Extract author name from itemprop='name'. (#943)

* Extract author name from itemprop='name'. Fixes #935

* De-dupe textContent.trim()

Author: danielnixon

Readability.js

@@ -940,6 +940,10 @@ Readability.prototype = {
    * (and its kids) are going away, and we want the next node over.
    *
    * Calling this in a loop will traverse the DOM depth-first.
+   *
+   * @param {Element} node
+   * @param {boolean} ignoreSelfAndKids
+   * @return {Element}
    */
   _getNextNode(node, ignoreSelfAndKids) {
     // First check for kids if those aren't being ignored
@@ -1079,7 +1083,20 @@ Readability.prototype = {
           !this._metadata.byline &&
           this._isValidByline(node, matchString)
         ) {
-          this._articleByline = node.textContent.trim();
+          // Find child node matching [itemprop="name"] and use that if it exists for a more accurate author name byline
+          var endOfSearchMarkerNode = this._getNextNode(node, true);
+          var next = this._getNextNode(node);
+          var itemPropNameNode = null;
+          while (next && next != endOfSearchMarkerNode) {
+            var itemprop = next.getAttribute("itemprop");
+            if (itemprop && itemprop.includes("name")) {
+              itemPropNameNode = next;
+              break;
+            } else {
+              next = this._getNextNode(next);
+            }
+          }
+          this._articleByline = (itemPropNameNode ?? node).textContent.trim();
           node = this._removeAndGetNext(node);
           continue;
         }

test/test-pages/001/expected-metadata.json

@@ -1,6 +1,6 @@
 {
   "title": "Get your Frontend JavaScript Code Covered | Code",
-  "byline": "Nicolas Perriault —",
+  "byline": "Nicolas Perriault",
   "dir": null,
   "lang": "en",
   "excerpt": "Nicolas Perriault's homepage.",

test/test-pages/ars-1/expected-metadata.json

@@ -1,6 +1,6 @@
 {
   "title": "Just-released Minecraft exploit makes it easy to crash game servers",
-  "byline": "Dan Goodin - Apr 16, 2015 8:02 pm UTC",
+  "byline": "Dan Goodin",
   "dir": null,
   "lang": "en-us",
   "excerpt": "Two-year-old bug exposes thousands of servers to crippling attack.",

test/test-pages/msn/expected-metadata.json

@@ -1,6 +1,6 @@
 {
   "title": "Nintendo's first iPhone game will launch in December for $10",
-  "byline": "Alex Perry\n                        \n                        1 day ago",
+  "byline": "Alex Perry",
   "dir": "ltr",
   "lang": "en-US",
   "excerpt": "Nintendo and Apple shocked the world earlier this year by announcing \"Super Mario Run,\" the legendary gaming company's first foray into mobile gaming. ",

test/test-pages/nytimes-3/expected-metadata.json

@@ -1,6 +1,6 @@
 {
   "title": "Manhole Fires and Burst Pipes: How Winter Wreaks Havoc on What’s Underneath N.Y.C.",
-  "byline": "By Corey Kilgannon",
+  "byline": "Corey Kilgannon",
   "dir": null,
   "lang": "en",
   "excerpt": "New York’s aging below-street infrastructure is tough to maintain, and the corrosive rock salt and “freeze-thaw” cycles of winter make it even worse.",

test/test-pages/nytimes-4/expected-metadata.json

@@ -1,6 +1,6 @@
 {
   "title": "As Debt Rises, the Government Will Soon Spend More on Interest Than on the Military",
-  "byline": "By Nelson D. Schwartz",
+  "byline": "Nelson D. Schwartz",
   "dir": null,
   "lang": "en",
   "excerpt": "Tax cuts, spending increases and higher interest rates could make it harder to respond to future recessions and deal with other needs.",