Bug 7281 – With java disabled wrong information is displayed.

• Component: Java
• Status: RESOLVED FIXED

[Matthew McGillis]

Safari does not display proper content for applets when java is disabled. Safari fails test case specified in URL.

See instructions in URL for steps to perform test.

Safari should show both the text from the alt value of the applet tag as well as the text between the begin and end applet tags. However neither get displayed.

[Joost de Valk (AlthA)]

Confirmed, this needs to be fixed.

[Rob Buis]

Created attachment 8562 [details]
First attempt

After discussion with Mitz, it seems like this is the best thing to do. Mozilla
adds also the alt attribute contents (see testcase), WinIE does not. If the code
gets okayed, I can do a more complete patch with RenderEmptyApplet removal
and if needed a testcase.
Cheers,

Rob.

[Maciej Stachowiak]

Change looks good to me conceptually. Please post a version that includes a test case and removes the dead code.

[Rob Buis]

Created attachment 8664 [details]
Now with testcase

With a testcase and RenderEmptyApplet removed.
Cheers,

Rob.

[Maciej Stachowiak]

Comment on attachment 8664 [details]
Now with testcase

sweet, r=me

[Geoffrey Garen]

Does DumpRenderTree run with Java disabled? If it doesn't, this test won't work.

[Rob Buis]

Hi Geoff,

(In reply to comment #6)
> Does DumpRenderTree run with Java disabled? If it doesn't, this test won't
> work.

I think it does, I enabled java after a run-safari, then did DumpRenderTree
and it still generated the same.
Cheers,

Rob.

[David Kilzer (:ddkilzer)]

(In reply to comment #6)
> Does DumpRenderTree run with Java disabled? If it doesn't, this test won't
> work.

Yes, yes it does.

http://trac.webkit.org/dev/browser/trunk/WebKitTools/DumpRenderTree/DumpRenderTree.m#L234

[David Kilzer (:ddkilzer)]

(In reply to comment #5)
> (From update of attachment 8664 [details] [edit])
> sweet, r=me

New committer questions:

1. This patch doesn't change any build files (mac, win, bake) for the removed files.  Can I simply update those files as I see fit when committing?  (I can only test the mac build now.)

2. The LayoutTests/tables/mozilla_expected_failures/bugs/bug8499.html test is affected by this change (will fail if the patch is committed) since it has an applet in it, but the patch doesn't cover this change.  Should I update the test results (because it's an expected failure and my visual inspection of the results looks okay) and commit without further approval?

I've decided to hold off committing this patch until I get more instructions.

[Darin Adler]

(In reply to comment #9)
> 1. This patch doesn't change any build files (mac, win, bake) for the removed
> files.  Can I simply update those files as I see fit when committing?  (I can
> only test the mac build now.)

Yes.

> 2. The LayoutTests/tables/mozilla_expected_failures/bugs/bug8499.html test is
> affected by this change (will fail if the patch is committed) since it has an
> applet in it, but the patch doesn't cover this change.  Should I update the
> test results (because it's an expected failure and my visual inspection of the
> results looks okay) and commit without further approval?

Yes.

[David Kilzer (:ddkilzer)]

Committed revision 14715.