Changeset 60916 in webkit

Timestamp:

Jun 9, 2010 3:05:20 PM (14 years ago)

Author:

abarth@webkit.org

Message:

2010-06-09 Tony Gentilcore <​tonyg@chromium.org>

Reviewed by Adam Barth.

Fix fast/parser/hex-entities-length.html for HTML5 parser
​https://bugs.webkit.org/show_bug.cgi?id=40385

Stripping surrogate pair values appears to be a problem with the spec
in "Preprocessing the input stream." Minefield doesn't appear to
implement the part in question. So this patch removes that guard and
adds a FIXME to track the issue.

No new tests because covered by fast/parser/hex-entities-length.html

• html/HTML5Lexer.h: (WebCore::HTML5Lexer::InputStreamPreprocessor::peek):

Location:

trunk/WebCore

Files:

• ChangeLog (modified) (1 diff)
• html/HTML5Lexer.h (modified) (1 diff)

trunk/WebCore/ChangeLog

@@ +1 @@
+2010-06-09  Tony Gentilcore  <tonyg@chromium.org>
+
+        Reviewed by Adam Barth.
+
+        Fix fast/parser/hex-entities-length.html for HTML5 parser
+        https://bugs.webkit.org/show_bug.cgi?id=40385
+
+        Stripping surrogate pair values appears to be a problem with the spec
+        in "Preprocessing the input stream." Minefield doesn't appear to
+        implement the part in question. So this patch removes that guard and
+        adds a FIXME to track the issue.
+
+        No new tests because covered by fast/parser/hex-entities-length.html
+
+        * html/HTML5Lexer.h:
+        (WebCore::HTML5Lexer::InputStreamPreprocessor::peek):
+
 2010-06-09  Kenneth Russell  <kbr@google.com>
 

trunk/WebCore/html/HTML5Lexer.h

@@ -160 +160 @@
                 } else {
                     m_skipNextNewLine = false;
-                    if (m_nextInputCharacter == '\0' || (m_nextInputCharacter >= 0xD800 && m_nextInputCharacter <= 0xDFFF))
+                    // FIXME: The spec indicates that the surrogate pair range as well as
+                    // a number of specific character values are parse errors and should be replaced
+                    // by the replacement character. We suspect this is a problem with the spec as doing
+                    // that filtering breaks surrogate pair handling and causes us not to match Minefield.
+                    if (m_nextInputCharacter == '\0')
                         m_nextInputCharacter = 0xFFFD;
                 }