[SCM] WebKit Debian packaging branch, debian/experimental, updated. upstream/1.3.3-9427-gc2be6fc

[abarth at webkit.org]

The following commit has been merged in the debian/experimental branch:
commit aeb3e23d075125102646b0b9d3d323cd40ee2f2e
Author: abarth at webkit.org <abarth at webkit.org@268f45cc-cd09-0410-ab3c-d52691b4dbfc>
Date:   Thu Sep 9 00:18:53 2010 +0000

    2010-09-08  Adam Barth  <abarth at webkit.org>
    
            Reviewed by Eric Seidel.
    
            text/plain documents that start with \n trigger ASSERT
            https://bugs.webkit.org/show_bug.cgi?id=45409
    
            Because the TextDocumentParser uses a <pre> element to show the text,
            it ran into a requirement from HTML5 to skip any initial \n character
            tokens inside a <pre> element.  We don't want the behavior for
            TextDocuments, so I've loosened our ASSERTs and added a comment and a
            test to document the correct behavior.
    
            Test: fast/tokenizer/text-plain.html
    
            * html/parser/HTMLTokenizer.cpp:
            (WebCore::HTMLTokenizer::nextToken):
    2010-09-08  Adam Barth  <abarth at webkit.org>
    
            Reviewed by Eric Seidel.
    
            text/plain documents that start with \n trigger ASSERT
            https://bugs.webkit.org/show_bug.cgi?id=45409
    
            * fast/tokenizer/resources/text.txt: Added.
            * fast/tokenizer/text-plain.html: Added.
    
    
    git-svn-id: http://svn.webkit.org/repository/webkit/trunk@67041 268f45cc-cd09-0410-ab3c-d52691b4dbfc

diff --git a/LayoutTests/ChangeLog b/LayoutTests/ChangeLog
index 516a6a0..9d128c5 100644
--- a/LayoutTests/ChangeLog
+++ b/LayoutTests/ChangeLog
@@ -1,3 +1,13 @@
+2010-09-08  Adam Barth  <abarth at webkit.org>
+
+        Reviewed by Eric Seidel.
+
+        text/plain documents that start with \n trigger ASSERT
+        https://bugs.webkit.org/show_bug.cgi?id=45409
+
+        * fast/tokenizer/resources/text.txt: Added.
+        * fast/tokenizer/text-plain.html: Added.
+
 2010-09-08  Yury Semikhatsky  <yurys at chromium.org>
 
         Reviewed by Joseph Pecoraro.
diff --git a/LayoutTests/fast/tokenizer/resources/text.txt b/LayoutTests/fast/tokenizer/resources/text.txt
new file mode 100644
index 0000000..fa13561
--- /dev/null
+++ b/LayoutTests/fast/tokenizer/resources/text.txt
@@ -0,0 +1,8 @@
+
+hi
+</pre>
+<strong>more</strong>
+!!!
+</plaintext>
+
+ggg
diff --git a/LayoutTests/fast/tokenizer/text-plain-expected.txt b/LayoutTests/fast/tokenizer/text-plain-expected.txt
new file mode 100644
index 0000000..34b460e
--- /dev/null
+++ b/LayoutTests/fast/tokenizer/text-plain-expected.txt
@@ -0,0 +1,14 @@
+
+
+--------
+Frame: '<!--framePath //<!--frame0-->-->'
+--------
+
+hi
+</pre>
+<strong>more</strong>
+!!!
+</plaintext>
+
+ggg
+
diff --git a/LayoutTests/fast/tokenizer/text-plain.html b/LayoutTests/fast/tokenizer/text-plain.html
new file mode 100644
index 0000000..8d3b727
--- /dev/null
+++ b/LayoutTests/fast/tokenizer/text-plain.html
@@ -0,0 +1,7 @@
+<script>
+if (window.layoutTestController) {
+    layoutTestController.dumpAsText();
+    layoutTestController.dumpChildFramesAsText();
+}
+</script>
+<iframe src="resources/text.txt"></iframe>
diff --git a/WebCore/ChangeLog b/WebCore/ChangeLog
index 39d1c40..d6e657e 100644
--- a/WebCore/ChangeLog
+++ b/WebCore/ChangeLog
@@ -1,3 +1,21 @@
+2010-09-08  Adam Barth  <abarth at webkit.org>
+
+        Reviewed by Eric Seidel.
+
+        text/plain documents that start with \n trigger ASSERT
+        https://bugs.webkit.org/show_bug.cgi?id=45409
+
+        Because the TextDocumentParser uses a <pre> element to show the text,
+        it ran into a requirement from HTML5 to skip any initial \n character
+        tokens inside a <pre> element.  We don't want the behavior for
+        TextDocuments, so I've loosened our ASSERTs and added a comment and a
+        test to document the correct behavior.
+
+        Test: fast/tokenizer/text-plain.html
+
+        * html/parser/HTMLTokenizer.cpp:
+        (WebCore::HTMLTokenizer::nextToken):
+
 2010-09-08  James Robinson  <jamesr at chromium.org>
 
         Reviewed by Kenneth Russell.
diff --git a/WebCore/html/parser/HTMLTokenizer.cpp b/WebCore/html/parser/HTMLTokenizer.cpp
index 5791842..f5405ff 100644
--- a/WebCore/html/parser/HTMLTokenizer.cpp
+++ b/WebCore/html/parser/HTMLTokenizer.cpp
@@ -293,7 +293,9 @@ bool HTMLTokenizer::nextToken(SegmentedString& source, HTMLToken& token)
                 ADVANCE_TO(DataState);
             if (m_state == RCDATAState)
                 ADVANCE_TO(RCDATAState);
-            ASSERT_NOT_REACHED();
+            // When parsing text/plain documents, we run the tokenizer in the
+            // PLAINTEXTState and ignore m_skipLeadingNewLineForListing.
+            ASSERT(m_state == PLAINTEXTState);
         }
     }
 

-- 
WebKit Debian packaging