[ruby/prism] Fix incorrect paring when using invalid regexp options

Fixes https://github.com/ruby/prism/pull/2617.

There was an issue with the lexer as follows.
The following are valid regexp options:

```console
$ bundle exec ruby -Ilib -rprism -ve 'p Prism.lex("/x/io").value.map {|token| token[0].type }'
ruby 3.3.0 (2023-12-25 revision https://github.com/ruby/prism/commit/5124f9ac75) [x86_64-darwin22]
[:REGEXP_BEGIN, :STRING_CONTENT, :REGEXP_END, :EOF]
```

The following are invalid regexp options. Unnecessary the `IDENTIFIER` token is appearing:

```console
$ bundle exec ruby -Ilib -rprism -ve 'p Prism.lex("/x/az").value.map {|token| token[0].type }'
ruby 3.3.0 (2023-12-25 revision https://github.com/ruby/prism/commit/5124f9ac75) [x86_64-darwin22]
[:REGEXP_BEGIN, :STRING_CONTENT, :REGEXP_END, :IDENTIFIER, :EOF]
```

As a behavior of Ruby, when given `A` to `Z` and `a` to `z`, they act as invalid regexp options. e.g.,

```console
$ ruby -e '/regexp/az'
-e:1: unknown regexp options - az
/regexp/az
-e: compile error (SyntaxError)
```

Thus, it should probably not be construed as `IDENTIFIER` token.

Therefore, `pm_byte_table` has been adapted to accept those invalid regexp option values.
Whether it is a valid regexp option or not is checked by `pm_regular_expression_flags_create`.
For invalid regexp options, `PM_ERR_REGEXP_UNKNOWN_OPTIONS` is added to diagnostics.

https://github.com/ruby/prism/commit/d2a6096fcf

7 files changed, 39 insertions, 10 deletions

diff --git a/lib/prism/translation/parser.rb b/lib/prism/translation/parser.rb
index 8df7164688..0d11b8f566 100644
--- a/lib/prism/translation/parser.rb
+++ b/lib/prism/translation/parser.rb
@@ -173,6 +173,8 @@ module Prism
           Diagnostic.new(:error, :duplicate_argument, {}, diagnostic_location, [])
         when :parameter_numbered_reserved
           Diagnostic.new(:error, :reserved_for_numparam, { name: location.slice }, diagnostic_location, [])
+        when :regexp_unknown_options
+          Diagnostic.new(:error, :regexp_options, { options: location.slice[1..] }, diagnostic_location, [])
         when :singleton_for_literals
           Diagnostic.new(:error, :singleton_literal, {}, diagnostic_location, [])
         when :string_literal_eof
diff --git a/prism/config.yml b/prism/config.yml
index 269bfa73ec..d9e39460d1 100644
--- a/prism/config.yml
+++ b/prism/config.yml
@@ -199,6 +199,7 @@ errors:
   - REGEXP_INVALID_UNICODE_RANGE
   - REGEXP_NON_ESCAPED_MBC
   - REGEXP_TERM
+  - REGEXP_UNKNOWN_OPTIONS
   - REGEXP_UTF8_CHAR_NON_UTF8_REGEXP
   - RESCUE_EXPRESSION
   - RESCUE_MODIFIER_VALUE
diff --git a/prism/prism.c b/prism/prism.c
index 8a6ca0eccc..b410b0a510 100644
--- a/prism/prism.c
+++ b/prism/prism.c
@@ -1214,10 +1214,12 @@ pm_node_flag_set_repeated_parameter(pm_node_t *node) {
  * Parse out the options for a regular expression.
  */
 static inline pm_node_flags_t
-pm_regular_expression_flags_create(const pm_token_t *closing) {
+pm_regular_expression_flags_create(pm_parser_t *parser, const pm_token_t *closing) {
     pm_node_flags_t flags = 0;
 
     if (closing->type == PM_TOKEN_REGEXP_END) {
+        pm_buffer_t unknown_flags = { 0 };
+
         for (const uint8_t *flag = closing->start + 1; flag < closing->end; flag++) {
             switch (*flag) {
                 case 'i': flags |= PM_REGULAR_EXPRESSION_FLAGS_IGNORE_CASE; break;
@@ -1230,9 +1232,16 @@ pm_regular_expression_flags_create(const pm_token_t *closing) {
                 case 's': flags = (pm_node_flags_t) (((pm_node_flags_t) (flags & PM_REGULAR_EXPRESSION_ENCODING_MASK)) | PM_REGULAR_EXPRESSION_FLAGS_WINDOWS_31J); break;
                 case 'u': flags = (pm_node_flags_t) (((pm_node_flags_t) (flags & PM_REGULAR_EXPRESSION_ENCODING_MASK)) | PM_REGULAR_EXPRESSION_FLAGS_UTF_8); break;
 
-                default: assert(false && "unreachable");
+                default: pm_buffer_append_byte(&unknown_flags, *flag);
             }
         }
+
+        size_t unknown_flags_length = pm_buffer_length(&unknown_flags);
+        if (unknown_flags_length != 0) {
+            char *word = unknown_flags_length >= 2 ? "options" : "option";
+            PM_PARSER_ERR_TOKEN_FORMAT(parser, parser->previous, PM_ERR_REGEXP_UNKNOWN_OPTIONS, word, unknown_flags_length, pm_buffer_value(&unknown_flags));
+        }
+        pm_buffer_free(&unknown_flags);
     }
 
     return flags;
@@ -4297,10 +4306,10 @@ pm_interpolated_regular_expression_node_append(pm_interpolated_regular_expressio
 }
 
 static inline void
-pm_interpolated_regular_expression_node_closing_set(pm_interpolated_regular_expression_node_t *node, const pm_token_t *closing) {
+pm_interpolated_regular_expression_node_closing_set(pm_parser_t *parser, pm_interpolated_regular_expression_node_t *node, const pm_token_t *closing) {
     node->closing_loc = PM_LOCATION_TOKEN_VALUE(closing);
     node->base.location.end = closing->end;
-    pm_node_flag_set((pm_node_t *)node, pm_regular_expression_flags_create(closing));
+    pm_node_flag_set((pm_node_t *)node, pm_regular_expression_flags_create(parser, closing));
 }
 
 /**
@@ -5528,7 +5537,7 @@ pm_regular_expression_node_create_unescaped(pm_parser_t *parser, const pm_token_
     *node = (pm_regular_expression_node_t) {
         {
             .type = PM_REGULAR_EXPRESSION_NODE,
-            .flags = pm_regular_expression_flags_create(closing) | PM_NODE_FLAG_STATIC_LITERAL,
+            .flags = pm_regular_expression_flags_create(parser, closing) | PM_NODE_FLAG_STATIC_LITERAL,
             .location = {
                 .start = MIN(opening->start, closing->start),
                 .end = MAX(opening->end, closing->end)
@@ -17490,7 +17499,7 @@ parse_expression_prefix(pm_parser_t *parser, pm_binding_power_t binding_power, b
                 expect1(parser, PM_TOKEN_REGEXP_END, PM_ERR_REGEXP_TERM);
             }
 
-            pm_interpolated_regular_expression_node_closing_set(interpolated, &closing);
+            pm_interpolated_regular_expression_node_closing_set(parser, interpolated, &closing);
             return (pm_node_t *) interpolated;
         }
         case PM_TOKEN_BACKTICK:
diff --git a/prism/templates/src/diagnostic.c.erb b/prism/templates/src/diagnostic.c.erb
index 2a3ac19930..818b12d98b 100644
--- a/prism/templates/src/diagnostic.c.erb
+++ b/prism/templates/src/diagnostic.c.erb
@@ -277,6 +277,7 @@ static const pm_diagnostic_data_t diagnostic_messages[PM_DIAGNOSTIC_ID_MAX] = {
     [PM_ERR_REGEXP_INCOMPAT_CHAR_ENCODING]      = { "incompatible character encoding: /%.*s/", PM_ERROR_LEVEL_FATAL },
     [PM_ERR_REGEXP_NON_ESCAPED_MBC]             = { "/.../n has a non escaped non ASCII character in non ASCII-8BIT script: /%.*s/", PM_ERROR_LEVEL_FATAL },
     [PM_ERR_REGEXP_INVALID_UNICODE_RANGE]       = { "invalid Unicode range: /%.*s/", PM_ERROR_LEVEL_FATAL },
+    [PM_ERR_REGEXP_UNKNOWN_OPTIONS]             = { "unknown regexp %s: %.*s", PM_ERROR_LEVEL_FATAL },
     [PM_ERR_REGEXP_TERM]                        = { "expected a closing delimiter for the regular expression", PM_ERROR_LEVEL_FATAL },
     [PM_ERR_REGEXP_UTF8_CHAR_NON_UTF8_REGEXP]   = { "UTF-8 character in non UTF-8 regexp: /%s/", PM_ERROR_LEVEL_FATAL },
     [PM_ERR_RESCUE_EXPRESSION]                  = { "expected a rescued expression", PM_ERROR_LEVEL_FATAL },
diff --git a/prism/util/pm_char.c b/prism/util/pm_char.c
index 13eddbba48..dce19abd1b 100644
--- a/prism/util/pm_char.c
+++ b/prism/util/pm_char.c
@@ -19,10 +19,10 @@ static const uint8_t pm_byte_table[256] = {
     0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, // 1x
     3, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, // 2x
     0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, // 3x
-    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, // 4x
-    0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, // 5x
-    0, 0, 0, 0, 0, 4, 0, 0, 0, 4, 0, 0, 0, 4, 4, 4, // 6x
-    0, 0, 0, 4, 0, 4, 0, 0, 4, 0, 0, 0, 0, 0, 0, 0, // 7x
+    0, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, // 4x
+    4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 0, 0, 0, 0, 0, // 5x
+    0, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, // 6x
+    4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 0, 0, 0, 0, 0, // 7x
     0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, // 8x
     0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, // 9x
     0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, // Ax
diff --git a/test/prism/errors_test.rb b/test/prism/errors_test.rb
index ccf7485c7b..9221d52ef3 100644
--- a/test/prism/errors_test.rb
+++ b/test/prism/errors_test.rb
@@ -2067,6 +2067,20 @@ module Prism
       assert_errors expression(source), source, errors, compare_ripper: false
     end
 
+    def test_regular_expression_with_unknown_regexp_options
+      source = "/foo/AZaz"
+      errors = [["unknown regexp options: AZaz", 4..9]]
+
+      assert_errors expression(source), source, errors
+    end
+
+    def test_interpolated_regular_expression_with_unknown_regexp_options
+      source = "/\#{foo}/AZaz"
+      errors = [["unknown regexp options: AZaz", 7..12]]
+
+      assert_errors expression(source), source, errors
+    end
+
     def test_singleton_method_for_literals
       source = <<~'RUBY'
         def (1).g; end
diff --git a/test/prism/location_test.rb b/test/prism/location_test.rb
index c7ce248b56..b7b9a754ca 100644
--- a/test/prism/location_test.rb
+++ b/test/prism/location_test.rb
@@ -527,6 +527,7 @@ module Prism
 
     def test_InterpolatedRegularExpressionNode
       assert_location(InterpolatedRegularExpressionNode, "/\#{foo}/")
+      assert_location(InterpolatedRegularExpressionNode, "/\#{foo}/io")
     end
 
     def test_InterpolatedStringNode
@@ -730,6 +731,7 @@ module Prism
 
     def test_RegularExpressionNode
       assert_location(RegularExpressionNode, "/foo/")
+      assert_location(RegularExpressionNode, "/foo/io")
     end
 
     def test_RequiredKeywordParameterNode