforeign" title="insertion mode: in
5374: foreign content">in foreign content</a>".</p>
5375:
5376: </dd>
5377:
5378: <dt>A start <!--or end--> tag whose tag name is one of: "caption",
5379: "col", "colgroup", "frame", "head", "tbody", "td", "tfoot", "th",
5380: "thead", "tr"</dt>
5381: <!--<dt>An end tag whose tag name is one of: "area", "base",
5382: "basefont", "bgsound", "command", "embed", "hr", "iframe", "image",
5383: "img", "input", "isindex", "keygen", "link", "meta", "noembed",
5384: "noframes", "param", "script", "select", "source", "spacer",
5385: "style", "table", "textarea", "title", "wbr"</dt>-->
5386: <!--<dt>An end tag whose tag name is "noscript", if the
5387: <span>scripting flag</span> is enabled</dt>-->
5388: <dd>
5389: <p><a href="#parse-error">Parse error</a>. Ignore the token.</p>
5390: <!-- end tags are commented out because since they can never end
5391: up on the stack anyway, the default end tag clause will
5392: automatically handle them. we don't want to have text in the spec
5393: that is just an optimisation, as that detracts from the spec
5394: itself -->
5395: </dd>
5396:
5397: <dt>Any other start tag</dt>
5398: <dd>
5399:
5400: <p><a href="#reconstruct-the-active-formatting-elements">Reconstruct the active formatting elements</a>, if
5401: any.</p>
5402:
5403: <p><a href="#insert-an-html-element">Insert an HTML element</a> for the token.</p>
5404:
5405: <p class="note">This element will be a <a href="#phrasing">phrasing</a>
5406: element.</p>
5407:
5408: </dd>
5409:
5410: <dt>Any other end tag</dt>
5411: <dd>
5412:
5413: <p>Run the following steps:</p>
5414:
5415: <ol><li><p>Initialize <var title="">node</var> to be the <a href="#current-node">current
5416: node</a> (the bottommost node of the stack).</p></li>
5417:
5418: <li><p>If <var title="">node</var> has the same tag name as
5419: the end tag token, then:</p>
5420:
5421: <ol><li><p><a href="#generate-implied-end-tags">Generate implied end tags</a>.</p></li>
5422:
5423: <li><p>If the tag name of the end tag token does not match
5424: the tag name of the <a href="#current-node">current node</a>, this is a
5425: <a href="#parse-error">parse error</a>.</p></li>
5426:
5427: <li><p>Pop all the nodes from the <a href="#current-node">current node</a> up
5428: to <var title="">node</var>, including <var title="">node</var>, then stop these steps.</p></li>
5429:
5430: </ol></li>
5431:
5432: <li><p>Otherwise, if <var title="">node</var> is in neither the
5433: <a href="#formatting">formatting</a> category nor the <a href="#phrasing">phrasing</a>
5434: category, then this is a <a href="#parse-error">parse error</a>; ignore the
5435: token, and abort these steps.</p></li>
5436:
5437: <li><p>Set <var title="">node</var> to the previous entry in the
5438: <a href="#stack-of-open-elements">stack of open elements</a>.</p></li>
5439:
5440: <li><p>Return to step 2.</p></li>
5441:
5442: </ol></dd>
5443:
1.91 mike 5444: </dl><h5 id="parsing-main-incdata"><span class="secno">9.2.5.11 </span>The "<dfn title="insertion mode: in RAWTEXT/RCDATA">in RAWTEXT/RCDATA</dfn>" insertion mode</h5>
1.1 mike 5445:
5446: <p>When the <a href="#insertion-mode">insertion mode</a> is "<a href="#parsing-main-incdata" title="insertion
1.91 mike 5447: mode: in RAWTEXT/RCDATA">in RAWTEXT/RCDATA</a>", tokens must be
1.1 mike 5448: handled as follows:</p>
5449:
5450: <dl class="switch"><dt>A character token</dt>
5451: <dd>
5452:
5453: <p><a href="#insert-a-character" title="insert a character">Insert the token's
5454: character</a> into the <a href="#current-node">current node</a>.</p>
5455:
5456: </dd>
5457:
5458: <dt>An end-of-file token</dt>
5459: <dd>
5460:
5461: <!-- can't be the fragment case -->
5462: <p><a href="#parse-error">Parse error</a>.</p>
5463:
5464: <p>If the <a href="#current-node">current node</a> is a <code><a href="semantics.html#script">script</a></code>
5465: element, mark the <code><a href="semantics.html#script">script</a></code> element as <a href="semantics.html#already-executed">"already
5466: executed"</a>.</p>
5467:
5468: <p>Pop the <a href="#current-node">current node</a> off the <a href="#stack-of-open-elements">stack of open
5469: elements</a>.</p>
5470:
5471: <p>Switch the <a href="#insertion-mode">insertion mode</a> to the <a href="#original-insertion-mode">original
5472: insertion mode</a> and reprocess the current token.</p>
5473:
5474: </dd>
5475:
5476: <dt>An end tag whose tag name is "script"</dt>
5477: <dd>
5478:
5479: <p>Let <var title="">script</var> be the <a href="#current-node">current node</a>
5480: (which will be a <code><a href="semantics.html#script">script</a></code> element).</p>
5481:
5482: <p>Pop the <a href="#current-node">current node</a> off the <a href="#stack-of-open-elements">stack of open
5483: elements</a>.</p>
5484:
5485: <p>Switch the <a href="#insertion-mode">insertion mode</a> to the <a href="#original-insertion-mode">original
5486: insertion mode</a>.</p>
5487:
5488: <p>Let the <var title="">old insertion point</var> have the
5489: same value as the current <a href="#insertion-point">insertion point</a>. Let
5490: the <a href="#insertion-point">insertion point</a> be just before the <a href="#next-input-character">next
5491: input character</a>.</p>
5492:
5493: <p>Increment the parser's <a href="#script-nesting-level">script nesting level</a> by
5494: one.</p>
5495:
5496: <p><a href="semantics.html#running-a-script" title="running a script">Run</a> the <var title="">script</var>. This might cause some script to execute,
1.159 mike 5497: which might cause <a href="embedded-content-0.html#dom-document-write" title="dom-document-write">new characters
1.1 mike 5498: to be inserted into the tokenizer</a>, and might cause the
5499: tokenizer to output more tokens, resulting in a <a href="#nestedParsing">reentrant invocation of the parser</a>.</p>
5500:
5501: <p>Decrement the parser's <a href="#script-nesting-level">script nesting level</a> by
5502: one. If the parser's <a href="#script-nesting-level">script nesting level</a> is zero,
5503: then set the <a href="#parser-pause-flag">parser pause flag</a> to false.</p>
5504:
5505: <p>Let the <a href="#insertion-point">insertion point</a> have the value of the <var title="">old insertion point</var>. (In other words, restore the
5506: <a href="#insertion-point">insertion point</a> to its previous value. This value
5507: might be the "undefined" value.)</p>
5508:
5509: <p id="scriptTagParserResumes">At this stage, if there is a
5510: <a href="semantics.html#pending-external-script">pending external script</a>, then:</p>
5511:
5512: <dl class="switch"><dt>If the <a href="#script-nesting-level">script nesting level</a> is not zero:</dt>
5513:
5514: <dd>
5515:
5516: <p>Set the <a href="#parser-pause-flag">parser pause flag</a> to true, and abort the
5517: processing of any nested invocations of the tokenizer, yielding
5518: control back to the caller. (Tokenization will resume when the
5519: caller returns to the "outer" tree construction stage.)</p>
5520:
5521: <p class="note">The tree construction stage of this particular
5522: parser is <a href="#nestedParsing">being called reentrantly</a>,
1.159 mike 5523: say from a call to <code title="dom-document-write"><a href="embedded-content-0.html#dom-document-write">document.write()</a></code>.</p>
1.1 mike 5524:
5525: </dd>
5526:
5527:
5528: <dt>Otherwise:</dt>
5529:
5530: <dd>
5531:
5532: <p>Follow these steps:</p>
5533:
5534: <ol><li><p>Let <var title="">the script</var> be the <a href="semantics.html#pending-external-script">pending
5535: external script</a>. There is no longer a <a href="semantics.html#pending-external-script">pending
5536: external script</a>.</p></li>
5537:
5538: <li><p><a href="browsers.html#pause">Pause</a> until the script has <a href="semantics.html#completed-loading">completed
5539: loading</a>.</p></li>
5540:
5541: <li><p>Let the <a href="#insertion-point">insertion point</a> be just before the
5542: <a href="#next-input-character">next input character</a>.</p></li>
5543:
5544: <li><p>Increment the parser's <a href="#script-nesting-level">script nesting level</a>
5545: by one (it should be zero before this step, so this sets it to
5546: one).</p></li>
5547:
5548: <li><p><a href="semantics.html#executing-a-script-block" title="executing a script block">Execute the
5549: script</a>.</p></li>
5550:
5551: <li><p>Decrement the parser's <a href="#script-nesting-level">script nesting level</a>
5552: by one. If the parser's <a href="#script-nesting-level">script nesting level</a> is
5553: zero (which it always should be at this point), then set the
5554: <a href="#parser-pause-flag">parser pause flag</a> to false.</p>
5555:
5556: </li><li><p>Let the <a href="#insertion-point">insertion point</a> be undefined
5557: again.</p></li>
5558:
5559: <li><p>If there is once again a <a href="semantics.html#pending-external-script">pending external
5560: script</a>, then repeat these steps from step 1.</p></li>
5561:
5562: </ol></dd>
5563:
5564: </dl></dd>
5565:
5566: <dt>Any other end tag</dt>
5567: <dd>
5568:
5569: <p>Pop the <a href="#current-node">current node</a> off the <a href="#stack-of-open-elements">stack of open
5570: elements</a>.</p>
5571:
5572: <p>Switch the <a href="#insertion-mode">insertion mode</a> to the <a href="#original-insertion-mode">original
5573: insertion mode</a>.</p>
5574:
5575: </dd>
5576:
5577: </dl><h5 id="parsing-main-intable"><span class="secno">9.2.5.12 </span>The "<dfn title="insertion mode: in table">in table</dfn>" insertion mode</h5>
5578:
5579: <p>When the <a href="#insertion-mode">insertion mode</a> is "<a href="#parsing-main-intable" title="insertion
5580: mode: in table">in table</a>", tokens must be handled as follows:</p>
5581:
5582: <dl class="switch"><dt>A character token</dt>
5583: <dd>
5584:
5585: <p>Let the <dfn id="pending-table-character-tokens"><var>pending table character tokens</var></dfn>
5586: be an empty list of tokens.</p>
5587:
5588: <p>Let the <a href="#original-insertion-mode">original insertion mode</a> be the current
5589: <a href="#insertion-mode">insertion mode</a>.</p>
5590:
5591: <p>Switch the <a href="#insertion-mode">insertion mode</a> to "<a href="#parsing-main-intabletext" title="insertion mode: in table text">in table text</a>" and
5592: reprocess the token.</p>
5593:
5594: </dd>
5595:
5596: <dt>A comment token</dt>
5597: <dd>
5598: <p>Append a <code>Comment</code> node to the <a href="#current-node">current
5599: node</a> with the <code title="">data</code> attribute set to
5600: the data given in the comment token.</p>
5601: </dd>
5602:
5603: <dt>A DOCTYPE token</dt>
5604: <dd>
5605: <p><a href="#parse-error">Parse error</a>. Ignore the token.</p>
5606: </dd>
5607:
5608: <dt>A start tag whose tag name is "caption"</dt>
5609: <dd>
5610:
5611: <p><a href="#clear-the-stack-back-to-a-table-context">Clear the stack back to a table context</a>. (See
5612: below.)</p>
5613:
5614: <p>Insert a marker at the end of the <a href="#list-of-active-formatting-elements">list of active
5615: formatting elements</a>.</p>
5616:
5617: <p><a href="#insert-an-html-element">Insert an HTML element</a> for the token, then
5618: switch the <a href="#insertion-mode">insertion mode</a> to "<a href="#parsing-main-incaption" title="insertion mode: in caption">in caption</a>".</p>
5619:
5620: </dd>
5621:
5622: <dt>A start tag whose tag name is "colgroup"</dt>
5623: <dd>
5624:
5625: <p><a href="#clear-the-stack-back-to-a-table-context">Clear the stack back to a table context</a>. (See
5626: below.)</p>
5627:
5628: <p><a href="#insert-an-html-element">Insert an HTML element</a> for the token, then
5629: switch the <a href="#insertion-mode">insertion mode</a> to "<a href="#parsing-main-incolgroup" title="insertion mode: in column group">in column
5630: group</a>".</p>
5631:
5632: </dd>
5633:
5634: <dt>A start tag whose tag name is "col"</dt>
5635: <dd>
5636: <p>Act as if a start tag token with the tag name "colgroup"
5637: had been seen, then reprocess the current token.</p>
5638: </dd>
5639:
5640: <dt>A start tag whose tag name is one of: "tbody", "tfoot", "thead"</dt>
5641: <dd>
5642:
5643: <p><a href="#clear-the-stack-back-to-a-table-context">Clear the stack back to a table context</a>. (See
5644: below.)</p>
5645:
5646: <p><a href="#insert-an-html-element">Insert an HTML element</a> for the token, then
5647: switch the <a href="#insertion-mode">insertion mode</a> to "<a href="#parsing-main-intbody" title="insertion mode: in table body">in table
5648: body</a>".</p>
5649:
5650: </dd>
5651:
5652: <dt>A start tag whose tag name is one of: "td", "th", "tr"</dt>
5653: <dd>
5654: <p>Act as if a start tag token with the tag name "tbody" had
5655: been seen, then reprocess the current token.</p>
5656: </dd>
5657:
5658: <dt>A start tag whose tag name is "table"</dt>
5659: <dd>
5660:
5661: <p><a href="#parse-error">Parse error</a>. Act as if an end tag token with
5662: the tag name "table" had been seen, then, if that token wasn't
5663: ignored, reprocess the current token.</p>
5664:
5665: <p class="note">The fake end tag token here can only be
5666: ignored in the <a href="#fragment-case">fragment case</a>.</p>
5667:
5668: </dd>
5669:
5670: <dt>An end tag whose tag name is "table"</dt>
5671: <dd>
5672:
5673: <p>If the <a href="#stack-of-open-elements">stack of open elements</a> does not <a href="#has-an-element-in-table-scope" title="has an element in table scope">have an element in table
5674: scope</a> with the same tag name as the token, this is a
5675: <a href="#parse-error">parse error</a>. Ignore the token. (<a href="#fragment-case">fragment
5676: case</a>)</p>
5677:
5678: <p>Otherwise:</p>
5679:
5680: <p>Pop elements from this stack until a <code><a href="tabular-data.html#the-table-element">table</a></code>
5681: element has been popped from the stack.</p>
5682:
5683: <p><a href="#reset-the-insertion-mode-appropriately">Reset the insertion mode appropriately</a>.</p>
5684:
5685: </dd>
5686:
5687: <dt>An end tag whose tag name is one of: "body", "caption",
5688: "col", "colgroup", "html", "tbody", "td", "tfoot", "th",
5689: "thead", "tr"</dt>
5690: <dd>
5691: <p><a href="#parse-error">Parse error</a>. Ignore the token.</p>
5692: </dd>
5693:
5694: <dt>A start tag whose tag name is one of: "style", "script"</dt>
5695: <dd>
5696:
5697: <p>Process the token <a href="#using-the-rules-for">using the rules for</a> the "<a href="#parsing-main-inhead" title="insertion mode: in head">in head</a>" <a href="#insertion-mode">insertion
5698: mode</a>.</p>
5699:
5700: </dd>
5701:
5702: <dt>A start tag whose tag name is "input"</dt>
5703: <dd>
5704:
5705: <p>If the token does not have an attribute with the name "type",
5706: or if it does, but that attribute's value is not an <a href="infrastructure.html#ascii-case-insensitive">ASCII
5707: case-insensitive</a> match for the string "<code title="">hidden</code>", then: act as described in the "anything
5708: else" entry below.</p>
5709:
5710: <p>Otherwise:</p>
5711:
5712: <p><a href="#parse-error">Parse error</a>.</p>
5713:
5714: <p><a href="#insert-an-html-element">Insert an HTML element</a> for the token.</p>
5715:
5716: <p>Pop that <code><a href="forms.html#the-input-element">input</a></code> element off the <a href="#stack-of-open-elements">stack of
5717: open elements</a>.</p>
5718:
5719: </dd>
5720:
5721: <dt>An end-of-file token</dt>
5722: <dd>
5723:
5724: <p>If the <a href="#current-node">current node</a> is not the root
1.159 mike 5725: <code><a href="semantics.html#the-html-element-0">html</a></code> element, then this is a <a href="#parse-error">parse
1.1 mike 5726: error</a>.</p>
5727:
5728: <p class="note">It can only be the <a href="#current-node">current node</a> in
5729: the <a href="#fragment-case">fragment case</a>.</p>
5730:
5731: <p><a href="#stop-parsing">Stop parsing</a>.</p>
5732:
5733: </dd>
5734:
5735: <dt>Anything else</dt>
5736: <dd>
5737:
5738: <p><a href="#parse-error">Parse error</a>. Process the token <a href="#using-the-rules-for">using the
5739: rules for</a> the "<a href="#parsing-main-inbody" title="insertion mode: in body">in
5740: body</a>" <a href="#insertion-mode">insertion mode</a>, except that if the
5741: <a href="#current-node">current node</a> is a <code><a href="tabular-data.html#the-table-element">table</a></code>,
5742: <code><a href="tabular-data.html#the-tbody-element">tbody</a></code>, <code><a href="tabular-data.html#the-tfoot-element">tfoot</a></code>, <code><a href="tabular-data.html#the-thead-element">thead</a></code>, or
5743: <code><a href="tabular-data.html#the-tr-element">tr</a></code> element, then, whenever a node would be inserted
5744: into the <a href="#current-node">current node</a>, it must instead be <a href="#foster-parent" title="foster parent">foster parented</a>.</p>
5745:
5746: </dd>
5747:
5748: </dl><p>When the steps above require the UA to <dfn id="clear-the-stack-back-to-a-table-context">clear the stack
5749: back to a table context</dfn>, it means that the UA must, while
5750: the <a href="#current-node">current node</a> is not a <code><a href="tabular-data.html#the-table-element">table</a></code>
1.159 mike 5751: element or an <code><a href="semantics.html#the-html-element-0">html</a></code> element, pop elements from the
1.1 mike 5752: <a href="#stack-of-open-elements">stack of open elements</a>.</p>
5753:
5754: <p class="note">The <a href="#current-node">current node</a> being an
1.159 mike 5755: <code><a href="semantics.html#the-html-element-0">html</a></code> element after this process is a <a href="#fragment-case">fragment
1.1 mike 5756: case</a>.</p>
5757:
5758:
5759:
5760: <h5 id="parsing-main-intabletext"><span class="secno">9.2.5.13 </span>The "<dfn title="insertion mode: in table text">in table text</dfn>" insertion mode</h5>
5761:
5762: <p>When the <a href="#insertion-mode">insertion mode</a> is "<a href="#parsing-main-intabletext" title="insertion
5763: mode: in table text">in table text</a>", tokens must be handled
5764: as follows:</p>
5765:
5766: <dl class="switch"><dt>A character token</dt>
5767: <dd>
5768:
5769: <p>Append the character token to the <var><a href="#pending-table-character-tokens">pending table character
5770: tokens</a></var> list.</p>
5771:
5772: </dd>
5773:
5774:
5775: <dt>Anything else</dt>
5776: <dd>
5777:
5778: <p>If any of the tokens in the <var><a href="#pending-table-character-tokens">pending table character
5779: tokens</a></var> list are character tokens that are not one of U+0009
5780: CHARACTER TABULATION, U+000A LINE FEED (LF), U+000C FORM FEED
5781: (FF), <!--U+000D CARRIAGE RETURN (CR),--> or U+0020 SPACE, then
5782: reprocess those character tokens using the rules given in the
5783: "anything else" entry in the <a href="#parsing-main-intable" title="insertion mode: in
5784: table">in table</a>" insertion mode.</p>
5785:
5786: <p>Otherwise, <a href="#insert-a-character" title="insert a character">insert the
5787: characters</a> given by the <var><a href="#pending-table-character-tokens">pending table character
5788: tokens</a></var> list into the <a href="#current-node">current node</a>.</p>
5789:
5790: <p>Switch the <a href="#insertion-mode">insertion mode</a> to the <a href="#original-insertion-mode">original
5791: insertion mode</a> and reprocess the token.</p>
5792:
5793: </dd>
5794:
5795: </dl><h5 id="parsing-main-incaption"><span class="secno">9.2.5.14 </span>The "<dfn title="insertion mode: in caption">in caption</dfn>" insertion mode</h5>
5796:
5797: <p>When the <a href="#insertion-mode">insertion mode</a> is "<a href="#parsing-main-incaption" title="insertion
5798: mode: in caption">in caption</a>", tokens must be handled as follows:</p>
5799:
5800: <dl class="switch"><dt>An end tag whose tag name is "caption"</dt>
5801: <dd>
5802:
5803: <p>If the <a href="#stack-of-open-elements">stack of open elements</a> does not <a href="#has-an-element-in-table-scope" title="has an element in table scope">have an element in table
5804: scope</a> with the same tag name as the token, this is a
5805: <a href="#parse-error">parse error</a>. Ignore the token. (<a href="#fragment-case">fragment
5806: case</a>)</p>
5807:
5808: <p>Otherwise:</p>
5809:
5810: <p><a href="#generate-implied-end-tags">Generate implied end tags</a>.</p>
5811:
5812: <p>Now, if the <a href="#current-node">current node</a> is not a
5813: <code><a href="tabular-data.html#the-caption-element">caption</a></code> element, then this is a <a href="#parse-error">parse
5814: error</a>.</p>
5815:
5816: <p>Pop elements from this stack until a <code><a href="tabular-data.html#the-caption-element">caption</a></code>
5817: element has been popped from the stack.</p>
5818:
5819: <p><a href="#clear-the-list-of-active-formatting-elements-up-to-the-last-marker">Clear the list of active formatting elements up to
5820: the last marker</a>.</p>
5821:
5822: <p>Switch the <a href="#insertion-mode">insertion mode</a> to "<a href="#parsing-main-intable" title="insertion mode: in table">in table</a>".</p>
5823:
5824: </dd>
5825:
5826: <dt>A start tag whose tag name is one of: "caption", "col",
5827: "colgroup", "tbody", "td", "tfoot", "th", "thead", "tr"</dt>
5828: <dt>An end tag whose tag name is "table"</dt>
5829: <dd>
5830:
5831: <p><a href="#parse-error">Parse error</a>. Act as if an end tag with the tag
5832: name "caption" had been seen, then, if that token wasn't
5833: ignored, reprocess the current token.</p>
5834:
5835: <p class="note">The fake end tag token here can only be
5836: ignored in the <a href="#fragment-case">fragment case</a>.</p>
5837:
5838: </dd>
5839:
5840: <dt>An end tag whose tag name is one of: "body", "col",
5841: "colgroup", "html", "tbody", "td", "tfoot", "th", "thead",
5842: "tr"</dt>
5843: <dd>
5844: <p><a href="#parse-error">Parse error</a>. Ignore the token.</p>
5845: </dd>
5846:
5847: <dt>Anything else</dt>
5848: <dd>
5849: <p>Process the token <a href="#using-the-rules-for">using the rules for</a> the "<a href="#parsing-main-inbody" title="insertion mode: in body">in body</a>" <a href="#insertion-mode">insertion
5850: mode</a>.</p>
5851: </dd>
5852:
5853: </dl><h5 id="parsing-main-incolgroup"><span class="secno">9.2.5.15 </span>The "<dfn title="insertion mode: in column group">in column group</dfn>" insertion mode</h5>
5854:
5855: <p>When the <a href="#insertion-mode">insertion mode</a> is "<a href="#parsing-main-incolgroup" title="insertion
5856: mode: in column group">in column group</a>", tokens must be handled as follows:</p>
5857:
5858: <dl class="switch"><dt>A character token that is one of U+0009 CHARACTER
5859: TABULATION, U+000A LINE FEED (LF), U+000C FORM FEED (FF),
5860: <!--U+000D CARRIAGE RETURN (CR),--> or U+0020 SPACE</dt>
5861: <dd>
5862: <p><a href="#insert-a-character" title="insert a character">Insert the character</a> into
5863: the <a href="#current-node">current node</a>.</p>
5864: </dd>
5865:
5866: <dt>A comment token</dt>
5867: <dd>
5868: <p>Append a <code>Comment</code> node to the <a href="#current-node">current
5869: node</a> with the <code title="">data</code> attribute set to
5870: the data given in the comment token.</p>
5871: </dd>
5872:
5873: <dt>A DOCTYPE token</dt>
5874: <dd>
5875: <p><a href="#parse-error">Parse error</a>. Ignore the token.</p>
5876: </dd>
5877:
5878: <dt>A start tag whose tag name is "html"</dt>
5879: <dd>
5880: <p>Process the token <a href="#using-the-rules-for">using the rules for</a> the "<a href="#parsing-main-inbody" title="insertion mode: in body">in body</a>" <a href="#insertion-mode">insertion
5881: mode</a>.</p>
5882: </dd>
5883:
5884: <dt>A start tag whose tag name is "col"</dt>
5885: <dd>
5886:
5887: <p><a href="#insert-an-html-element">Insert an HTML element</a> for the token. Immediately
5888: pop the <a href="#current-node">current node</a> off the <a href="#stack-of-open-elements">stack of open
5889: elements</a>.</p>
5890:
5891: <p><a href="#acknowledge-self-closing-flag" title="acknowledge self-closing flag">Acknowledge the
5892: token's <i>self-closing flag</i></a>, if it is set.</p>
5893:
5894: </dd>
5895:
5896: <dt>An end tag whose tag name is "colgroup"</dt>
5897: <dd>
5898:
5899: <p>If the <a href="#current-node">current node</a> is the root
1.159 mike 5900: <code><a href="semantics.html#the-html-element-0">html</a></code> element, then this is a <a href="#parse-error">parse
1.1 mike 5901: error</a>; ignore the token. (<a href="#fragment-case">fragment
5902: case</a>)</p>
5903:
5904: <p>Otherwise, pop the <a href="#current-node">current node</a> (which will be
5905: a <code><a href="tabular-data.html#the-colgroup-element">colgroup</a></code> element) from the <a href="#stack-of-open-elements">stack of open
5906: elements</a>. Switch the <a href="#insertion-mode">insertion mode</a> to
5907: "<a href="#parsing-main-intable" title="insertion mode: in table">in table</a>".</p>
5908:
5909: </dd>
5910:
5911: <dt>An end tag whose tag name is "col"</dt>
5912: <dd>
5913: <p><a href="#parse-error">Parse error</a>. Ignore the token.</p>
5914: </dd>
5915:
5916: <dt>An end-of-file token</dt>
5917: <dd>
5918:
1.159 mike 5919: <p>If the <a href="#current-node">current node</a> is the root <code><a href="semantics.html#the-html-element-0">html</a></code>
1.1 mike 5920: element, then <a href="#stop-parsing">stop parsing</a>. (<a href="#fragment-case">fragment
5921: case</a>)</p>
5922:
5923: <p>Otherwise, act as described in the "anything else" entry
5924: below.</p>
5925:
5926: </dd>
5927:
5928: <dt>Anything else</dt>
5929: <dd>
5930:
5931: <p>Act as if an end tag with the tag name "colgroup" had been
5932: seen, and then, if that token wasn't ignored, reprocess the
5933: current token.</p>
5934:
5935: <p class="note">The fake end tag token here can only be
5936: ignored in the <a href="#fragment-case">fragment case</a>.</p>
5937:
5938: </dd>
5939:
5940: </dl><h5 id="parsing-main-intbody"><span class="secno">9.2.5.16 </span>The "<dfn title="insertion mode: in table body">in table body</dfn>" insertion mode</h5>
5941:
5942: <p>When the <a href="#insertion-mode">insertion mode</a> is "<a href="#parsing-main-intbody" title="insertion
5943: mode: in table body">in table body</a>", tokens must be handled as follows:</p>
5944:
5945: <dl class="switch"><dt>A start tag whose tag name is "tr"</dt>
5946: <dd>
5947:
5948: <p><a href="#clear-the-stack-back-to-a-table-body-context">Clear the stack back to a table body
5949: context</a>. (See below.)</p>
5950:
5951: <p><a href="#insert-an-html-element">Insert an HTML element</a> for the token, then switch
5952: the <a href="#insertion-mode">insertion mode</a> to "<a href="#parsing-main-intr" title="insertion mode:
5953: in row">in row</a>".</p>
5954:
5955: </dd>
5956:
5957: <dt>A start tag whose tag name is one of: "th", "td"</dt>
5958: <dd>
5959: <p><a href="#parse-error">Parse error</a>. Act as if a start tag with
5960: the tag name "tr" had been seen, then reprocess the current
5961: token.</p>
5962: </dd>
5963:
5964: <dt>An end tag whose tag name is one of: "tbody", "tfoot",
5965: "thead"</dt>
5966: <dd>
5967:
5968: <p>If the <a href="#stack-of-open-elements">stack of open elements</a> does not <a href="#has-an-element-in-table-scope" title="has an element in table scope">have an element in table
5969: scope</a> with the same tag name as the token, this is a
5970: <a href="#parse-error">parse error</a>. Ignore the token.</p>
5971:
5972: <p>Otherwise:</p>
5973:
5974: <p><a href="#clear-the-stack-back-to-a-table-body-context">Clear the stack back to a table body
5975: context</a>. (See below.)</p>
5976:
5977: <p>Pop the <a href="#current-node">current node</a> from the <a href="#stack-of-open-elements">stack of
5978: open elements</a>. Switch the <a href="#insertion-mode">insertion mode</a>
5979: to "<a href="#parsing-main-intable" title="insertion mode: in table">in table</a>".</p>
5980:
5981: </dd>
5982:
5983: <dt>A start tag whose tag name is one of: "caption", "col",
5984: "colgroup", "tbody", "tfoot", "thead"</dt>
5985: <dt>An end tag whose tag name is "table"</dt>
5986: <dd>
5987:
5988: <p>If the <a href="#stack-of-open-elements">stack of open elements</a> does not <a href="#has-an-element-in-table-scope" title="has an element in table scope">have a
5989: <code>tbody</code>, <code>thead</code>, or <code>tfoot</code>
5990: element in table scope</a>, this is a <a href="#parse-error">parse
5991: error</a>. Ignore the token. (<a href="#fragment-case">fragment
5992: case</a>)</p>
5993:
5994: <p>Otherwise:</p>
5995:
5996: <p><a href="#clear-the-stack-back-to-a-table-body-context">Clear the stack back to a table body
5997: context</a>. (See below.)</p>
5998:
5999: <p>Act as if an end tag with the same tag name as the
6000: <a href="#current-node">current node</a> ("tbody", "tfoot", or "thead") had
6001: been seen, then reprocess the current token.</p>
6002:
6003: </dd>
6004:
6005: <dt>An end tag whose tag name is one of: "body", "caption",
6006: "col", "colgroup", "html", "td", "th", "tr"</dt>
6007: <dd>
6008: <p><a href="#parse-error">Parse error</a>. Ignore the token.</p>
6009: </dd>
6010:
6011: <dt>Anything else</dt>
6012: <dd>
6013: <p>Process the token <a href="#using-the-rules-for">using the rules for</a> the "<a href="#parsing-main-intable" title="insertion mode: in table">in table</a>" <a href="#insertion-mode">insertion
6014: mode</a>.</p>
6015: </dd>
6016:
6017: </dl><p>When the steps above require the UA to <dfn id="clear-the-stack-back-to-a-table-body-context">clear the stack
6018: back to a table body context</dfn>, it means that the UA must,
6019: while the <a href="#current-node">current node</a> is not a <code><a href="tabular-data.html#the-tbody-element">tbody</a></code>,
1.159 mike 6020: <code><a href="tabular-data.html#the-tfoot-element">tfoot</a></code>, <code><a href="tabular-data.html#the-thead-element">thead</a></code>, or <code><a href="semantics.html#the-html-element-0">html</a></code>
1.1 mike 6021: element, pop elements from the <a href="#stack-of-open-elements">stack of open
6022: elements</a>.</p>
6023:
6024: <p class="note">The <a href="#current-node">current node</a> being an
1.159 mike 6025: <code><a href="semantics.html#the-html-element-0">html</a></code> element after this process is a <a href="#fragment-case">fragment
1.1 mike 6026: case</a>.</p>
6027:
6028:
6029: <h5 id="parsing-main-intr"><span class="secno">9.2.5.17 </span>The "<dfn title="insertion mode: in row">in row</dfn>" insertion mode</h5>
6030:
6031: <p>When the <a href="#insertion-mode">insertion mode</a> is "<a href="#parsing-main-intr" title="insertion
6032: mode: in row">in row</a>", tokens must be handled as follows:</p>
6033:
6034: <dl class="switch"><dt>A start tag whose tag name is one of: "th", "td"</dt>
6035: <dd>
6036:
6037: <p><a href="#clear-the-stack-back-to-a-table-row-context">Clear the stack back to a table row
6038: context</a>. (See below.)</p>
6039:
6040: <p><a href="#insert-an-html-element">Insert an HTML element</a> for the token, then switch
6041: the <a href="#insertion-mode">insertion mode</a> to "<a href="#parsing-main-intd" title="insertion mode:
6042: in cell">in cell</a>".</p>
6043:
6044: <p>Insert a marker at the end of the <a href="#list-of-active-formatting-elements">list of active
6045: formatting elements</a>.</p>
6046:
6047: </dd>
6048:
6049: <dt>An end tag whose tag name is "tr"</dt>
6050: <dd>
6051:
6052: <p>If the <a href="#stack-of-open-elements">stack of open elements</a> does not <a href="#has-an-element-in-table-scope" title="has an element in table scope">have an element in table
6053: scope</a> with the same tag name as the token, this is a
6054: <a href="#parse-error">parse error</a>. Ignore the token. (<a href="#fragment-case">fragment
6055: case</a>)</p>
6056:
6057: <p>Otherwise:</p>
6058:
6059: <p><a href="#clear-the-stack-back-to-a-table-row-context">Clear the stack back to a table row
6060: context</a>. (See below.)</p>
6061:
6062: <p>Pop the <a href="#current-node">current node</a> (which will be a
6063: <code><a href="tabular-data.html#the-tr-element">tr</a></code> element) from the <a href="#stack-of-open-elements">stack of open
6064: elements</a>. Switch the <a href="#insertion-mode">insertion mode</a> to
6065: "<a href="#parsing-main-intbody" title="insertion mode: in table body">in table
6066: body</a>".</p>
6067:
6068: </dd>
6069:
6070: <dt>A start tag whose tag name is one of: "caption", "col",
6071: "colgroup", "tbody", "tfoot", "thead", "tr"</dt>
6072: <dt>An end tag whose tag name is "table"</dt>
6073: <dd>
6074:
6075: <p>Act as if an end tag with the tag name "tr" had been seen,
6076: then, if that token wasn't ignored, reprocess the current
6077: token.</p>
6078:
6079: <p class="note">The fake end tag token here can only be
6080: ignored in the <a href="#fragment-case">fragment case</a>.</p>
6081:
6082: </dd>
6083:
6084: <dt>An end tag whose tag name is one of: "tbody", "tfoot",
6085: "thead"</dt>
6086: <dd>
6087:
6088: <p>If the <a href="#stack-of-open-elements">stack of open elements</a> does not <a href="#has-an-element-in-table-scope" title="has an element in table scope">have an element in table
6089: scope</a> with the same tag name as the token, this is a
6090: <a href="#parse-error">parse error</a>. Ignore the token.</p>
6091:
6092: <p>Otherwise, act as if an end tag with the tag name "tr" had
6093: been seen, then reprocess the current token.</p>
6094:
6095: </dd>
6096:
6097: <dt>An end tag whose tag name is one of: "body", "caption",
6098: "col", "colgroup", "html", "td", "th"</dt>
6099: <dd>
6100: <p><a href="#parse-error">Parse error</a>. Ignore the token.</p>
6101: </dd>
6102:
6103: <dt>Anything else</dt>
6104: <dd>
6105: <p>Process the token <a href="#using-the-rules-for">using the rules for</a> the "<a href="#parsing-main-intable" title="insertion mode: in table">in table</a>" <a href="#insertion-mode">insertion
6106: mode</a>.</p>
6107: </dd>
6108:
6109: </dl><p>When the steps above require the UA to <dfn id="clear-the-stack-back-to-a-table-row-context">clear the stack
6110: back to a table row context</dfn>, it means that the UA must,
6111: while the <a href="#current-node">current node</a> is not a <code><a href="tabular-data.html#the-tr-element">tr</a></code>
1.159 mike 6112: element or an <code><a href="semantics.html#the-html-element-0">html</a></code> element, pop elements from the
1.1 mike 6113: <a href="#stack-of-open-elements">stack of open elements</a>.</p>
6114:
6115: <p class="note">The <a href="#current-node">current node</a> being an
1.159 mike 6116: <code><a href="semantics.html#the-html-element-0">html</a></code> element after this process is a <a href="#fragment-case">fragment
1.1 mike 6117: case</a>.</p>
6118:
6119:
6120: <h5 id="parsing-main-intd"><span class="secno">9.2.5.18 </span>The "<dfn title="insertion mode: in cell">in cell</dfn>" insertion mode</h5>
6121:
6122: <p>When the <a href="#insertion-mode">insertion mode</a> is "<a href="#parsing-main-intd" title="insertion
6123: mode: in cell">in cell</a>", tokens must be handled as follows:</p>
6124:
6125: <dl class="switch"><dt>An end tag whose tag name is one of: "td", "th"</dt>
6126: <dd>
6127:
6128: <p>If the <a href="#stack-of-open-elements">stack of open elements</a> does not <a href="#has-an-element-in-table-scope" title="has an element in table scope">have an element in table
6129: scope</a> with the same tag name as that of the token, then
6130: this is a <a href="#parse-error">parse error</a> and the token must be
6131: ignored.</p>
6132:
6133: <p>Otherwise:</p>
6134:
6135: <p><a href="#generate-implied-end-tags">Generate implied end tags</a>.</p>
6136:
6137: <p>Now, if the <a href="#current-node">current node</a> is not an element
6138: with the same tag name as the token, then this is a
6139: <a href="#parse-error">parse error</a>.</p>
6140:
6141: <p>Pop elements from this stack until an element with the same
6142: tag name as the token has been popped from the stack.</p>
6143:
6144: <p><a href="#clear-the-list-of-active-formatting-elements-up-to-the-last-marker">Clear the list of active formatting elements up to
6145: the last marker</a>.</p>
6146:
6147: <p>Switch the <a href="#insertion-mode">insertion mode</a> to "<a href="#parsing-main-intr" title="insertion mode: in row">in row</a>". (The
6148: <a href="#current-node">current node</a> will be a <code><a href="tabular-data.html#the-tr-element">tr</a></code> element at
6149: this point.)</p>
6150:
6151: </dd>
6152:
6153: <dt>A start tag whose tag name is one of: "caption", "col",
6154: "colgroup", "tbody", "td", "tfoot", "th", "thead", "tr"</dt>
6155: <dd>
6156:
6157: <p>If the <a href="#stack-of-open-elements">stack of open elements</a> does
6158: <em>not</em> <a href="#has-an-element-in-table-scope" title="has an element in table scope">have
6159: a <code>td</code> or <code>th</code> element in table
6160: scope</a>, then this is a <a href="#parse-error">parse error</a>; ignore
6161: the token. (<a href="#fragment-case">fragment case</a>)</p>
6162:
6163: <p>Otherwise, <a href="#close-the-cell">close the cell</a> (see below) and
6164: reprocess the current token.</p>
6165:
6166: </dd>
6167:
6168: <dt>An end tag whose tag name is one of: "body", "caption",
6169: "col", "colgroup", "html"</dt>
6170: <dd>
6171: <p><a href="#parse-error">Parse error</a>. Ignore the token.</p>
6172: </dd>
6173:
6174: <dt>An end tag whose tag name is one of: "table", "tbody",
6175: "tfoot", "thead", "tr"</dt>
6176: <dd>
6177:
6178: <p>If the <a href="#stack-of-open-elements">stack of open elements</a> does not <a href="#has-an-element-in-table-scope" title="has an element in table scope">have an element in table
6179: scope</a> with the same tag name as that of the token
6180: (which can only happen for "tbody", "tfoot" and "thead", or,
6181: in the <a href="#fragment-case">fragment case</a>), then this is a <a href="#parse-error">parse
6182: error</a> and the token must be ignored.</p>
6183:
6184: <p>Otherwise, <a href="#close-the-cell">close the cell</a> (see below) and
6185: reprocess the current token.</p>
6186:
6187: </dd>
6188:
6189: <dt>Anything else</dt>
6190: <dd>
6191: <p>Process the token <a href="#using-the-rules-for">using the rules for</a> the "<a href="#parsing-main-inbody" title="insertion mode: in body">in body</a>" <a href="#insertion-mode">insertion
6192: mode</a>.</p>
6193: </dd>
6194:
6195: </dl><p>Where the steps above say to <dfn id="close-the-cell">close the cell</dfn>, they
6196: mean to run the following algorithm:</p>
6197:
6198: <ol><li><p>If the <a href="#stack-of-open-elements">stack of open elements</a> <a href="#has-an-element-in-table-scope" title="has an element in table scope">has a <code>td</code>
6199: element in table scope</a>, then act as if an end tag token
6200: with the tag name "td" had been seen.</p></li>
6201:
6202: <li><p>Otherwise, the <a href="#stack-of-open-elements">stack of open elements</a> will
6203: <a href="#has-an-element-in-table-scope" title="has an element in table scope">have a
6204: <code>th</code> element in table scope</a>; act as if an end
6205: tag token with the tag name "th" had been seen.</p></li>
6206:
6207: </ol><p class="note">The <a href="#stack-of-open-elements">stack of open elements</a> cannot
6208: have both a <code><a href="tabular-data.html#the-td-element">td</a></code> and a <code><a href="tabular-data.html#the-th-element">th</a></code> element <a href="#has-an-element-in-table-scope" title="has an element in table scope">in table scope</a> at
6209: the same time, nor can it have neither when the <a href="#insertion-mode">insertion
6210: mode</a> is "<a href="#parsing-main-intd" title="insertion mode: in cell">in
6211: cell</a>".</p>
6212:
6213:
6214: <h5 id="parsing-main-inselect"><span class="secno">9.2.5.19 </span>The "<dfn title="insertion mode: in select">in select</dfn>" insertion mode</h5>
6215:
6216: <p>When the <a href="#insertion-mode">insertion mode</a> is "<a href="#parsing-main-inselect" title="insertion
6217: mode: in select">in select</a>", tokens must be handled as follows:</p>
6218:
6219: <dl class="switch"><dt>A character token</dt>
6220: <dd>
6221: <p><a href="#insert-a-character" title="insert a character">Insert the token's
6222: character</a> into the <a href="#current-node">current node</a>.</p>
6223: </dd>
6224:
6225: <dt>A comment token</dt>
6226: <dd>
6227: <p>Append a <code>Comment</code> node to the <a href="#current-node">current
6228: node</a> with the <code title="">data</code> attribute set to
6229: the data given in the comment token.</p>
6230: </dd>
6231:
6232: <dt>A DOCTYPE token</dt>
6233: <dd>
6234: <p><a href="#parse-error">Parse error</a>. Ignore the token.</p>
6235: </dd>
6236:
6237: <dt>A start tag whose tag name is "html"</dt>
6238: <dd>
6239: <p>Process the token <a href="#using-the-rules-for">using the rules for</a> the "<a href="#parsing-main-inbody" title="insertion mode: in body">in body</a>" <a href="#insertion-mode">insertion
6240: mode</a>.</p>
6241: </dd>
6242:
6243: <dt>A start tag whose tag name is "option"</dt>
6244: <dd>
6245:
6246: <p>If the <a href="#current-node">current node</a> is an <code><a href="forms.html#the-option-element">option</a></code>
6247: element, act as if an end tag with the tag name "option" had
6248: been seen.</p>
6249:
6250: <p><a href="#insert-an-html-element">Insert an HTML element</a> for the token.</p>
6251:
6252: </dd>
6253:
6254: <dt>A start tag whose tag name is "optgroup"</dt>
6255: <dd>
6256:
6257: <p>If the <a href="#current-node">current node</a> is an <code><a href="forms.html#the-option-element">option</a></code>
6258: element, act as if an end tag with the tag name "option" had
6259: been seen.</p>
6260:
6261: <p>If the <a href="#current-node">current node</a> is an
6262: <code><a href="forms.html#the-optgroup-element">optgroup</a></code> element, act as if an end tag with the
6263: tag name "optgroup" had been seen.</p>
6264:
6265: <p><a href="#insert-an-html-element">Insert an HTML element</a> for the token.</p>
6266:
6267: </dd>
6268:
6269: <dt>An end tag whose tag name is "optgroup"</dt>
6270: <dd>
6271:
6272: <p>First, if the <a href="#current-node">current node</a> is an
6273: <code><a href="forms.html#the-option-element">option</a></code> element, and the node immediately before
6274: it in the <a href="#stack-of-open-elements">stack of open elements</a> is an
6275: <code><a href="forms.html#the-optgroup-element">optgroup</a></code> element, then act as if an end tag with
6276: the tag name "option" had been seen.</p>
6277:
6278: <p>If the <a href="#current-node">current node</a> is an
6279: <code><a href="forms.html#the-optgroup-element">optgroup</a></code> element, then pop that node from the
6280: <a href="#stack-of-open-elements">stack of open elements</a>. Otherwise, this is a
6281: <a href="#parse-error">parse error</a>; ignore the token.</p>
6282:
6283: </dd>
6284:
6285: <dt>An end tag whose tag name is "option"</dt>
6286: <dd>
6287:
6288: <p>If the <a href="#current-node">current node</a> is an <code><a href="forms.html#the-option-element">option</a></code>
6289: element, then pop that node from the <a href="#stack-of-open-elements">stack of open
6290: elements</a>. Otherwise, this is a <a href="#parse-error">parse
6291: error</a>; ignore the token.</p>
6292:
6293: </dd>
6294:
6295: <dt>An end tag whose tag name is "select"</dt>
6296: <dd>
6297:
6298: <p>If the <a href="#stack-of-open-elements">stack of open elements</a> does not <a href="#has-an-element-in-table-scope" title="has an element in table scope">have an element in table
6299: scope</a> with the same tag name as the token, this is a
6300: <a href="#parse-error">parse error</a>. Ignore the token. (<a href="#fragment-case">fragment
6301: case</a>)</p>
6302:
6303: <p>Otherwise:</p>
6304:
6305: <p>Pop elements from the <a href="#stack-of-open-elements">stack of open elements</a>
6306: until a <code><a href="forms.html#the-select-element">select</a></code> element has been popped from the
6307: stack.</p>
6308:
6309: <p><a href="#reset-the-insertion-mode-appropriately">Reset the insertion mode appropriately</a>.</p>
6310:
6311: </dd>
6312:
6313: <dt>A start tag whose tag name is "select"</dt>
6314: <dd>
6315:
6316: <p><a href="#parse-error">Parse error</a>. Act as if the token had been
6317: an end tag with the tag name "select" instead.</p>
6318:
6319: </dd>
6320:
6321: <dt>A start tag whose tag name is one of: "input", "keygen", "textarea"</dt>
6322: <dd>
1.73 mike 6323:
6324: <p><a href="#parse-error">Parse error</a>.</p>
6325:
1.124 mike 6326: <p>If the <a href="#stack-of-open-elements">stack of open elements</a> does not <a href="#has-an-element-in-table-scope" title="has an element in table scope">have a <code>select</code>
6327: element in table scope</a>, ignore the token. (<a href="#fragment-case">fragment
6328: case</a>)</p>
1.73 mike 6329:
6330: <p>Otherwise, act as if an end tag with the tag name "select" had
6331: been seen, and reprocess the token.</p>
6332:
1.1 mike 6333: </dd>
6334:
6335: <dt>A start tag token whose tag name is "script"</dt>
6336: <dd>
6337: <p>Process the token <a href="#using-the-rules-for">using the rules for</a> the "<a href="#parsing-main-inhead" title="insertion mode: in head">in head</a>" <a href="#insertion-mode">insertion
6338: mode</a>.</p>
6339: </dd>
6340:
6341: <dt>An end-of-file token</dt>
6342: <dd>
6343:
6344: <p>If the <a href="#current-node">current node</a> is not the root
1.159 mike 6345: <code><a href="semantics.html#the-html-element-0">html</a></code> element, then this is a <a href="#parse-error">parse
1.1 mike 6346: error</a>.</p>
6347:
6348: <p class="note">It can only be the <a href="#current-node">current node</a> in
6349: the <a href="#fragment-case">fragment case</a>.</p>
6350:
6351: <p><a href="#stop-parsing">Stop parsing</a>.</p>
6352:
6353: </dd>
6354:
6355: <dt>Anything else</dt>
6356: <dd>
6357: <p><a href="#parse-error">Parse error</a>. Ignore the token.</p>
6358: </dd>
6359:
6360: </dl><h5 id="parsing-main-inselectintable"><span class="secno">9.2.5.20 </span>The "<dfn title="insertion mode: in select in table">in select in table</dfn>" insertion mode</h5>
6361:
6362: <p>When the <a href="#insertion-mode">insertion mode</a> is "<a href="#parsing-main-inselectintable" title="insertion
6363: mode: in select in table">in select in table</a>", tokens must be handled as follows:</p>
6364:
6365: <dl class="switch"><dt>A start tag whose tag name is one of: "caption", "table",
6366: "tbody", "tfoot", "thead", "tr", "td", "th"</dt>
6367: <dd>
6368: <p><a href="#parse-error">Parse error</a>. Act as if an end tag with the tag
6369: name "select" had been seen, and reprocess the token.</p>
6370: </dd>
6371:
6372: <dt>An end tag whose tag name is one of: "caption", "table",
6373: "tbody", "tfoot", "thead", "tr", "td", "th"</dt>
6374: <dd>
6375:
6376: <p><a href="#parse-error">Parse error</a>.</p>
6377:
6378: <p>If the <a href="#stack-of-open-elements">stack of open elements</a> <a href="#has-an-element-in-table-scope">has an
6379: element in table scope</a> with the same tag name as that
6380: of the token, then act as if an end tag with the tag name
6381: "select" had been seen, and reprocess the token. Otherwise,
6382: ignore the token.</p>
6383:
6384: </dd>
6385:
6386: <dt>Anything else</dt>
6387: <dd>
6388: <p>Process the token <a href="#using-the-rules-for">using the rules for</a> the "<a href="#parsing-main-inselect" title="insertion mode: in select">in select</a>" <a href="#insertion-mode">insertion
6389: mode</a>.</p>
6390: </dd>
6391:
1.168 mike 6392: </dl><h5 id="parsing-main-inforeign"><span class="secno">9.2.5.21 </span>The "<dfn title="insertion mode: in foreign content">in foreign content</dfn>" insertion mode</h5>
1.1 mike 6393:
6394: <p>When the <a href="#insertion-mode">insertion mode</a> is "<a href="#parsing-main-inforeign" title="insertion
6395: mode: in foreign content">in foreign content</a>", tokens must be
6396: handled as follows:</p>
6397:
6398: <dl class="switch"><dt>A character token</dt>
6399: <dd>
6400:
6401: <p><a href="#insert-a-character" title="insert a character">Insert the token's
6402: character</a> into the <a href="#current-node">current node</a>.</p>
6403:
6404: <p>If the token is not one of U+0009 CHARACTER TABULATION, U+000A
6405: LINE FEED (LF), U+000C FORM FEED (FF), <!--U+000D CARRIAGE RETURN
6406: (CR),--> or U+0020 SPACE, then set the <a href="#frameset-ok-flag">frameset-ok
6407: flag</a> to "not ok".</p>
6408:
6409: </dd>
6410:
6411: <dt>A comment token</dt>
6412: <dd>
6413: <p>Append a <code>Comment</code> node to the <a href="#current-node">current
6414: node</a> with the <code title="">data</code> attribute set to
6415: the data given in the comment token.</p>
6416: </dd>
6417:
6418: <dt>A DOCTYPE token</dt>
6419: <dd>
6420: <p><a href="#parse-error">Parse error</a>. Ignore the token.</p>
6421: </dd>
6422:
6423: <dt>An end tag whose tag name is "script", if the <a href="#current-node">current node</a> is a <code title="">script</code> element in the <a href="#svg-namespace">SVG namespace</a>.</dt>
6424: <dd>
6425:
6426: <p>Pop the <a href="#current-node">current node</a> off the <a href="#stack-of-open-elements">stack of open
6427: elements</a>.</p>
6428:
6429: <p>Let the <var title="">old insertion point</var> have the
6430: same value as the current <a href="#insertion-point">insertion point</a>. Let
6431: the <a href="#insertion-point">insertion point</a> be just before the <a href="#next-input-character">next
6432: input character</a>.</p>
6433:
6434: <p>Increment the parser's <a href="#script-nesting-level">script nesting level</a> by
6435: one. Set the <a href="#parser-pause-flag">parser pause flag</a> to true.</p>
6436:
6437: <p><a href="http://www.w3.org/TR/SVGMobile12/script.html#ScriptContentProcessing">Process
6438: the <code title="">script</code> element</a> according to the SVG
1.187 mike 6439: rules, if the user agent supports SVG. <a href="references.html#refsSVG">[SVG]</a></p>
1.1 mike 6440:
1.159 mike 6441: <p class="note">Even if this causes <a href="embedded-content-0.html#dom-document-write" title="dom-document-write">new characters to be inserted into the
1.1 mike 6442: tokenizer</a>, the parser will not be executed reentrantly,
6443: since the <a href="#parser-pause-flag">parser pause flag</a> is true.</p>
6444:
6445: <p>Decrement the parser's <a href="#script-nesting-level">script nesting level</a> by
6446: one. If the parser's <a href="#script-nesting-level">script nesting level</a> is zero,
6447: then set the <a href="#parser-pause-flag">parser pause flag</a> to false.</p>
6448:
6449: <p>Let the <a href="#insertion-point">insertion point</a> have the value of the <var title="">old insertion point</var>. (In other words, restore the
6450: <a href="#insertion-point">insertion point</a> to its previous value. This value
6451: might be the "undefined" value.)</p>
6452:
6453: </dd>
6454:
6455: <dt>A start tag whose tag name is neither "mglyph" nor "malignmark", if the <a href="#current-node">current node</a> is an <code title="">mi</code> element in the <a href="#mathml-namespace">MathML namespace</a>.</dt>
6456: <dt>A start tag whose tag name is neither "mglyph" nor "malignmark", if the <a href="#current-node">current node</a> is an <code title="">mo</code> element in the <a href="#mathml-namespace">MathML namespace</a>.</dt>
6457: <dt>A start tag whose tag name is neither "mglyph" nor "malignmark", if the <a href="#current-node">current node</a> is an <code title="">mn</code> element in the <a href="#mathml-namespace">MathML namespace</a>.</dt>
6458: <dt>A start tag whose tag name is neither "mglyph" nor "malignmark", if the <a href="#current-node">current node</a> is an <code title="">ms</code> element in the <a href="#mathml-namespace">MathML namespace</a>.</dt>
6459: <dt>A start tag whose tag name is neither "mglyph" nor "malignmark", if the <a href="#current-node">current node</a> is an <code title="">mtext</code> element in the <a href="#mathml-namespace">MathML namespace</a>.</dt>
6460: <dt>A start tag whose tag name is "svg", if the <a href="#current-node">current node</a> is an <code title="">annotation-xml</code> element in the <a href="#mathml-namespace">MathML namespace</a>.</dt>
6461: <dt>A start tag, if the <a href="#current-node">current node</a> is a <code title="">foreignObject</code> element in the <a href="#svg-namespace">SVG namespace</a>.</dt>
6462: <dt>A start tag, if the <a href="#current-node">current node</a> is a <code title="">desc</code> element in the <a href="#svg-namespace">SVG namespace</a>.</dt>
6463: <dt>A start tag, if the <a href="#current-node">current node</a> is a <code title="">title</code> element in the <a href="#svg-namespace">SVG namespace</a>.</dt>
6464: <dt>A start tag, if the <a href="#current-node">current node</a> is an element in the <a href="#html-namespace-0">HTML namespace</a>.</dt>
6465: <dt>An end tag</dt>
6466: <dd>
6467:
6468: <p>Process the token <a href="#using-the-rules-for">using the rules for</a> the
6469: <a href="#secondary-insertion-mode">secondary insertion mode</a>.</p>
6470:
6471: <p>If, after doing so, the <a href="#insertion-mode">insertion mode</a> is still
6472: "<a href="#parsing-main-inforeign" title="insertion mode: in foreign content">in foreign
6473: content</a>", but there is no element in scope that has a
6474: namespace other than the <a href="#html-namespace-0">HTML namespace</a><!-- XXX this
6475: isn't well-defined; it means check to see if the bottom-most HTML
6476: element on the stack is in scope -->, switch the <a href="#insertion-mode">insertion
6477: mode</a> to the <a href="#secondary-insertion-mode">secondary insertion mode</a>.</p>
6478:
6479: </dd>
6480:
6481: <dt>A start tag whose tag name is one of: <!--"a",--> "b", "big",
6482: "blockquote", "body"<!--by inspection-->, "br", "center", "code",
6483: "dd", "div", "dl", "dt"<!-- so that dd and dt can be handled
6484: uniformly throughout the parser -->, "em", "embed", "h1", "h2",
6485: "h3", "h4"<!--for completeness-->, "h5", "h6"<!--for
6486: completeness-->, "head"<!--by inspection-->, "hr", "i", "img",
6487: "li", "listing"<!-- so that pre and listing can be handled
6488: uniformly throughout the parser -->, "menu", "meta", "nobr",
6489: "ol"<!-- so that dl, ul, and ol can be handled uniformly throughout
6490: the parser -->, "p", "pre", "ruby", "s", <!--"script",--> "small",
6491: "span", "strong", "strike"<!-- so that s and strike can be handled
6492: uniformly throughout the parser -->, <!--"style",--> "sub", "sup",
6493: "table"<!--by inspection-->, "tt", "u", "ul", "var"</dt>
6494: <!-- this list was determined empirically by studying over
6495: 6,000,000,000 pages that were specifically not XML pages -->
6496: <dt>A start tag whose tag name is "font", if the token has
6497: any attributes named "color", "face", or "size"</dt>
6498: <!-- the attributes here are required so that SVG <font> will go
6499: through as SVG but legacy <font>s won't -->
6500: <dt>An end-of-file token</dt>
6501: <dd>
6502:
6503: <p><a href="#parse-error">Parse error</a>.</p>
6504:
6505: <p>Pop elements from the <a href="#stack-of-open-elements">stack of open elements</a> until
6506: the <a href="#current-node">current node</a> is in the <a href="#html-namespace-0">HTML
6507: namespace</a>.</p>
6508:
6509: <p>Switch the <a href="#insertion-mode">insertion mode</a> to the <a href="#secondary-insertion-mode">secondary
6510: insertion mode</a>, and reprocess the token.</p>
6511:
6512: </dd>
6513:
6514: <dt>Any other start tag</dt>
6515: <dd>
6516:
6517: <p>If the <a href="#current-node">current node</a> is an element in the
6518: <a href="#mathml-namespace">MathML namespace</a>, <a href="#adjust-mathml-attributes">adjust MathML
6519: attributes</a> for the token. (This fixes the case of MathML
6520: attributes that are not all lowercase.)</p>
6521:
6522: <p>If the <a href="#current-node">current node</a> is an element in the <a href="#svg-namespace">SVG
6523: namespace</a>, and the token's tag name is one of the ones in
6524: the first column of the following table, change the tag name to
6525: the name given in the corresponding cell in the second
6526: column. (This fixes the case of SVG elements that are not all
6527: lowercase.)</p>
6528:
6529: <table><thead><tr><th> Tag name </th><th> Element name
6530: </th></tr></thead><tbody><tr><td> <code title="">altglyph</code> </td><td> <code title="">altGlyph</code>
6531: </td></tr><tr><td> <code title="">altglyphdef</code> </td><td> <code title="">altGlyphDef</code>
6532: </td></tr><tr><td> <code title="">altglyphitem</code> </td><td> <code title="">altGlyphItem</code>
6533: </td></tr><tr><td> <code title="">animatecolor</code> </td><td> <code title="">animateColor</code>
6534: </td></tr><tr><td> <code title="">animatemotion</code> </td><td> <code title="">animateMotion</code>
6535: </td></tr><tr><td> <code title="">animatetransform</code> </td><td> <code title="">animateTransform</code>
6536: </td></tr><tr><td> <code title="">clippath</code> </td><td> <code title="">clipPath</code>
6537: </td></tr><tr><td> <code title="">feblend</code> </td><td> <code title="">feBlend</code>
6538: </td></tr><tr><td> <code title="">fecolormatrix</code> </td><td> <code title="">feColorMatrix</code>
6539: </td></tr><tr><td> <code title="">fecomponenttransfer</code> </td><td> <code title="">feComponentTransfer</code>
6540: </td></tr><tr><td> <code title="">fecomposite</code> </td><td> <code title="">feComposite</code>
6541: </td></tr><tr><td> <code title="">feconvolvematrix</code> </td><td> <code title="">feConvolveMatrix</code>
6542: </td></tr><tr><td> <code title="">fediffuselighting</code> </td><td> <code title="">feDiffuseLighting</code>
6543: </td></tr><tr><td> <code title="">fedisplacementmap</code> </td><td> <code title="">feDisplacementMap</code>
6544: </td></tr><tr><td> <code title="">fedistantlight</code> </td><td> <code title="">feDistantLight</code>
6545: </td></tr><tr><td> <code title="">feflood</code> </td><td> <code title="">feFlood</code>
6546: </td></tr><tr><td> <code title="">fefunca</code> </td><td> <code title="">feFuncA</code>
6547: </td></tr><tr><td> <code title="">fefuncb</code> </td><td> <code title="">feFuncB</code>
6548: </td></tr><tr><td> <code title="">fefuncg</code> </td><td> <code title="">feFuncG</code>
6549: </td></tr><tr><td> <code title="">fefuncr</code> </td><td> <code title="">feFuncR</code>
6550: </td></tr><tr><td> <code title="">fegaussianblur</code> </td><td> <code title="">feGaussianBlur</code>
6551: </td></tr><tr><td> <code title="">feimage</code> </td><td> <code title="">feImage</code>
6552: </td></tr><tr><td> <code title="">femerge</code> </td><td> <code title="">feMerge</code>
6553: </td></tr><tr><td> <code title="">femergenode</code> </td><td> <code title="">feMergeNode</code>
6554: </td></tr><tr><td> <code title="">femorphology</code> </td><td> <code title="">feMorphology</code>
6555: </td></tr><tr><td> <code title="">feoffset</code> </td><td> <code title="">feOffset</code>
6556: </td></tr><tr><td> <code title="">fepointlight</code> </td><td> <code title="">fePointLight</code>
6557: </td></tr><tr><td> <code title="">fespecularlighting</code> </td><td> <code title="">feSpecularLighting</code>
6558: </td></tr><tr><td> <code title="">fespotlight</code> </td><td> <code title="">feSpotLight</code>
6559: </td></tr><tr><td> <code title="">fetile</code> </td><td> <code title="">feTile</code>
6560: </td></tr><tr><td> <code title="">feturbulence</code> </td><td> <code title="">feTurbulence</code>
6561: </td></tr><tr><td> <code title="">foreignobject</code> </td><td> <code title="">foreignObject</code>
6562: </td></tr><tr><td> <code title="">glyphref</code> </td><td> <code title="">glyphRef</code>
6563: </td></tr><tr><td> <code title="">lineargradient</code> </td><td> <code title="">linearGradient</code>
6564: </td></tr><tr><td> <code title="">radialgradient</code> </td><td> <code title="">radialGradient</code>
6565: <!--<tr> <td> <code title="">solidcolor</code> <td> <code title="">solidColor</code> (SVG 1.2)-->
6566: </td></tr><tr><td> <code title="">textpath</code> </td><td> <code title="">textPath</code>
6567: </td></tr></tbody></table><p>If the <a href="#current-node">current node</a> is an element in the <a href="#svg-namespace">SVG
6568: namespace</a>, <a href="#adjust-svg-attributes">adjust SVG attributes</a> for the
6569: token. (This fixes the case of SVG attributes that are not all
6570: lowercase.)</p>
6571:
6572: <p><a href="#adjust-foreign-attributes">Adjust foreign attributes</a> for the token. (This
6573: fixes the use of namespaced attributes, in particular XLink in
6574: SVG.)</p>
6575:
6576: <p><a href="#insert-a-foreign-element">Insert a foreign element</a> for the token, in the
6577: same namespace as the <a href="#current-node">current node</a>.</p>
6578:
6579: <p>If the token has its <i>self-closing flag</i> set, pop the
6580: <a href="#current-node">current node</a> off the <a href="#stack-of-open-elements">stack of open
6581: elements</a> and <a href="#acknowledge-self-closing-flag" title="acknowledge self-closing
6582: flag">acknowledge the token's <i>self-closing flag</i></a>.</p>
6583:
6584: </dd>
6585:
6586: </dl><h5 id="parsing-main-afterbody"><span class="secno">9.2.5.22 </span>The "<dfn title="insertion mode: after body">after body</dfn>" insertion mode</h5>
6587:
6588: <p>When the <a href="#insertion-mode">insertion mode</a> is "<a href="#parsing-main-afterbody" title="insertion
6589: mode: after body">after body</a>", tokens must be handled as follows:</p>
6590:
6591: <dl class="switch"><dt>A character token that is one of U+0009 CHARACTER
6592: TABULATION, U+000A LINE FEED (LF), U+000C FORM FEED (FF),
6593: <!--U+000D CARRIAGE RETURN (CR),--> or U+0020 SPACE</dt>
6594: <dd>
6595: <p>Process the token <a href="#using-the-rules-for">using the rules for</a> the "<a href="#parsing-main-inbody" title="insertion mode: in body">in body</a>" <a href="#insertion-mode">insertion
6596: mode</a>.</p>
6597: </dd>
6598:
6599: <dt>A comment token</dt>
6600: <dd>
6601: <p>Append a <code>Comment</code> node to the first element in
1.159 mike 6602: the <a href="#stack-of-open-elements">stack of open elements</a> (the <code><a href="semantics.html#the-html-element-0">html</a></code>
1.1 mike 6603: element), with the <code title="">data</code> attribute set to
6604: the data given in the comment token.</p>
6605: </dd>
6606:
6607: <dt>A DOCTYPE token</dt>
6608: <dd>
6609: <p><a href="#parse-error">Parse error</a>. Ignore the token.</p>
6610: </dd>
6611:
6612: <dt>A start tag whose tag name is "html"</dt>
6613: <dd>
6614: <p>Process the token <a href="#using-the-rules-for">using the rules for</a> the "<a href="#parsing-main-inbody" title="insertion mode: in body">in body</a>" <a href="#insertion-mode">insertion
6615: mode</a>.</p>
6616: </dd>
6617:
6618: <dt>An end tag whose tag name is "html"</dt>
6619: <dd>
6620:
6621: <p>If the parser was originally created as part of the <a href="#html-fragment-parsing-algorithm">HTML
6622: fragment parsing algorithm</a>, this is a <a href="#parse-error">parse
6623: error</a>; ignore the token. (<a href="#fragment-case">fragment case</a>)</p>
6624:
6625: <p>Otherwise, switch the <a href="#insertion-mode">insertion mode</a> to "<a href="#the-after-after-body-insertion-mode" title="insertion mode: after after body">after after
6626: body</a>".</p>
6627:
6628: </dd>
6629:
6630: <dt>An end-of-file token</dt>
6631: <dd>
6632: <p><a href="#stop-parsing">Stop parsing</a>.</p>
6633: </dd>
6634:
6635: <dt>Anything else</dt>
6636: <dd>
6637:
6638: <p><a href="#parse-error">Parse error</a>. Switch the <a href="#insertion-mode">insertion
6639: mode</a> to "<a href="#parsing-main-inbody" title="insertion mode: in body">in
6640: body</a>" and reprocess the token.</p>
6641:
6642: </dd>
6643:
6644: </dl><h5 id="parsing-main-inframeset"><span class="secno">9.2.5.23 </span>The "<dfn title="insertion mode: in frameset">in frameset</dfn>" insertion mode</h5>
6645:
6646: <p>When the <a href="#insertion-mode">insertion mode</a> is "<a href="#parsing-main-inframeset" title="insertion
6647: mode: in frameset">in frameset</a>", tokens must be handled as follows:</p>
6648:
6649: <dl class="switch"><dt>A character token that is one of U+0009 CHARACTER
6650: TABULATION, U+000A LINE FEED (LF), U+000C FORM FEED (FF),
6651: <!--U+000D CARRIAGE RETURN (CR),--> or U+0020 SPACE</dt>
6652: <dd>
6653: <p><a href="#insert-a-character" title="insert a character">Insert the character</a> into
6654: the <a href="#current-node">current node</a>.</p>
6655: </dd>
6656:
6657: <dt>A comment token</dt>
6658: <dd>
6659: <p>Append a <code>Comment</code> node to the <a href="#current-node">current
6660: node</a> with the <code title="">data</code> attribute set to
6661: the data given in the comment token.</p>
6662: </dd>
6663:
6664: <dt>A DOCTYPE token</dt>
6665: <dd>
6666: <p><a href="#parse-error">Parse error</a>. Ignore the token.</p>
6667: </dd>
6668:
6669: <dt>A start tag whose tag name is "html"</dt>
6670: <dd>
6671: <p>Process the token <a href="#using-the-rules-for">using the rules for</a> the "<a href="#parsing-main-inbody" title="insertion mode: in body">in body</a>" <a href="#insertion-mode">insertion
6672: mode</a>.</p>
6673: </dd>
6674:
6675: <dt>A start tag whose tag name is "frameset"</dt>
6676: <dd>
6677: <p><a href="#insert-an-html-element">Insert an HTML element</a> for the token.</p>
6678: </dd>
6679:
6680: <dt>An end tag whose tag name is "frameset"</dt>
6681: <dd>
6682:
6683: <p>If the <a href="#current-node">current node</a> is the root
1.159 mike 6684: <code><a href="semantics.html#the-html-element-0">html</a></code> element, then this is a <a href="#parse-error">parse
1.1 mike 6685: error</a>; ignore the token. (<a href="#fragment-case">fragment
6686: case</a>)</p>
6687:
6688: <p>Otherwise, pop the <a href="#current-node">current node</a> from the
6689: <a href="#stack-of-open-elements">stack of open elements</a>.</p>
6690:
6691: <p>If the parser was <em>not</em> originally created as part
6692: of the <a href="#html-fragment-parsing-algorithm">HTML fragment parsing algorithm</a>
6693: (<a href="#fragment-case">fragment case</a>), and the <a href="#current-node">current
1.5 mike 6694: node</a> is no longer a <code><a href="obsolete.html#frameset">frameset</a></code> element, then
1.1 mike 6695: switch the <a href="#insertion-mode">insertion mode</a> to "<a href="#parsing-main-afterframeset" title="insertion mode: after frameset">after
6696: frameset</a>".</p>
6697:
6698: </dd>
6699:
6700: <dt>A start tag whose tag name is "frame"</dt>
6701: <dd>
6702:
6703: <p><a href="#insert-an-html-element">Insert an HTML element</a> for the token.
6704: Immediately pop the <a href="#current-node">current node</a> off the
6705: <a href="#stack-of-open-elements">stack of open elements</a>.</p>
6706:
6707: <p><a href="#acknowledge-self-closing-flag" title="acknowledge self-closing flag">Acknowledge the
6708: token's <i>self-closing flag</i></a>, if it is set.</p>
6709:
6710: </dd>
6711:
6712: <dt>A start tag whose tag name is "noframes"</dt>
6713: <dd>
6714: <p>Process the token <a href="#using-the-rules-for">using the rules for</a> the "<a href="#parsing-main-inhead" title="insertion mode: in head">in head</a>" <a href="#insertion-mode">insertion
6715: mode</a>.</p>
6716: </dd>
6717:
6718: <dt>An end-of-file token</dt>
6719: <dd>
6720:
6721: <p>If the <a href="#current-node">current node</a> is not the root
1.159 mike 6722: <code><a href="semantics.html#the-html-element-0">html</a></code> element, then this is a <a href="#parse-error">parse
1.1 mike 6723: error</a>.</p>
6724:
6725: <p class="note">It can only be the <a href="#current-node">current node</a> in
6726: the <a href="#fragment-case">fragment case</a>.</p>
6727:
6728: <p><a href="#stop-parsing">Stop parsing</a>.</p>
6729:
6730: </dd>
6731:
6732: <dt>Anything else</dt>
6733: <dd>
6734: <p><a href="#parse-error">Parse error</a>. Ignore the token.</p>
6735: </dd>
6736:
6737: </dl><h5 id="parsing-main-afterframeset"><span class="secno">9.2.5.24 </span>The "<dfn title="insertion mode: after frameset">after frameset</dfn>" insertion mode</h5>
6738:
6739: <p>When the <a href="#insertion-mode">insertion mode</a> is "<a href="#parsing-main-afterframeset" title="insertion
6740: mode: after frameset">after frameset</a>", tokens must be handled as follows:</p>
6741:
6742: <!-- due to rules in the "in frameset" mode, this can't be entered in the fragment case -->
6743: <dl class="switch"><dt>A character token that is one of U+0009 CHARACTER
6744: TABULATION, U+000A LINE FEED (LF), U+000C FORM FEED (FF),
6745: <!--U+000D CARRIAGE RETURN (CR),--> or U+0020 SPACE</dt>
6746: <dd>
6747: <p><a href="#insert-a-character" title="insert a character">Insert the character</a> into
6748: the <a href="#current-node">current node</a>.</p>
6749: </dd>
6750:
6751: <dt>A comment token</dt>
6752: <dd>
6753: <p>Append a <code>Comment</code> node to the <a href="#current-node">current
6754: node</a> with the <code title="">data</code> attribute set to
6755: the data given in the comment token.</p>
6756: </dd>
6757:
6758: <dt>A DOCTYPE token</dt>
6759: <dd>
6760: <p><a href="#parse-error">Parse error</a>. Ignore the token.</p>
6761: </dd>
6762:
6763: <dt>A start tag whose tag name is "html"</dt>
6764: <dd>
6765: <p>Process the token <a href="#using-the-rules-for">using the rules for</a> the "<a href="#parsing-main-inbody" title="insertion mode: in body">in body</a>" <a href="#insertion-mode">insertion
6766: mode</a>.</p>
6767: </dd>
6768:
6769: <dt>An end tag whose tag name is "html"</dt>
6770: <dd>
6771: <p>Switch the <a href="#insertion-mode">insertion mode</a> to "<a href="#the-after-after-frameset-insertion-mode" title="insertion mode: after after frameset">after after
6772: frameset</a>".</p>
6773: </dd>
6774:
6775: <dt>A start tag whose tag name is "noframes"</dt>
6776: <dd>
6777: <p>Process the token <a href="#using-the-rules-for">using the rules for</a> the "<a href="#parsing-main-inhead" title="insertion mode: in head">in head</a>" <a href="#insertion-mode">insertion
6778: mode</a>.</p>
6779: </dd>
6780:
6781: <dt>An end-of-file token</dt>
6782: <dd>
6783: <p><a href="#stop-parsing">Stop parsing</a>.</p>
6784: </dd>
6785:
6786: <dt>Anything else</dt>
6787: <dd>
6788: <p><a href="#parse-error">Parse error</a>. Ignore the token.</p>
6789: </dd>
6790:
6791: </dl><p class="XXX">This doesn't handle UAs that don't support
6792: frames, or that do support frames but want to show the NOFRAMES
6793: content. Supporting the former is easy; supporting the latter is
6794: harder.</p>
6795:
6796:
6797: <h5 id="the-after-after-body-insertion-mode"><span class="secno">9.2.5.25 </span>The "<dfn title="insertion mode: after after body">after after body</dfn>" insertion mode</h5>
6798:
6799: <p>When the <a href="#insertion-mode">insertion mode</a> is "<a href="#the-after-after-body-insertion-mode" title="insertion
6800: mode: after after body">after after body</a>", tokens must be handled as follows:</p>
6801:
6802: <dl class="switch"><dt>A comment token</dt>
6803: <dd>
6804: <p>Append a <code>Comment</code> node to the <code>Document</code>
6805: object with the <code title="">data</code> attribute set to the
6806: data given in the comment token.</p>
6807: </dd>
6808:
6809: <dt>A DOCTYPE token</dt>
6810: <dt>A character token that is one of U+0009 CHARACTER
6811: TABULATION, U+000A LINE FEED (LF), U+000C FORM FEED (FF),
6812: <!--U+000D CARRIAGE RETURN (CR),--> or U+0020 SPACE</dt>
6813: <dt>A start tag whose tag name is "html"</dt>
6814: <dd>
6815: <p>Process the token <a href="#using-the-rules-for">using the rules for</a> the "<a href="#parsing-main-inbody" title="insertion mode: in body">in body</a>" <a href="#insertion-mode">insertion
6816: mode</a>.</p>
6817: </dd>
6818:
6819: <dt>An end-of-file token</dt>
6820: <dd>
6821: <p><a href="#stop-parsing">Stop parsing</a>.</p>
6822: </dd>
6823:
6824: <dt>Anything else</dt>
6825: <dd>
6826: <p><a href="#parse-error">Parse error</a>. Switch the <a href="#insertion-mode">insertion mode</a>
6827: to "<a href="#parsing-main-inbody" title="insertion mode: in body">in body</a>" and
6828: reprocess the token.</p>
6829: </dd>
6830:
6831: </dl><h5 id="the-after-after-frameset-insertion-mode"><span class="secno">9.2.5.26 </span>The "<dfn title="insertion mode: after after frameset">after after frameset</dfn>" insertion mode</h5>
6832:
6833: <p>When the <a href="#insertion-mode">insertion mode</a> is "<a href="#the-after-after-frameset-insertion-mode" title="insertion
6834: mode: after after frameset">after after frameset</a>", tokens must be handled as follows:</p>
6835:
6836: <dl class="switch"><dt>A comment token</dt>
6837: <dd>
6838: <p>Append a <code>Comment</code> node to the <code>Document</code>
6839: object with the <code title="">data</code> attribute set to the
6840: data given in the comment token.</p>
6841: </dd>
6842:
6843: <dt>A DOCTYPE token</dt>
6844: <dt>A character token that is one of U+0009 CHARACTER
6845: TABULATION, U+000A LINE FEED (LF), U+000C FORM FEED (FF),
6846: <!--U+000D CARRIAGE RETURN (CR),--> or U+0020 SPACE</dt>
6847: <dt>A start tag whose tag name is "html"</dt>
6848: <dd>
6849: <p>Process the token <a href="#using-the-rules-for">using the rules for</a> the "<a href="#parsing-main-inbody" title="insertion mode: in body">in body</a>" <a href="#insertion-mode">insertion
6850: mode</a>.</p>
6851: </dd>
6852:
6853: <dt>An end-of-file token</dt>
6854: <dd>
6855: <p><a href="#stop-parsing">Stop parsing</a>.</p>
6856: </dd>
6857:
6858: <dt>A start tag whose tag name is "noframes"</dt>
6859: <dd>
6860: <p>Process the token <a href="#using-the-rules-for">using the rules for</a> the "<a href="#parsing-main-inhead" title="insertion mode: in head">in head</a>" <a href="#insertion-mode">insertion
6861: mode</a>.</p>
6862: </dd>
6863:
6864: <dt>Anything else</dt>
6865: <dd>
6866: <p><a href="#parse-error">Parse error</a>. Ignore the token.</p>
6867: </dd>
6868:
6869: </dl><h4 id="the-end"><span class="secno">9.2.6 </span>The end</h4>
6870:
6871: <p>Once the user agent <dfn id="stop-parsing" title="stop parsing">stops parsing</dfn>
6872: the document, the user agent must follow the steps in this
6873: section.</p>
6874:
1.10 mike 6875: <!-- this happens as part of one of the tasks that runs the parser -->
1.1 mike 6876:
1.10 mike 6877: <p>First, the user agent must set the <a href="dom.html#current-document-readiness">current document
6878: readiness</a> to "interactive" and the <a href="#insertion-point">insertion
6879: point</a> to undefined.</p>
6880:
6881: <p>Then, the user agent must then make a list of all the scripts
6882: that are in the <a href="semantics.html#list-of-scripts-that-will-execute-when-the-document-has-finished-parsing">list of scripts that will execute when the
6883: document has finished parsing</a>, the <a href="semantics.html#list-of-scripts-that-will-execute-asynchronously">list of scripts that
6884: will execute asynchronously</a>, and the <a href="semantics.html#list-of-scripts-that-will-execute-as-soon-as-possible">list of scripts
6885: that will execute as soon as possible</a>. This is the <dfn id="list-of-scripts-pending-after-the-parser-stopped">list
6886: of scripts pending after the parser stopped</dfn>.</p>
6887:
6888: <p>The rules for <a href="semantics.html#when-a-script-completes-loading">when a script completes loading</a> start
6889: applying (script execution is no longer managed by the parser).</p>
1.1 mike 6890:
6891: <p>If any of the scripts in the <a href="semantics.html#list-of-scripts-that-will-execute-as-soon-as-possible">list of scripts that will
6892: execute as soon as possible</a> have <a href="semantics.html#completed-loading">completed
6893: loading</a>, or if the <a href="semantics.html#list-of-scripts-that-will-execute-asynchronously">list of scripts that will execute
6894: asynchronously</a> is not empty and the first script in that list
6895: has <a href="semantics.html#completed-loading">completed loading</a>, then the user agent must act as
6896: if those scripts just completed loading, following the rules given
6897: for that in the <code><a href="semantics.html#script">script</a></code> element definition.</p>
6898:
1.10 mike 6899: <p>If the <a href="semantics.html#list-of-scripts-that-will-execute-when-the-document-has-finished-parsing">list of scripts that will execute when the document
6900: has finished parsing</a> is not empty, and the first item in this
6901: list has already <a href="semantics.html#completed-loading">completed loading</a>, then the user agent
6902: must act as if that script just finished loading.</p>
6903:
6904: <p class="note">By this point, there will be no scripts that have
6905: loaded but have not yet been executed.</p>
6906:
6907: <hr><!-- async --><p>Once all the scripts on the <a href="#list-of-scripts-pending-after-the-parser-stopped">list of scripts pending after
6908: the parser stopped</a> have <a href="semantics.html#completed-loading">completed loading</a> and
6909: been <a href="semantics.html#executing-a-script-block" title="executing a script block">executed</a>, the
6910: user agent must <a href="browsers.html#queue-a-task">queue a task</a> to <a href="browsers.html#fire-a-simple-event">fire a simple
6911: event</a> called <code title="event-DOMContentLoaded">DOMContentLoaded</code> at the
6912: <code>Document</code>. (If the list is empty, this happens
6913: immediately.)</p>
1.1 mike 6914:
1.10 mike 6915: <hr><!-- async --><p>Once everything that <dfn id="delay-the-load-event" title="delay the load event">delays the
1.1 mike 6916: load event</dfn> of the document has completed, the user agent must
6917: run the following steps:</p>
6918:
6919: <ol><li><a href="browsers.html#queue-a-task">Queue a task</a> to set the <a href="dom.html#current-document-readiness">current document
6920: readiness</a> to "complete".</li>
6921:
6922: <li>If the <code>Document</code> is in a <a href="browsers.html#browsing-context">browsing
6923: context</a>, then <a href="browsers.html#queue-a-task">queue a task</a> to <a href="browsers.html#fire-a-simple-event">fire a
6924: simple event</a> called <code title="event-load"><a href="video.html#event-load">load</a></code> at
6925: the <code>Document</code>'s <code><a href="browsers.html#window">Window</a></code> object, but with
6926: its <code title="dom-event-target">target</code> set to the
6927: <code>Document</code> object (and the <code title="dom-event-currentTarget">currentTarget</code> set to the
6928: <code><a href="browsers.html#window">Window</a></code> object).</li>
6929:
6930: <li>If the <code>Document</code> has a <a href="history.html#pending-state-object">pending state
6931: object</a>, then <a href="browsers.html#queue-a-task">queue a task</a> to fire a <code title="event-popstate"><a href="history.html#event-popstate">popstate</a></code> event in no namespace on the
6932: <code>Document</code>'s <code><a href="browsers.html#window">Window</a></code> object using the
6933: <code><a href="history.html#popstateevent">PopStateEvent</a></code> interface, with the <code title="dom-PopStateEvent-state"><a href="history.html#dom-popstateevent-state">state</a></code> attribute set to the
6934: current value of the <a href="history.html#pending-state-object">pending state object</a>. This event
6935: must bubble but not be cancelable and has no default action.</li>
6936:
1.228 mike 6937: </ol><p>The <a href="browsers.html#task-source">task source</a> for these <span title="concept-tasks">tasks</span> is the <a href="browsers.html#dom-manipulation-task-source">DOM manipulation
6938: task source</a>.</p>
1.1 mike 6939:
6940: <!-- XXX make sure things "delay the load event", e.g. applet and
6941: object element's frame contents -->
6942:
6943: <p class="XXX">delaying the load event for things like image
6944: loads allows for intranet port scans (even without
6945: javascript!). Should we really encode that into the spec?</p>
6946:
6947:
6948: <!--XXX need to handle
6949: http://lxr.mozilla.org/mozilla/source/parser/htmlparser/src/CNavDTD.cpp#2354
6950: 2354 // Don't open transient styles if it makes the stack deep, bug 58917.
6951: -->
6952:
6953: <!--XXX
6954: http://lxr.mozilla.org/mozilla/source/parser/htmlparser/src/nsHTMLTokenizer.cpp#749
6955: -->
6956:
6957: <!--
6958: see also CTextToken::ConsumeCharacterData() for CDATA parsing?
6959:
6960: 1212 1 Here's a tricky case from bug 22596: <h5><li><h5>
6961: 1213 How do we know that the 2nd <h5> should close the <LI> rather than nest inside the <LI>?
6962: 1214 (Afterall, the <h5> is a legal child of the <LI>).
6963: 1215
6964: 1216 The way you know is that there is no root between the two, so the <h5> binds more
6965: 1217 tightly to the 1st <h5> than to the <LI>.
6966: 1218 2. Also, bug 6148 shows this case: <SPAN><DIV><SPAN>
6967: 1219 From this case we learned not to execute this logic if the parent is a block.
6968: 1220
6969: 1221 3. Fix for 26583
6970: 1222 Ex. <A href=foo.html><B>foo<A href-bar.html>bar</A></B></A> <- A legal HTML
6971: 1223 In the above example clicking on "foo" or "bar" should link to
6972: 1224 foo.html or bar.html respectively. That is, the inner <A> should be informed
6973: 1225 about the presence of an open <A> above <B>..so that the inner <A> can close out
6974: 1226 the outer <A>. The following code does it for us.
6975: 1227
6976: 1228 4. Fix for 27865 [ similer to 22596 ]. Ex: <DL><DD><LI>one<DD><LI>two
6977: - http://lxr.mozilla.org/mozilla/source/parser/htmlparser/src/CNavDTD.cpp#1211
6978:
6979: 815 // Here's a problem. If theTag is legal in here, we don't move it
6980: 816 // out. So if we're moving stuff out of here, the parent of theTag
6981: 817 // gets closed at this point. But some things are legal
6982: 818 // _everywhere_ and hence would effectively close out misplaced
6983: 819 // content in tables. This is undesirable, so treat them as
6984: 820 // illegal here so they'll be shipped out with their parents and
6985: 821 // siblings. See bug 40855 for an explanation (that bug was for
6986: 822 // comments, but the same issues arise with whitespace, newlines,
6987: 823 // noscript, etc). Script is special, though. Shipping it out
6988: 824 // breaks document.write stuff. See bug 243064.
6989: - http://lxr.mozilla.org/mozilla/source/parser/htmlparser/src/CNavDTD.cpp#825
6990:
6991:
6992: 1326 /**************************************************************************************
6993: 1327 *
6994: 1328 * Now a little code to deal with bug #49687 (crash when layout stack gets too deep)
6995: 1329 * I've also opened this up to any container (not just inlines): re bug 55095
6996: 1330 * Improved to handle bug 55980 (infinite loop caused when DEPTH is exceeded and
6997: 1331 * </P> is encountered by itself (<P>) is continuously produced.
6998: 1332 *
6999: 1333 **************************************************************************************/
7000:
7001: 1912 // Oh boy!! we found a "stray" tag. Nav4.x and IE introduce line break in
7002: 1913 // such cases. So, let's simulate that effect for compatibility.
7003: 1914 // Ex. <html><body>Hello</P>There</body></html>
7004: http://lxr.mozilla.org/mozilla/source/parser/htmlparser/src/CNavDTD.cpp#1912
7005:
7006: http://lxr.mozilla.org/seamonkey/search?string=nested
7007: /parser/htmlparser/src/CNavDTD.cpp, line 791 - * 2. <CENTER><DL><DT><A><CENTER> allow nested <CENTER>
7008: /parser/htmlparser/src/CNavDTD.cpp, line 792 - * 3. <TABLE><TR><TD><TABLE>... allow nested <TABLE>
7009: /parser/htmlparser/src/CNavDTD.cpp, line 2562 - // Discard nested forms - bug 72639
7010: /parser/htmlparser/src/nsElementTable.cpp, line 1453 - * 2. <CENTER><DL><DT><A><CENTER> allow nested <CENTER>
7011: /parser/htmlparser/src/nsElementTable.cpp, line 1454 - * 3. <TABLE><TR><TD><TABLE>... allow nested <TABLE>
7012: /parser/htmlparser/src/nsElementTable.cpp, line 1901 - // Ex: <H1><LI><H1><LI>. Inner LI has the potential of getting nested
7013: -->
7014:
7015:
7016: <h4 id="coercing-an-html-dom-into-an-infoset"><span class="secno">9.2.7 </span>Coercing an HTML DOM into an infoset</h4>
7017:
7018: <p>When an application uses an <a href="#html-parser">HTML parser</a> in
7019: conjunction with an XML pipeline, it is possible that the
7020: constructed DOM is not compatible with the XML tool chain in certain
7021: subtle ways. For example, an XML toolchain might not be able to
7022: represent attributes with the name <code title="">xmlns</code>,
7023: since they conflict with the Namespaces in XML syntax. There is also
7024: some data that the <a href="#html-parser">HTML parser</a> generates that isn't
7025: included in the DOM itself. This section specifies some rules for
7026: handling these issues.</p>
7027:
7028: <p>If the XML API being used doesn't support DOCTYPEs, the tool may
7029: drop DOCTYPEs altogether.</p>
7030:
7031: <p>If the XML API doesn't support attributes in no namespace that
7032: are named "<code title="">xmlns</code>", attributes whose names
7033: start with "<code title="">xmlns:</code>", or attributes in the
7034: <a href="#xmlns-namespace">XMLNS namespace</a>, then the tool may drop such
7035: attributes.</p>
7036:
7037: <p>The tool may annotate the output with any namespace declarations
7038: required for proper operation.</p>
7039:
7040: <p>If the XML API being used restricts the allowable characters in
7041: the local names of elements and attributes, then the tool may map
7042: all element and attribute local names that the API wouldn't support
7043: to a set of names that <em>are</em> allowed, by replacing any
7044: character that isn't supported with the uppercase letter U and the
7045: six digits of the character's Unicode code point when expressed in
7046: hexadecimal, using digits 0-9 and capital letters A-F as the
7047: symbols, in increasing numeric order.</p>
7048:
7049: <p class="example">For example, the element name <code title="">foo<bar</code>, which can be output by the <a href="#html-parser">HTML
7050: parser</a>, though it is neither a legal HTML element name nor a
7051: well-formed XML element name, would be converted into <code title="">fooU00003Cbar</code>, which <em>is</em> a well-formed XML
7052: element name (though it's still not legal in HTML by any means).</p>
7053:
7054: <p class="example">As another example, consider the attribute
7055: <code>xlink:href</code>. Used on a MathML element, it becomes, after
7056: being <a href="#adjust-foreign-attributes" title="adjust foreign attributes">adjusted</a>, an
7057: attribute with a prefix "<code title="">xlink</code>" and a local
7058: name "<code title="">href</code>". However, used on an HTML element,
7059: it becomes an attribute with no prefix and the local name "<code title="">xlink:href</code>", which is not a valid NCName, and thus
7060: might not be accepted by an XML API. It could thus get converted,
7061: becoming "<code title="">xlinkU00003Ahref</code>".</p>
7062:
7063: <p class="note">The resulting names from this conversion
7064: conveniently can't clash with any attribute generated by the
7065: <a href="#html-parser">HTML parser</a>, since those are all either lowercase or
7066: those listed in the <a href="#adjust-foreign-attributes">adjust foreign attributes</a>
7067: algorithm's table.</p>
7068:
7069: <p>If the XML API restricts comments from having two consecutive
7070: U+002D HYPHEN-MINUS characters (--), the tool may insert a single
7071: U+0020 SPACE character between any such offending characters.</p>
7072:
7073: <p>If the XML API restricts comments from ending in a
7074: U+002D HYPHEN-MINUS character (-), the tool may insert a single
7075: U+0020 SPACE character at the end of such comments.</p>
7076:
7077: <p>If the XML API restricts allowed characters in character data,
7078: the tool may replace any U+000C FORM FEED (FF) character with a
7079: U+0020 SPACE character, and any other literal non-XML character with
7080: a U+FFFD REPLACEMENT CHARACTER.</p>
7081:
7082: <p>If the tool has no way to convey out-of-band information, then
7083: the tool may drop the following information:</p>
7084:
7085: <ul><li>Whether the document is set to <i><a href="dom.html#no-quirks-mode">no quirks mode</a></i>,
7086: <i><a href="dom.html#limited-quirks-mode">limited quirks mode</a></i>, or <i><a href="dom.html#quirks-mode">quirks mode</a></i></li>
7087:
7088: <li>The association between form controls and forms that aren't
7089: their nearest <code><a href="forms.html#the-form-element">form</a></code> element ancestor (use of the
7090: <a href="#form-element-pointer"><code>form</code> element pointer</a> in the parser)</li>
7091:
7092: </ul><p class="note">The mutations allowed by this section apply
7093: <em>after</em> the <a href="#html-parser">HTML parser</a>'s rules have been
7094: applied. For example, a <code title=""><a::></code> start tag
7095: will be closed by a <code title=""></a::></code> end tag, and
7096: never by a <code title=""></aU00003AU00003A></code> end tag, even
7097: if the user agent is using the rules above to then generate an
7098: actual element in the DOM with the name <code title="">aU00003AU00003A</code> for that start tag.</p>
7099:
7100:
7101:
1.35 mike 7102: <h4 id="an-introduction-to-error-handling-and-strange-cases-in-the-parser"><span class="secno">9.2.8 </span>An introduction to error handling and strange cases in the parser</h4>
1.1 mike 7103:
7104: <p><i>This section is non-normative.</i></p>
7105:
7106: <p>This section examines some erroneous markup and discusses how
7107: the <a href="#html-parser">HTML parser</a> handles these cases.</p>
7108:
7109:
7110: <h5 id="misnested-tags:-b-i-b-i"><span class="secno">9.2.8.1 </span>Misnested tags: <b><i></b></i></h5>
7111:
7112: <p><i>This section is non-normative.</i></p>
7113:
7114: <p>The most-often discussed example of erroneous markup is as
7115: follows:</p>
7116:
7117: <pre><p>1<b>2<i>3</b>4</i>5</p></pre>
7118:
7119: <p>The parsing of this markup is straightforward up to the "3". At
7120: this point, the DOM looks like this:</p>
7121:
1.159 mike 7122: <ul class="domTree"><li class="t1"><code><a href="semantics.html#the-html-element-0">html</a></code><ul><li class="t1"><code><a href="semantics.html#the-head-element-0">head</a></code></li><li class="t1"><code><a href="semantics.html#the-body-element-0">body</a></code><ul><li class="t1"><code><a href="semantics.html#the-p-element">p</a></code><ul><li class="t3"><code>#text</code>: <span title="">1</span></li><li class="t1"><code><a href="text-level-semantics.html#the-b-element">b</a></code><ul><li class="t3"><code>#text</code>: <span title="">2</span></li><li class="t1"><code><a href="text-level-semantics.html#the-i-element">i</a></code><ul><li class="t3"><code>#text</code>: <span title="">3</span></li></ul></li></ul></li></ul></li></ul></li></ul></li></ul><p>Here, the <a href="#stack-of-open-elements">stack of open elements</a> has five elements
7123: on it: <code><a href="semantics.html#the-html-element-0">html</a></code>, <code><a href="semantics.html#the-body-element-0">body</a></code>, <code><a href="semantics.html#the-p-element">p</a></code>,
1.1 mike 7124: <code><a href="text-level-semantics.html#the-b-element">b</a></code>, and <code><a href="text-level-semantics.html#the-i-element">i</a></code>. The <a href="#list-of-active-formatting-elements">list of active
7125: formatting elements</a> just has two: <code><a href="text-level-semantics.html#the-b-element">b</a></code> and
7126: <code><a href="text-level-semantics.html#the-i-element">i</a></code>. The <a href="#insertion-mode">insertion mode</a> is "<a href="#parsing-main-inbody" title="insertion mode: in body">in body</a>".</p>
7127:
7128: <p>Upon receiving the end tag token with the tag name "b", the "<a href="#adoptionAgency">adoption agency algorithm</a>" is
7129: invoked. This is a simple case, in that the <var title="">formatting
7130: element</var> is the <code><a href="text-level-semantics.html#the-b-element">b</a></code> element, and there is no
7131: <var title="">furthest block</var>. Thus, the <a href="#stack-of-open-elements">stack of open
1.159 mike 7132: elements</a> ends up with just three elements: <code><a href="semantics.html#the-html-element-0">html</a></code>,
7133: <code><a href="semantics.html#the-body-element-0">body</a></code>, and <code><a href="semantics.html#the-p-element">p</a></code>, while the <a href="#list-of-active-formatting-elements">list of
1.1 mike 7134: active formatting elements</a> has just one: <code><a href="text-level-semantics.html#the-i-element">i</a></code>. The
7135: DOM tree is unmodified at this point.</p>
7136:
7137: <p>The next token is a character ("4"), triggers the <a href="#reconstruct-the-active-formatting-elements" title="reconstruct the active formatting elements">reconstruction of
7138: the active formatting elements</a>, in this case just the
7139: <code><a href="text-level-semantics.html#the-i-element">i</a></code> element. A new <code><a href="text-level-semantics.html#the-i-element">i</a></code> element is thus created
7140: for the "4" text node. After the end tag token for the "i" is also
7141: received, and the "5" text node is inserted, the DOM looks as
7142: follows:</p>
7143:
1.159 mike 7144: <ul class="domTree"><li class="t1"><code><a href="semantics.html#the-html-element-0">html</a></code><ul><li class="t1"><code><a href="semantics.html#the-head-element-0">head</a></code></li><li class="t1"><code><a href="semantics.html#the-body-element-0">body</a></code><ul><li class="t1"><code><a href="semantics.html#the-p-element">p</a></code><ul><li class="t3"><code>#text</code>: <span title="">1</span></li><li class="t1"><code><a href="text-level-semantics.html#the-b-element">b</a></code><ul><li class="t3"><code>#text</code>: <span title="">2</span></li><li class="t1"><code><a href="text-level-semantics.html#the-i-element">i</a></code><ul><li class="t3"><code>#text</code>: <span title="">3</span></li></ul></li></ul></li><li class="t1"><code><a href="text-level-semantics.html#the-i-element">i</a></code><ul><li class="t3"><code>#text</code>: <span title="">4</span></li></ul></li><li class="t3"><code>#text</code>: <span title="">5</span></li></ul></li></ul></li></ul></li></ul><h5 id="misnested-tags:-b-p-b-p"><span class="secno">9.2.8.2 </span>Misnested tags: <b><p></b></p></h5>
1.1 mike 7145:
7146: <p><i>This section is non-normative.</i></p>
7147:
7148: <p>A case similar to the previous one is the following:</p>
7149:
7150: <pre><b>1<p>2</b>3</p></pre>
7151:
7152: <p>Up to the "2" the parsing here is straightforward:</p>
7153:
1.159 mike 7154: <ul class="domTree"><li class="t1"><code><a href="semantics.html#the-html-element-0">html</a></code><ul><li class="t1"><code><a href="semantics.html#the-head-element-0">head</a></code></li><li class="t1"><code><a href="semantics.html#the-body-element-0">body</a></code><ul><li class="t1"><code><a href="text-level-semantics.html#the-b-element">b</a></code><ul><li class="t3"><code>#text</code>: <span title="">1</span></li><li class="t1"><code><a href="semantics.html#the-p-element">p</a></code><ul><li class="t3"><code>#text</code>: <span title="">2</span></li></ul></li></ul></li></ul></li></ul></li></ul><p>The interesting part is when the end tag token with the tag name
1.1 mike 7155: "b" is parsed.</p>
7156:
7157: <p>Before that token is seen, the <a href="#stack-of-open-elements">stack of open
1.159 mike 7158: elements</a> has four elements on it: <code><a href="semantics.html#the-html-element-0">html</a></code>,
7159: <code><a href="semantics.html#the-body-element-0">body</a></code>, <code><a href="text-level-semantics.html#the-b-element">b</a></code>, and <code><a href="semantics.html#the-p-element">p</a></code>. The
1.1 mike 7160: <a href="#list-of-active-formatting-elements">list of active formatting elements</a> just has the one:
7161: <code><a href="text-level-semantics.html#the-b-element">b</a></code>. The <a href="#insertion-mode">insertion mode</a> is "<a href="#parsing-main-inbody" title="insertion mode: in body">in body</a>".</p>
7162:
7163: <p>Upon receiving the end tag token with the tag name "b", the "<a href="#adoptionAgency">adoption agency algorithm</a>" is invoked, as
7164: in the previous example. However, in this case, there <em>is</em> a
7165: <var title="">furthest block</var>, namely the <code><a href="semantics.html#the-p-element">p</a></code> element. Thus,
7166: this time the adoption agency algorithm isn't skipped over.</p>
7167:
1.159 mike 7168: <p>The <var title="">common ancestor</var> is the <code><a href="semantics.html#the-body-element-0">body</a></code>
1.1 mike 7169: element. A conceptual "bookmark" marks the position of the
7170: <code><a href="text-level-semantics.html#the-b-element">b</a></code> in the <a href="#list-of-active-formatting-elements">list of active formatting
7171: elements</a>, but since that list has only one element in it,
7172: it won't have much effect.</p>
7173:
7174: <p>As the algorithm progresses, <var title="">node</var> ends up set
7175: to the formatting element (<code><a href="text-level-semantics.html#the-b-element">b</a></code>), and <var title="">last
7176: node</var> ends up set to the <var title="">furthest block</var>
7177: (<code><a href="semantics.html#the-p-element">p</a></code>).</p>
7178:
7179: <p>The <var title="">last node</var> gets appended (moved) to the
7180: <var title="">common ancestor</var>, so that the DOM looks like:</p>
7181:
1.159 mike 7182: <ul class="domTree"><li class="t1"><code><a href="semantics.html#the-html-element-0">html</a></code><ul><li class="t1"><code><a href="semantics.html#the-head-element-0">head</a></code></li><li class="t1"><code><a href="semantics.html#the-body-element-0">body</a></code><ul><li class="t1"><code><a href="text-level-semantics.html#the-b-element">b</a></code><ul><li class="t3"><code>#text</code>: <span title="">1</span></li></ul></li><li class="t1"><code><a href="semantics.html#the-p-element">p</a></code><ul><li class="t3"><code>#text</code>: <span title="">2</span></li></ul></li></ul></li></ul></li></ul><p>A new <code><a href="text-level-semantics.html#the-b-element">b</a></code> element is created, and the children of the
1.1 mike 7183: <code><a href="semantics.html#the-p-element">p</a></code> element are moved to it:</p>
7184:
1.159 mike 7185: <ul class="domTree"><li class="t1"><code><a href="semantics.html#the-html-element-0">html</a></code><ul><li class="t1"><code><a href="semantics.html#the-head-element-0">head</a></code></li><li class="t1"><code><a href="semantics.html#the-body-element-0">body</a></code><ul><li class="t1"><code><a href="text-level-semantics.html#the-b-element">b</a></code><ul><li class="t3"><code>#text</code>: <span title="">1</span></li></ul></li><li class="t1"><code><a href="semantics.html#the-p-element">p</a></code></li></ul></li></ul></li></ul><ul class="domTree"><li class="t1"><code><a href="text-level-semantics.html#the-b-element">b</a></code><ul><li class="t3"><code>#text</code>: <span title="">2</span></li></ul></li></ul><p>Finally, the new <code><a href="text-level-semantics.html#the-b-element">b</a></code> element is appended to the
1.1 mike 7186: <code><a href="semantics.html#the-p-element">p</a></code> element, so that the DOM looks like:</p>
7187:
1.159 mike 7188: <ul class="domTree"><li class="t1"><code><a href="semantics.html#the-html-element-0">html</a></code><ul><li class="t1"><code><a href="semantics.html#the-head-element-0">head</a></code></li><li class="t1"><code><a href="semantics.html#the-body-element-0">body</a></code><ul><li class="t1"><code><a href="text-level-semantics.html#the-b-element">b</a></code><ul><li class="t3"><code>#text</code>: <span title="">1</span></li></ul></li><li class="t1"><code><a href="semantics.html#the-p-element">p</a></code><ul><li class="t1"><code><a href="text-level-semantics.html#the-b-element">b</a></code><ul><li class="t3"><code>#text</code>: <span title="">2</span></li></ul></li></ul></li></ul></li></ul></li></ul><p>The <code><a href="text-level-semantics.html#the-b-element">b</a></code> element is removed from the <a href="#list-of-active-formatting-elements">list of
1.1 mike 7189: active formatting elements</a> and the <a href="#stack-of-open-elements">stack of open
7190: elements</a>, so that when the "3" is parsed, it is appended to
7191: the <code><a href="semantics.html#the-p-element">p</a></code> element:</p>
7192:
1.159 mike 7193: <ul class="domTree"><li class="t1"><code><a href="semantics.html#the-html-element-0">html</a></code><ul><li class="t1"><code><a href="semantics.html#the-head-element-0">head</a></code></li><li class="t1"><code><a href="semantics.html#the-body-element-0">body</a></code><ul><li class="t1"><code><a href="text-level-semantics.html#the-b-element">b</a></code><ul><li class="t3"><code>#text</code>: <span title="">1</span></li></ul></li><li class="t1"><code><a href="semantics.html#the-p-element">p</a></code><ul><li class="t1"><code><a href="text-level-semantics.html#the-b-element">b</a></code><ul><li class="t3"><code>#text</code>: <span title="">2</span></li></ul></li><li class="t3"><code>#text</code>: <span title="">3</span></li></ul></li></ul></li></ul></li></ul><h5 id="unexpected-markup-in-tables"><span class="secno">9.2.8.3 </span>Unexpected markup in tables</h5>
1.1 mike 7194:
7195: <p><i>This section is non-normative.</i></p>
7196:
7197: <p>Error handling in tables is, for historical reasons, especially
7198: strange. For example, consider the following markup:</p>
7199:
7200: <pre><table><strong><b></strong><tr><td>aaa</td></tr><strong>bbb</strong></table>ccc</pre>
7201:
7202: <p>The highlighted <code><a href="text-level-semantics.html#the-b-element">b</a></code> element start tag is not allowed
7203: directly inside a table like that, and the parser handles this case
7204: by placing the element <em>before</em> the table. (This is called <i title="foster parent"><a href="#foster-parent">foster parenting</a></i>.) This can be seen by
7205: examining the DOM tree as it stands just after the
7206: <code><a href="tabular-data.html#the-table-element">table</a></code> element's start tag has been seen:</p>
7207:
1.159 mike 7208: <ul class="domTree"><li class="t1"><code><a href="semantics.html#the-html-element-0">html</a></code><ul><li class="t1"><code><a href="semantics.html#the-head-element-0">head</a></code></li><li class="t1"><code><a href="semantics.html#the-body-element-0">body</a></code><ul><li class="t1"><code><a href="tabular-data.html#the-table-element">table</a></code></li></ul></li></ul></li></ul><p>...and then immediately after the <code><a href="text-level-semantics.html#the-b-element">b</a></code> element start
1.1 mike 7209: tag has been seen:</p>
7210:
1.159 mike 7211: <ul class="domTree"><li class="t1"><code><a href="semantics.html#the-html-element-0">html</a></code><ul><li class="t1"><code><a href="semantics.html#the-head-element-0">head</a></code></li><li class="t1"><code><a href="semantics.html#the-body-element-0">body</a></code><ul><li class="t1"><code><a href="text-level-semantics.html#the-b-element">b</a></code></li><li class="t1"><code><a href="tabular-data.html#the-table-element">table</a></code></li></ul></li></ul></li></ul><p>At this point, the <a href="#stack-of-open-elements">stack of open elements</a> has on it
7212: the elements <code><a href="semantics.html#the-html-element-0">html</a></code>, <code><a href="semantics.html#the-body-element-0">body</a></code>,
1.1 mike 7213: <code><a href="tabular-data.html#the-table-element">table</a></code>, and <code><a href="text-level-semantics.html#the-b-element">b</a></code> (in that order, despite the
7214: resulting DOM tree); the <a href="#list-of-active-formatting-elements">list of active formatting
7215: elements</a> just has the <code><a href="text-level-semantics.html#the-b-element">b</a></code> element in it; and the
7216: <a href="#insertion-mode">insertion mode</a> is "<a href="#parsing-main-intable" title="insertion mode: in
7217: table">in table</a>".</p>
7218:
7219: <p>The <code><a href="tabular-data.html#the-tr-element">tr</a></code> start tag causes the <code><a href="text-level-semantics.html#the-b-element">b</a></code> element
7220: to be popped off the stack and a <code><a href="tabular-data.html#the-tbody-element">tbody</a></code> start tag to be
7221: implied; the <code><a href="tabular-data.html#the-tbody-element">tbody</a></code> and <code><a href="tabular-data.html#the-tr-element">tr</a></code> elements are
7222: then handled in a rather straight-forward manner, taking the parser
7223: through the "<a href="#parsing-main-intbody" title="insertion mode: in table body">in table
7224: body</a>" and "<a href="#parsing-main-intr" title="insertion mode: in row">in
7225: row</a>" insertion modes, after which the DOM looks as
7226: follows:</p>
7227:
1.159 mike 7228: <ul class="domTree"><li class="t1"><code><a href="semantics.html#the-html-element-0">html</a></code><ul><li class="t1"><code><a href="semantics.html#the-head-element-0">head</a></code></li><li class="t1"><code><a href="semantics.html#the-body-element-0">body</a></code><ul><li class="t1"><code><a href="text-level-semantics.html#the-b-element">b</a></code></li><li class="t1"><code><a href="tabular-data.html#the-table-element">table</a></code><ul><li class="t1"><code><a href="tabular-data.html#the-tbody-element">tbody</a></code><ul><li class="t1"><code><a href="tabular-data.html#the-tr-element">tr</a></code></li></ul></li></ul></li></ul></li></ul></li></ul><p>Here, the <a href="#stack-of-open-elements">stack of open elements</a> has on it the
7229: elements <code><a href="semantics.html#the-html-element-0">html</a></code>, <code><a href="semantics.html#the-body-element-0">body</a></code>, <code><a href="tabular-data.html#the-table-element">table</a></code>,
1.1 mike 7230: <code><a href="tabular-data.html#the-tbody-element">tbody</a></code>, and <code><a href="tabular-data.html#the-tr-element">tr</a></code>; the <a href="#list-of-active-formatting-elements">list of active
7231: formatting elements</a> still has the <code><a href="text-level-semantics.html#the-b-element">b</a></code> element in
7232: it; and the <a href="#insertion-mode">insertion mode</a> is "<a href="#parsing-main-intr" title="insertion
7233: mode: in row">in row</a>".</p>
7234:
7235: <p>The <code><a href="tabular-data.html#the-td-element">td</a></code> element start tag token, after putting a
7236: <code><a href="tabular-data.html#the-td-element">td</a></code> element on the tree, puts a marker on the <a href="#list-of-active-formatting-elements">list
7237: of active formatting elements</a> (it also switches to the "<a href="#parsing-main-intd" title="insertion mode: in cell">in cell</a>" <a href="#insertion-mode">insertion
7238: mode</a>).</p>
7239:
1.159 mike 7240: <ul class="domTree"><li class="t1"><code><a href="semantics.html#the-html-element-0">html</a></code><ul><li class="t1"><code><a href="semantics.html#the-head-element-0">head</a></code></li><li class="t1"><code><a href="semantics.html#the-body-element-0">body</a></code><ul><li class="t1"><code><a href="text-level-semantics.html#the-b-element">b</a></code></li><li class="t1"><code><a href="tabular-data.html#the-table-element">table</a></code><ul><li class="t1"><code><a href="tabular-data.html#the-tbody-element">tbody</a></code><ul><li class="t1"><code><a href="tabular-data.html#the-tr-element">tr</a></code><ul><li class="t1"><code><a href="tabular-data.html#the-td-element">td</a></code></li></ul></li></ul></li></ul></li></ul></li></ul></li></ul><p>The marker means that when the "aaa" character tokens are seen,
1.1 mike 7241: no <code><a href="text-level-semantics.html#the-b-element">b</a></code> element is created to hold the resulting text
7242: node:</p>
7243:
1.159 mike 7244: <ul class="domTree"><li class="t1"><code><a href="semantics.html#the-html-element-0">html</a></code><ul><li class="t1"><code><a href="semantics.html#the-head-element-0">head</a></code></li><li class="t1"><code><a href="semantics.html#the-body-element-0">body</a></code><ul><li class="t1"><code><a href="text-level-semantics.html#the-b-element">b</a></code></li><li class="t1"><code><a href="tabular-data.html#the-table-element">table</a></code><ul><li class="t1"><code><a href="tabular-data.html#the-tbody-element">tbody</a></code><ul><li class="t1"><code><a href="tabular-data.html#the-tr-element">tr</a></code><ul><li class="t1"><code><a href="tabular-data.html#the-td-element">td</a></code><ul><li class="t3"><code>#text</code>: <span title="">aaa</span></li></ul></li></ul></li></ul></li></ul></li></ul></li></ul></li></ul><p>The end tags are handled in a straight-forward manner; after
1.1 mike 7245: handling them, the <a href="#stack-of-open-elements">stack of open elements</a> has on it the
1.159 mike 7246: elements <code><a href="semantics.html#the-html-element-0">html</a></code>, <code><a href="semantics.html#the-body-element-0">body</a></code>, <code><a href="tabular-data.html#the-table-element">table</a></code>,
1.1 mike 7247: and <code><a href="tabular-data.html#the-tbody-element">tbody</a></code>; the <a href="#list-of-active-formatting-elements">list of active formatting
7248: elements</a> still has the <code><a href="text-level-semantics.html#the-b-element">b</a></code> element in it (the
7249: marker having been removed by the "td" end tag token); and the
7250: <a href="#insertion-mode">insertion mode</a> is "<a href="#parsing-main-intbody" title="insertion mode: in
7251: table body">in table body</a>".</p>
7252:
7253: <p>Thus it is that the "bbb" character tokens are found. These
7254: trigger the "<a href="#parsing-main-intabletext" title="insertion mode: in table text">in table
7255: text</a>" insertion mode to be used (with the <a href="#original-insertion-mode">original
7256: insertion mode</a> set to "<a href="#parsing-main-intbody" title="insertion mode: in table
7257: body">in table body</a>"). The character tokens are collected,
7258: and when the next token (the <code><a href="tabular-data.html#the-table-element">table</a></code> element end tag) is
7259: seen, they are processed as a group. Since they are not all spaces,
7260: they are handled as per the "anything else" rules in the "<a href="#parsing-main-intable" title="insertion mode: in table">in table</a>" insertion mode,
7261: which defer to the "<a href="#parsing-main-inbody" title="insertion mode: in body">in
7262: body</a>" insertion mode but with <a href="#foster-parent" title="foster
7263: parent">foster parenting</a>.</p>
7264:
7265: <p>When <a href="#reconstruct-the-active-formatting-elements" title="reconstruct the active formatting elements">the
7266: active formatting elements are reconstructed</a>, a
7267: <code><a href="text-level-semantics.html#the-b-element">b</a></code> element is created and <a href="#foster-parent" title="foster
7268: parent">foster parented</a>, and then the "bbb" text node is
7269: appended to it:</p>
7270:
1.159 mike 7271: <ul class="domTree"><li class="t1"><code><a href="semantics.html#the-html-element-0">html</a></code><ul><li class="t1"><code><a href="semantics.html#the-head-element-0">head</a></code></li><li class="t1"><code><a href="semantics.html#the-body-element-0">body</a></code><ul><li class="t1"><code><a href="text-level-semantics.html#the-b-element">b</a></code></li><li class="t1"><code><a href="text-level-semantics.html#the-b-element">b</a></code><ul><li class="t3"><code>#text</code>: <span title="">bbb</span></li></ul></li><li class="t1"><code><a href="tabular-data.html#the-table-element">table</a></code><ul><li class="t1"><code><a href="tabular-data.html#the-tbody-element">tbody</a></code><ul><li class="t1"><code><a href="tabular-data.html#the-tr-element">tr</a></code><ul><li class="t1"><code><a href="tabular-data.html#the-td-element">td</a></code><ul><li class="t3"><code>#text</code>: <span title="">aaa</span></li></ul></li></ul></li></ul></li></ul></li></ul></li></ul></li></ul><p>The <a href="#stack-of-open-elements">stack of open elements</a> has on it the elements
7272: <code><a href="semantics.html#the-html-element-0">html</a></code>, <code><a href="semantics.html#the-body-element-0">body</a></code>, <code><a href="tabular-data.html#the-table-element">table</a></code>,
1.1 mike 7273: <code><a href="tabular-data.html#the-tbody-element">tbody</a></code>, and the new <code><a href="text-level-semantics.html#the-b-element">b</a></code> (again, note that
7274: this doesn't match the resulting tree!); the <a href="#list-of-active-formatting-elements">list of active
7275: formatting elements</a> has the new <code><a href="text-level-semantics.html#the-b-element">b</a></code> element in it;
7276: and the <a href="#insertion-mode">insertion mode</a> is still "<a href="#parsing-main-intbody" title="insertion
7277: mode: in table body">in table body</a>".</p>
7278:
7279: <p>Had the character tokens been only <a href="infrastructure.html#space-character" title="space
7280: character">space characters</a> instead of "bbb", then those
7281: <a href="infrastructure.html#space-character" title="space character">space characters</a> would just be
7282: appended to the <code><a href="tabular-data.html#the-tbody-element">tbody</a></code> element.</p>
7283:
7284: <p>Finally, the <code><a href="tabular-data.html#the-table-element">table</a></code> is closed by a "table" end
7285: tag. This pops all the nodes from the <a href="#stack-of-open-elements">stack of open
7286: elements</a> up to and including the <code><a href="tabular-data.html#the-table-element">table</a></code> element,
7287: but it doesn't affect the <a href="#list-of-active-formatting-elements">list of active formatting
7288: elements</a>, so the "ccc" character tokens after the table
7289: result in yet another <code><a href="text-level-semantics.html#the-b-element">b</a></code> element being created, this
7290: time after the table:</p>
7291:
1.159 mike 7292: <ul class="domTree"><li class="t1"><code><a href="semantics.html#the-html-element-0">html</a></code><ul><li class="t1"><code><a href="semantics.html#the-head-element-0">head</a></code></li><li class="t1"><code><a href="semantics.html#the-body-element-0">body</a></code><ul><li class="t1"><code><a href="text-level-semantics.html#the-b-element">b</a></code></li><li class="t1"><code><a href="text-level-semantics.html#the-b-element">b</a></code><ul><li class="t3"><code>#text</code>: <span title="">bbb</span></li></ul></li><li class="t1"><code><a href="tabular-data.html#the-table-element">table</a></code><ul><li class="t1"><code><a href="tabular-data.html#the-tbody-element">tbody</a></code><ul><li class="t1"><code><a href="tabular-data.html#the-tr-element">tr</a></code><ul><li class="t1"><code><a href="tabular-data.html#the-td-element">td</a></code><ul><li class="t3"><code>#text</code>: <span title="">aaa</span></li></ul></li></ul></li></ul></li></ul></li><li class="t1"><code><a href="text-level-semantics.html#the-b-element">b</a></code><ul><li class="t3"><code>#text</code>: <span title="">ccc</span></li></ul></li></ul></li></ul></li></ul><h5 id="scripts-that-modify-the-page-as-it-is-being-parsed"><span class="secno">9.2.8.4 </span>Scripts that modify the page as it is being parsed</h5>
1.1 mike 7293:
7294: <p><i>This section is non-normative.</i></p>
7295:
7296: <p>Consider the following markup, which for this example we will
7297: assume is the document with <a href="infrastructure.html#url">URL</a> <code title="">http://example.com/inner</code>, being rendered as the
1.159 mike 7298: content of an <code><a href="text-level-semantics.html#the-iframe-element">iframe</a></code> in another document with the
1.1 mike 7299: <a href="infrastructure.html#url">URL</a> <code title="">http://example.com/outer</code>:</p>
7300:
7301: <pre><div id=a>
7302: <script>
7303: var div = document.getElementById('a');
7304: parent.document.body.appendChild(div);
7305: </script>
7306: <script>
7307: alert(document.URL);
7308: </script>
7309: </div>
7310: <script>
7311: alert(document.URL);
7312: </script></pre>
7313:
7314: <p>Up to the first "script" end tag, before the script is parsed,
7315: the result is relatively straightforward:</p>
7316:
1.159 mike 7317: <ul class="domTree"><li class="t1"><code><a href="semantics.html#the-html-element-0">html</a></code><ul><li class="t1"><code><a href="semantics.html#the-head-element-0">head</a></code></li><li class="t1"><code><a href="semantics.html#the-body-element-0">body</a></code><ul><li class="t1"><code><a href="interactive-elements.html#the-div-element">div</a></code> <span class="t2" title=""><code class="attribute name">id</code>="<code class="attribute value"><a href="text-level-semantics.html#the-a-element">a</a></code>"</span><ul><li class="t3"><code>#text</code>: <span title="">
1.1 mike 7318: </span></li><li class="t1"><code><a href="semantics.html#script">script</a></code><ul><li class="t3"><code>#text</code>: <span title="">var div = document.getElementById('a'); ⏎ parent.document.body.appendChild(div);</span></li></ul></li></ul></li></ul></li></ul></li></ul><p>After the script is parsed, though, the <code><a href="interactive-elements.html#the-div-element">div</a></code> element
7319: and its child <code><a href="semantics.html#script">script</a></code> element are gone:</p>
7320:
1.159 mike 7321: <ul class="domTree"><li class="t1"><code><a href="semantics.html#the-html-element-0">html</a></code><ul><li class="t1"><code><a href="semantics.html#the-head-element-0">head</a></code></li><li class="t1"><code><a href="semantics.html#the-body-element-0">body</a></code></li></ul></li></ul><p>They are, at this point, in the <code>Document</code> of the
1.1 mike 7322: aforementioned outer <a href="browsers.html#browsing-context">browsing context</a>. However, the
7323: <a href="#stack-of-open-elements">stack of open elements</a> <em>still contains the
7324: <code><a href="interactive-elements.html#the-div-element">div</a></code> element</em>.</p>
7325:
7326: <p>Thus, when the second <code><a href="semantics.html#script">script</a></code> element is parsed, it
7327: is inserted <em>into the outer <code>Document</code>
7328: object</em>.</p>
7329:
7330: <p>This also means that the <a href="browsers.html#script-s-global-object">script's global object</a> is
7331: the outer <a href="browsers.html#browsing-context">browsing context</a>'s <code><a href="browsers.html#window">Window</a></code>
7332: object, <em>not</em> the <code><a href="browsers.html#window">Window</a></code> object inside the
1.159 mike 7333: <code><a href="text-level-semantics.html#the-iframe-element">iframe</a></code>.</p>
1.1 mike 7334:
7335: <p class="note">This isn't a security problem since the script that
7336: moves the <code><a href="interactive-elements.html#the-div-element">div</a></code> into the outer <code>Document</code> can
7337: only do so because they have the two <code>Document</code> object
1.159 mike 7338: have the same <a href="browsers.html#origin">origin</a>.</p>
1.1 mike 7339:
7340: <p>Thus, the first alert says "http://example.com/outer".</p>
7341:
7342: <p>Once the <code><a href="interactive-elements.html#the-div-element">div</a></code> element's end tag is parsed, the
7343: <code><a href="interactive-elements.html#the-div-element">div</a></code> element is popped off the stack, and so the next
7344: <code><a href="semantics.html#script">script</a></code> element is in the inner <code>Document</code>:</p>
7345:
1.159 mike 7346: <ul class="domTree"><li class="t1"><code><a href="semantics.html#the-html-element-0">html</a></code><ul><li class="t1"><code><a href="semantics.html#the-head-element-0">head</a></code></li><li class="t1"><code><a href="semantics.html#the-body-element-0">body</a></code><ul><li class="t1"><code><a href="semantics.html#script">script</a></code><ul><li class="t3"><code>#text</code>: <span title="">alert(document.URL);</span></li></ul></li></ul></li></ul></li></ul><p>This second alert will say "http://example.com/inner".</p>
1.1 mike 7347:
7348:
7349:
1.215 mike 7350: <h3 id="namespaces"><span class="secno">9.3 </span>Namespaces</h3>
1.1 mike 7351:
7352: <p>The <dfn id="html-namespace-0">HTML namespace</dfn> is: <code>http://www.w3.org/1999/xhtml</code></p>
7353:
7354: <p>The <dfn id="mathml-namespace">MathML namespace</dfn> is: <code>http://www.w3.org/1998/Math/MathML</code></p>
7355:
7356: <p>The <dfn id="svg-namespace">SVG namespace</dfn> is: <code>http://www.w3.org/2000/svg</code></p>
7357:
7358: <p>The <dfn id="xlink-namespace">XLink namespace</dfn> is: <code>http://www.w3.org/1999/xlink</code></p>
7359:
7360: <p>The <dfn id="xml-namespace">XML namespace</dfn> is: <code>http://www.w3.org/XML/1998/namespace</code></p>
7361:
7362: <p>The <dfn id="xmlns-namespace">XMLNS namespace</dfn> is: <code>http://www.w3.org/2000/xmlns/</code></p>
7363:
7364: <hr><p>Data mining tools and other user agents that perform operations
1.83 mike 7365: on <code><a href="iana-considerations.html#text-html">text/html</a></code> content without running scripts,
1.1 mike 7366: evaluating CSS or XPath expressions, or otherwise exposing the
7367: resulting DOM to arbitrary content, may "support namespaces" by just
7368: asserting that their DOM node analogues are in certain namespaces,
7369: without actually exposing the above strings.</p>
7370:
7371:
7372:
7373:
7374: <h3 id="serializing-html-fragments"><span class="secno">9.4 </span>Serializing HTML fragments</h3>
7375:
7376: <p>The following steps form the <dfn id="html-fragment-serialization-algorithm">HTML fragment serialization
7377: algorithm</dfn>. The algorithm takes as input a DOM
7378: <code>Element</code> or <code>Document</code>, referred to as <var title="">the node</var>, and either returns a string or raises an
7379: exception.</p>
7380:
7381: <p class="note">This algorithm serializes the <em>children</em> of
7382: the node being serialized, not the node itself.</p>
7383:
7384: <ol><li><p>Let <var title="">s</var> be a string, and initialize it to
7385: the empty string.</p></li>
7386:
7387: <li>
7388:
7389: <p>For each child node of <var title="">the node</var>, in
7390: <a href="infrastructure.html#tree-order">tree order</a>, run the following steps:
7391:
7392: </p><ol><li><p>Let <var title="">current node</var> be the child node
7393: being processed.</p></li>
7394:
7395: <li>
7396:
7397: <p>Append the appropriate string from the following list to
7398: <var title="">s</var>:</p>
7399:
7400: <dl class="switch"><dt>If <var title="">current node</var> is an <code title="">Element</code></dt>
7401:
7402: <dd>
7403:
7404: <p>Append a U+003C LESS-THAN SIGN (<code title=""><</code>)
7405: character, followed by the element's tag name. (For nodes
7406: created by the <a href="#html-parser">HTML parser</a> or <code title="">Document.createElement()</code>, the tag name will be
7407: lowercase.)</p>
7408:
7409: <p>For each attribute that the element has, append a U+0020
7410: SPACE character, the attribute's name (which, for attributes
7411: set by the <a href="#html-parser">HTML parser</a> or by <code title="">Element.setAttributeNode()</code> or <code title="">Element.setAttribute()</code>, will be lowercase), a
7412: U+003D EQUALS SIGN (<code title="">=</code>) character, a
7413: U+0022 QUOTATION MARK (<code title="">"</code>)
7414: character, the attribute's value, <a href="#escapingString" title="escaping a
7415: string">escaped as described below</a> in <i>attribute
7416: mode</i>, and a second U+0022 QUOTATION MARK (<code title="">"</code>) character.</p>
7417:
7418: <p>While the exact order of attributes is UA-defined, and may
7419: depend on factors such as the order that the attributes were
7420: given in the original markup, the sort order must be stable,
7421: such that consecutive invocations of this algorithm serialize an
7422: element's attributes in the same order.</p>
7423:
7424: <p>Append a U+003E GREATER-THAN SIGN (<code title="">></code>)
7425: character.</p>
7426:
7427: <p>If <var title="">current node</var> is an
1.5 mike 7428: <code><a href="the-canvas-element.html#the-area-element">area</a></code>, <code><a href="semantics.html#the-base-element">base</a></code>, <code><a href="obsolete.html#basefont">basefont</a></code>,
1.1 mike 7429: <code>bgsound</code>, <code><a href="semantics.html#the-br-element">br</a></code>, <code><a href="tabular-data.html#the-col-element">col</a></code>,
1.159 mike 7430: <code><a href="text-level-semantics.html#the-embed-element">embed</a></code>, <code><a href="obsolete.html#frame">frame</a></code>, <code><a href="semantics.html#the-hr-element">hr</a></code>,
7431: <code><a href="text-level-semantics.html#the-img-element">img</a></code>, <code><a href="forms.html#the-input-element">input</a></code>, <code><a href="forms.html#the-keygen-element">keygen</a></code>,
7432: <code><a href="semantics.html#the-link-element">link</a></code>, <code><a href="semantics.html#meta">meta</a></code>, <code><a href="text-level-semantics.html#the-param-element">param</a></code>,
1.5 mike 7433: <code><a href="obsolete.html#spacer">spacer</a></code>, or <code>wbr</code> element, then
1.1 mike 7434: continue on to the next child node at this point.</p> <!--
7435: also, i guess: image and isindex, but we don't list those
7436: because we don't consider those "elements", more "macros", and
7437: thus we should never serialize them -->
7438:
7439: <p>If <var title="">current node</var> is a <code><a href="semantics.html#the-pre-element">pre</a></code>,
1.42 mike 7440: <code><a href="forms.html#the-textarea-element">textarea</a></code>, or <code><a href="obsolete.html#listing">listing</a></code> element, append
1.1 mike 7441: a U+000A LINE FEED (LF) character.</p>
7442:
7443: <p>Append the value of running the <a href="#html-fragment-serialization-algorithm">HTML fragment
7444: serialization algorithm</a> on the <var title="">current
7445: node</var> element (thus recursing into this algorithm for
7446: that element), followed by a U+003C LESS-THAN SIGN (<code title=""><</code>) character, a U+002F SOLIDUS (<code title="">/</code>) character, the element's tag name again,
7447: and finally a U+003E GREATER-THAN SIGN (<code title="">></code>) character.</p>
7448:
7449: </dd>
7450:
7451:
7452: <dt>If <var title="">current node</var> is a <code title="">Text</code> or <code title="">CDATASection</code>
7453: node</dt>
7454:
7455: <dd>
7456:
1.19 mike 7457: <p>If the parent of <var title="">current node</var> is a
1.42 mike 7458: <code><a href="semantics.html#the-style-element">style</a></code>, <code><a href="semantics.html#script">script</a></code>, <code><a href="obsolete.html#xmp">xmp</a></code>,
1.159 mike 7459: <code><a href="text-level-semantics.html#the-iframe-element">iframe</a></code>, <code><a href="obsolete.html#noembed">noembed</a></code>,
1.222 mike 7460: <code><a href="obsolete.html#noframes">noframes</a></code>, or <code><a href="obsolete.html#plaintext">plaintext</a></code> element, or
7461: if the parent of <var title="">current node</var> is
7462: <code><a href="semantics.html#the-noscript-element">noscript</a></code> element and <a href="browsers.html#concept-n-script" title="concept-n-script">scripting is enabled</a> for the
7463: node, then append the value of <var title="">current
7464: node</var>'s <code title="">data</code> IDL attribute
7465: literally.</p>
1.1 mike 7466:
7467: <p>Otherwise, append the value of <var title="">current
1.192 mike 7468: node</var>'s <code title="">data</code> IDL attribute, <a href="#escapingString" title="escaping a string">escaped as described
1.1 mike 7469: below</a>.</p>
7470:
7471: </dd>
7472:
7473:
7474: <dt>If <var title="">current node</var> is a <code title="">Comment</code></dt>
7475:
7476: <dd>
7477:
7478: <p>Append the literal string <code><!--</code> (U+003C
7479: LESS-THAN SIGN, U+0021 EXCLAMATION MARK, U+002D HYPHEN-MINUS,
1.192 mike 7480: U+002D HYPHEN-MINUS), followed by the value of <var title="">current node</var>'s <code title="">data</code> IDL
1.1 mike 7481: attribute, followed by the literal string <code>--></code>
7482: (U+002D HYPHEN-MINUS, U+002D HYPHEN-MINUS, U+003E GREATER-THAN
7483: SIGN).</p>
7484:
7485: </dd>
7486:
7487:
7488: <dt>If <var title="">current node</var> is a <code title="">ProcessingInstruction</code></dt>
7489:
7490: <dd>
7491:
7492: <p>Append the literal string <code><?</code> (U+003C
7493: LESS-THAN SIGN, U+003F QUESTION MARK), followed by the value
1.192 mike 7494: of <var title="">current node</var>'s <code title="">target</code> IDL attribute, followed by a single
7495: U+0020 SPACE character, followed by the value of <var title="">current node</var>'s <code title="">data</code> IDL
1.1 mike 7496: attribute, followed by a single U+003E GREATER-THAN SIGN
7497: character ('>').</p>
7498:
7499: </dd>
7500:
7501:
7502: <dt>If <var title="">current node</var> is a <code title="">DocumentType</code></dt>
7503:
7504: <dd>
7505:
7506: <p>Append the literal string <code><!DOCTYPE</code> (U+003C
7507: LESS-THAN SIGN, U+0021 EXCLAMATION MARK, U+0044 LATIN CAPITAL
7508: LETTER D, U+004F LATIN CAPITAL LETTER O, U+0043 LATIN CAPITAL
7509: LETTER C, U+0054 LATIN CAPITAL LETTER T, U+0059 LATIN CAPITAL
7510: LETTER Y, U+0050 LATIN CAPITAL LETTER P, U+0045 LATIN CAPITAL
7511: LETTER E), followed by a space (U+0020 SPACE), followed by the
1.192 mike 7512: value of <var title="">current node</var>'s <code title="">name</code> IDL attribute, followed by the literal
1.1 mike 7513: string <code>></code> (U+003E GREATER-THAN SIGN).</p>
7514:
7515: </dd>
7516:
7517:
7518: </dl><p>Other node types (e.g. <code title="">Attr</code>) cannot
7519: occur as children of elements. If, despite this, they somehow do
7520: occur, this algorithm must raise an
7521: <code><a href="infrastructure.html#invalid_state_err">INVALID_STATE_ERR</a></code> exception.</p>
7522:
7523: </li>
7524:
7525: </ol></li>
7526:
7527: <li><p>The result of the algorithm is the string <var title="">s</var>.</p></li>
7528:
7529: </ol><p><dfn id="escapingString">Escaping a string</dfn> (for the
7530: purposes of the algorithm above) consists of replacing any
7531: occurrences of the "<code title="">&</code>" character by the
7532: string "<code title="">&amp;</code>", any occurrences of the
7533: U+00A0 NO-BREAK SPACE character by the string "<code title="">&nbsp;</code>", and, if the algorithm was invoked in
7534: the <i>attribute mode</i>, any occurrences of the "<code title="">"</code>" character by the string "<code title="">&quot;</code>", or if it was not, any occurrences of
7535: the "<code title=""><</code>" character by the string "<code title="">&lt;</code>", any occurrences of the "<code title="">></code>" character by the string "<code title="">&gt;</code>".</p>
7536:
7537: <p class="note">Entity reference nodes are <a href="infrastructure.html#entity-references">assumed to be expanded</a> by the user
7538: agent, and are therefore not covered in the algorithm above.</p>
7539:
7540: <p class="note">It is possible that the output of this algorithm, if
7541: parsed with an <a href="#html-parser">HTML parser</a>, will not return the
7542: original tree structure. For instance, if a <code><a href="forms.html#the-textarea-element">textarea</a></code>
7543: element to which a <code title="">Comment</code> node has been
7544: appended is serialized and the output is then reparsed, the comment
7545: will end up being displayed in the text field. Similarly, if, as a
7546: result of DOM manipulation, an element contains a comment that
7547: contains the literal string "<code title="">--></code>", then
7548: when the result of serializing the element is parsed, the comment
7549: will be truncated at that point and the rest of the comment will be
7550: interpreted as markup. More examples would be making a
7551: <code><a href="semantics.html#script">script</a></code> element contain a text node with the text string
7552: "<code></script></code>", or having a <code><a href="semantics.html#the-p-element">p</a></code> element that
7553: contains a <code><a href="semantics.html#the-ul-element">ul</a></code> element (as the <code><a href="semantics.html#the-ul-element">ul</a></code> element's
7554: <a href="#syntax-start-tag" title="syntax-start-tag">start tag</a> would imply the end
7555: tag for the <code><a href="semantics.html#the-p-element">p</a></code>).</p>
7556:
7557:
7558: <h3 id="parsing-html-fragments"><span class="secno">9.5 </span>Parsing HTML fragments</h3>
7559:
7560: <p>The following steps form the <dfn id="html-fragment-parsing-algorithm">HTML fragment parsing
7561: algorithm</dfn>. The algorithm optionally takes as input an
7562: <code>Element</code> node, referred to as the <var title="">context</var> element, which gives the context for the
7563: parser, as well as <var title="">input</var>, a string to parse, and
7564: returns a list of zero or more nodes.</p>
7565:
7566: <p class="note">Parts marked <dfn id="fragment-case">fragment case</dfn> in algorithms
7567: in the parser section are parts that only occur if the parser was
7568: created for the purposes of this algorithm (and with a <var title="">context</var> element). The algorithms have been annotated
7569: with such markings for informational purposes only; such markings
7570: have no normative weight. If it is possible for a condition
7571: described as a <a href="#fragment-case">fragment case</a> to occur even when the
7572: parser wasn't created for the purposes of handling this algorithm,
7573: then that is an error in the specification.</p>
7574:
7575: <ol><li>
7576:
7577: <p>Create a new <code>Document</code> node, and mark it as being
7578: an <a href="dom.html#html-documents" title="HTML documents">HTML document</a>.</p>
7579:
7580: </li>
7581:
7582: <li>
7583:
7584: <p>If there is a <var title="">context</var> element, and the
7585: <code>Document</code> of the <var title="">context</var> element
7586: is in <a href="dom.html#quirks-mode">quirks mode</a>, then let the <code>Document</code>
7587: be in <a href="dom.html#quirks-mode">quirks mode</a>. Otherwise, if there is a <var title="">context</var> element, and the <code>Document</code> of
7588: the <var title="">context</var> element is in <a href="dom.html#limited-quirks-mode">limited quirks
7589: mode</a>, then let the <code>Document</code> be in
7590: <a href="dom.html#limited-quirks-mode">limited quirks mode</a>. Otherwise, leave the
7591: <code>Document</code> in <a href="dom.html#no-quirks-mode">no quirks mode</a>.</p>
7592:
7593: </li>
7594:
7595: <li>
7596:
7597: <p>Create a new <a href="#html-parser">HTML parser</a>, and associate it with
7598: the just created <code>Document</code> node.</p>
7599:
7600: </li>
7601:
7602: <li>
7603:
7604: <p>If there is a <var title="">context</var> element, run these
7605: substeps:</p>
7606:
7607: <ol><li>
7608:
7609: <p>Set the <a href="#html-parser">HTML parser</a>'s <a href="#tokenization">tokenization</a>
7610: stage's <a href="#content-model-flag">content model flag</a> according to the <var title="">context</var> element, as follows:</p>
7611:
1.159 mike 7612: <dl class="switch"><dt>If it is a <code><a href="semantics.html#the-title-element-0">title</a></code> or <code><a href="forms.html#the-textarea-element">textarea</a></code>
1.1 mike 7613: element</dt>
7614:
7615: <dd>Set the <a href="#content-model-flag">content model flag</a> to
7616: the RCDATA state.</dd>
7617:
7618:
7619: <dt>If it is a <code><a href="semantics.html#the-style-element">style</a></code>, <code><a href="semantics.html#script">script</a></code>,
1.159 mike 7620: <code><a href="obsolete.html#xmp">xmp</a></code>, <code><a href="text-level-semantics.html#the-iframe-element">iframe</a></code>, <code><a href="obsolete.html#noembed">noembed</a></code>, or
1.5 mike 7621: <code><a href="obsolete.html#noframes">noframes</a></code> element</dt>
1.1 mike 7622:
7623: <dd>Set the <a href="#content-model-flag">content model flag</a> to
1.91 mike 7624: the RAWTEXT state.</dd>
1.1 mike 7625:
7626:
7627: <dt>If it is a <code><a href="semantics.html#the-noscript-element">noscript</a></code> element</dt>
7628:
7629: <dd>If the <a href="#scripting-flag">scripting flag</a> is enabled, set the
1.91 mike 7630: <a href="#content-model-flag">content model flag</a> to the RAWTEXT
1.1 mike 7631: state. Otherwise, set the <a href="#content-model-flag">content model flag</a> to the
7632: PCDATA state.</dd>
7633:
7634:
1.42 mike 7635: <dt>If it is a <code><a href="obsolete.html#plaintext">plaintext</a></code> element</dt>
1.1 mike 7636:
7637: <dd>Set the <a href="#content-model-flag">content model flag</a> to
7638: PLAINTEXT.</dd>
7639:
7640:
7641: <dt>Otherwise</dt>
7642:
7643: <dd>Leave the <a href="#content-model-flag">content model flag</a> in the PCDATA
7644: state.</dd>
7645:
7646: </dl></li>
7647:
7648: <li>
7649:
1.159 mike 7650: <p>Let <var title="">root</var> be a new <code><a href="semantics.html#the-html-element-0">html</a></code> element
1.1 mike 7651: with no attributes.</p>
7652:
7653: </li>
7654:
7655: <li>
7656:
7657: <p>Append the element <var title="">root</var> to the
7658: <code>Document</code> node created above.</p>
7659:
7660: </li>
7661:
7662: <li>
7663:
7664: <p>Set up the parser's <a href="#stack-of-open-elements">stack of open elements</a> so that
7665: it contains just the single element <var title="">root</var>.</p>
7666:
7667: </li>
7668:
7669: <li>
7670:
7671: <p><a href="#reset-the-insertion-mode-appropriately" title="reset the insertion mode appropriately">Reset the
7672: parser's insertion mode appropriately</a>.</p>
7673:
7674: <p class="note">The parser will reference the <var title="">context</var> element as part of that algorithm.</p>
7675:
7676: </li>
7677:
7678: <li>
7679:
7680: <p>Set the parser's <a href="#form-element-pointer"><code>form</code> element pointer</a>
7681: to the nearest node to the <var title="">context</var> element
7682: that is a <code><a href="forms.html#the-form-element">form</a></code> element (going straight up the
7683: ancestor chain, and including the element itself, if it is a
7684: <code><a href="forms.html#the-form-element">form</a></code> element), or, if there is no such
7685: <code><a href="forms.html#the-form-element">form</a></code> element, to null.</p>
7686:
7687: </li>
7688:
7689: </ol></li>
7690:
7691: <li>
7692:
7693: <p>Place into the <a href="#the-input-stream">input stream</a> for the <a href="#html-parser">HTML
7694: parser</a> just created the <var title="">input</var>. The
7695: encoding <a href="#concept-encoding-confidence" title="concept-encoding-confidence">confidence</a> is
7696: <i>irrelevant</i>.</p>
7697:
7698: </li>
7699:
7700: <li>
7701:
7702: <p>Start the parser and let it run until it has consumed all the
7703: characters just inserted into the input stream.</p>
7704:
7705: </li>
7706:
7707: <li>
7708:
7709: <p>If there is a <var title="">context</var> element, return the
7710: child nodes of <var title="">root</var>, in <a href="infrastructure.html#tree-order">tree
7711: order</a>.</p>
7712:
7713: <p>Otherwise, return the children of the <code>Document</code>
7714: object, in <a href="infrastructure.html#tree-order">tree order</a>.</p>
7715:
7716: </li>
7717:
1.123 mike 7718: </ol></div></body></html>
Webmaster
![[BACK]](/https/dev.w3.org/cvsweb/icons/back.gif){kind=icon}
Return to syntax.html CVS log ![[TXT]](/https/dev.w3.org/cvsweb/icons/text.gif){kind=icon}
![[DIR]](/https/dev.w3.org/cvsweb/icons/dir.gif){kind=icon}
Up to [Public] / html5 / spec