Bug 26156 – In view-source mode, always render the contents using HTMLViewSourceDocument

• Component: Layout and Rendering
• Status: RESOLVED FIXED

[Takeshi Yoshino]

When in view-source mode, render the contents using HTMLViewSourceDocument
regardless it's applicable for any plugin or not.

Chromium tells WebCore to render the contents of specified URL when
view-source: prefix is added to the URL. But currently, DOMImplementation
ignores inViewSourceMode() when the MIME type is indicating that the contents
are neither texts nor HTML family documents.

For example, we can check the contents of asf file without launching media
player. Rendering contents for view-source:-specified input is not what user
expects.

http://code.google.com/p/chromium/issues/detail?id=10545

I want to fix this issue by this patch. IMHO, regardless of this Chromium
specific issue, I think we should force use of HTMLViewSourceDocument when
inViewSourceMode() is specified.

[Takeshi Yoshino]

Created attachment 30895 [details]
Proposed fix for 26156

[Takeshi Yoshino]

This might break Web Inspector. Now investigating...

[Takeshi Yoshino]

I didn't see any difference in the Web Inspector, but the trunk HEAD seems to have some problem with the Web Inspector. After fix, I'll request review again

[Takeshi Yoshino]

Created attachment 30979 [details]
Layout tests for the proposed patch for 26156

[Takeshi Yoshino]

Comment on attachment 30895 [details]
Proposed fix for 26156

Resuming review.

[Eric Seidel (no email)]

I'm not sure who is the right person to review this.  Probably Tim?

[Timothy Hatcher]

Dave Hyatt should review this, he created the view source mode.

[Eric Seidel (no email)]

Comment on attachment 30895 [details]
Proposed fix for 26156

Please provide a single patch including both the layout tests and the code change.  Also please incldue the -expected results for the test so the reviewer can verify that they look correct.

[Takeshi Yoshino]

Created attachment 31524 [details]
Proposed fix for 26156 (rev 2)

[Takeshi Yoshino]

(In reply to comment #8)
> (From update of attachment 30895 [details] [review])
> Please provide a single patch including both the layout tests and the code
> change.  Also please incldue the -expected results for the test so the reviewer
> can verify that they look correct.
> 

All right. Done.
Thanks,

[Dave Hyatt]

Comment on attachment 31524 [details]
Proposed fix for 26156 (rev 2)

r=me

[Takeshi Yoshino]

(In reply to comment #11)
> (From update of attachment 31524 [details] [review])
> r=me
> 

Thank you for review.
Could someone land this for me? I'm not a committer.

[David Levin]

Assign to levin for landing.

[David Levin]

Actually, I'm not landing this one right now... It looks like it will need baselines for other platforms, so it needs a little more time to handle than I can devote at the moment.

[Eric Seidel (no email)]

patching file WebCore/ChangeLog
Hunk #1 FAILED at 1.
1 out of 1 hunk FAILED -- saving rejects to file WebCore/ChangeLog.rej

I think something is wrong with svn-apply.  I would have expected it to handle the ChangeLog fine.

[Takeshi Yoshino]

Created attachment 31910 [details]
Proposed fix for 26156 (rev 3)

Resolved conflicts.

Additionally, replaced trailing CR.

[Takeshi Yoshino]

Please take a look again. No code change from revision 2. Just resolved conflicts.

[Eric Seidel (no email)]

Committing to http://svn.webkit.org/repository/webkit/trunk ...
	M	LayoutTests/ChangeLog
	A	LayoutTests/fast/frames/resources/viewsource-fake-image-file.png
	A	LayoutTests/fast/frames/viewsource-on-image-file.html
	A	LayoutTests/platform/mac/fast/frames/viewsource-on-image-file-expected.txt
	M	WebCore/ChangeLog
	M	WebCore/dom/DOMImplementation.cpp
	M	WebCore/html/HTMLViewSourceDocument.cpp
	M	WebCore/html/HTMLViewSourceDocument.h
Committed r45255
http://trac.webkit.org/changeset/45255