[Webkit-unassigned] [Bug 31144] [Qt] The XML tokenizer reports a parse error twice if it occurs before the document element is found.

bugzilla-daemon at webkit.org bugzilla-daemon at webkit.org
Wed Nov 4 16:48:10 PST 2009


https://bugs.webkit.org/show_bug.cgi?id=31144


Holger Freyther <zecke at selfish.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
  Attachment #42530|review?                     |review-
               Flag|                            |




--- Comment #3 from Holger Freyther <zecke at selfish.org>  2009-11-04 16:48:10 PDT ---
(From update of attachment 42530)

> -    if (m_stream.error() == QXmlStreamReader::PrematureEndOfDocumentError
> -        || (m_wroteText && !m_sawFirstElement && !m_sawXSLTransform))
> +    if (m_stream.error() == QXmlStreamReader::PrematureEndOfDocumentError)

I'm not convinced. Why do you remove m_sawXSLTransform? What is with the
originally mentioned XML Http Request test, is it still passing, why it passing
without the change?

So the question is the following:
  "libxml2 has the semantic that when writing an empty string and finishing it,
it will report an error. For QXmlStreamReader this is valid."

Is that still true, was that true, was the semantic of QXmlStreamReader
changed?



> diff --git a/WebCore/manual-tests/qt/parse-error-reported-twice.xml b/WebCore/manual-tests/qt/parse-error-reported-twice.xml
> new file mode 100644
> index 0000000..0b97956
> --- /dev/null
> +++ b/WebCore/manual-tests/qt/parse-error-reported-twice.xml
> @@ -0,0 +1,5 @@
> +<?xml version="1.0" encoding="utf-8"?>
> +<!-- Trigger a parse error in the doctype. -->
> +<!DOCTYPE doc {}>
> +<doc></doc>
> +<!-- The tokenizer should report a parse failure _once_. -->
> -- 
> 1.6.4.2
> 

Make that a proper layout test, don't worry about the locale it is the same for
everyone and we can change it via the LayoutTestController

-- 
Configure bugmail: https://bugs.webkit.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.



More information about the webkit-unassigned mailing list