Support escaped char codes in WebVTT tokenizer.

Source/core/html/track/vtt/VTTTokenizer.cpp

 /*
  * Copyright (C) 2011 Google Inc.  All rights reserved.
  *
  * Redistribution and use in source and binary forms, with or without
  * modification, are permitted provided that the following conditions are
  * met:
  *
  *     * Redistributions of source code must retain the above copyright
  * notice, this list of conditions and the following disclaimer.
  *     * Redistributions in binary form must reproduce the above
@@ skipping 88 matching lines @@
     StringBuilder classes;
     enum {
         DataState,
         EscapeState,
         TagState,
         StartTagState,
         StartTagClassState,
         StartTagAnnotationState,
         EndTagState,
         TimestampTagState,
+        EscapeCharacterState,
     } state = DataState;
 
     // 4.8.10.13.4 WebVTT cue text tokenizer
     switch (state) {
         WEBVTT_BEGIN_STATE(DataState) {
             if (cc == '&') {
                 buffer.append(static_cast<LChar>(cc));
                 WEBVTT_ADVANCE_TO(EscapeState);
             } else if (cc == '<') {
                 if (result.isEmpty()) {
@@ skipping 25 matching lines @@
                 } else if (equalLiteral(buffer, "&rlm")) {
                     result.append(rightToLeftMark);
                 } else if (equalLiteral(buffer, "&nbsp")) {
                     result.append(noBreakSpace);
                 } else {
                     buffer.append(static_cast<LChar>(cc));
                     result.append(buffer);
                 }
                 buffer.clear();
                 WEBVTT_ADVANCE_TO(DataState);
+            } else if (cc == '#') {
+                buffer.clear();
+                WEBVTT_ADVANCE_TO(EscapeCharacterState);
             } else if (isASCIIAlphanumeric(cc)) {
                 buffer.append(static_cast<LChar>(cc));
                 WEBVTT_ADVANCE_TO(EscapeState);
             } else if (cc == '<') {
                 result.append(buffer);
                 return emitToken(token, VTTToken::StringToken(result.toString()));
             } else if (cc == kEndOfFileMarker) {
                 result.append(buffer);
                 return advanceAndEmitToken(m_input, token, VTTToken::StringToken(result.toString()));
             } else {
                 result.append(buffer);
                 buffer.clear();
 
                 if (cc == '&') {
                     buffer.append(static_cast<LChar>(cc));
                     WEBVTT_ADVANCE_TO(EscapeState);
                 }
                 result.append(cc);
                 WEBVTT_ADVANCE_TO(DataState);
             }
         }
         END_STATE()
 
+        WEBVTT_BEGIN_STATE(EscapeCharacterState) {
+            if (cc == ';') {
+                unsigned charCode = buffer.toString().toUIntStrict();
+                result.append(static_cast<UChar>(charCode));
+                buffer.clear();
+                WEBVTT_ADVANCE_TO(DataState);
+            } else if (WTF::isASCIIDigit(cc)) {
+                buffer.append(static_cast<LChar>(cc));
+                WEBVTT_ADVANCE_TO(EscapeCharacterState);
+            } else {
+                buffer.clear();
+                WEBVTT_ADVANCE_TO(DataState);
+            }
+        }
+        END_STATE()
+
         WEBVTT_BEGIN_STATE(TagState) {
             if (isTokenizerWhitespace(cc)) {
                 ASSERT(result.isEmpty());
                 WEBVTT_ADVANCE_TO(StartTagAnnotationState);
             } else if (cc == '.') {
                 ASSERT(result.isEmpty());
                 WEBVTT_ADVANCE_TO(StartTagClassState);
             } else if (cc == '/') {
                 WEBVTT_ADVANCE_TO(EndTagState);
             } else if (WTF::isASCIIDigit(cc)) {
@@ skipping 69 matching lines @@
         END_STATE()
 
     }
 
     ASSERT_NOT_REACHED();
     return false;
 }
 
 }