Chrome Frame test fixes for Vista/IE7.

Chrome Frame test fixes for Vista/IE7.

Write the ActiveX control from an external script in all test documents.  This works around a "feature" of IE versions released between April, 2006 and April, 2008 that required the user to click on an ActiveX control to "activate" it.  See http://msdn.microsoft.com/en-us/library/ms537508.aspx.

I also added asserts for the copying of CFInstance and CFInstall so that the whole test aborts if the files can't be copied, since otherwise test results are bogus.

BUG=60987
TEST=no manual testing needed; chrome_frame_tests.exe will either become better or worse.

Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=113291

[grt (UTC plus 2), 2011-12-02 21:29:17]

This is the first of several commits that I hope will make chrome_frame_tests a
bit more robust.  This change allows them to run on fresh machines with versions
of IE where a mouse click was required to "activate" a control on a page.

[robertshield, 2011-12-02 22:05:02]

lgtm, couple of nits

http://codereview.chromium.org/8777002/diff/2001/chrome_frame/test/data/chrom...
File chrome_frame/test/data/chrome_frame_tester_helpers.js (right):

http://codereview.chromium.org/8777002/diff/2001/chrome_frame/test/data/chrom...
chrome_frame/test/data/chrome_frame_tester_helpers.js:199: e.innerHTML =
innerHTML;
It seems at a glance that there's a fair amount of duplication in the html
passed in here (most of the callers pass most of the same text). Any way the
parts that change could become parameters and the rest moved here to shorten the
test pages?

http://codereview.chromium.org/8777002/diff/2001/chrome_frame/test/data/src_p...
File chrome_frame/test/data/src_property_host.html (right):

http://codereview.chromium.org/8777002/diff/2001/chrome_frame/test/data/src_p...
chrome_frame/test/data/src_property_host.html:37: '          <param
name="onload" value="return OnChromeFrameLoaded(arguments[0]);">' +
could you make these < 80 chars? maybe break the line before 'value'?

[grt (UTC plus 2), 2011-12-05 22:03:53]

PTAL

[ananta, 2011-12-05 22:45:02]

LGTM

[robertshield, 2011-12-06 19:57:27]

lgtm

[commit-bot: I haz the power, 2011-12-06 23:19:18]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/grt@chromium.org/8777002/10001

[commit-bot: I haz the power, 2011-12-06 23:19:20]

Can't apply patch for file chrome_frame/test/data/chrome_frame_resize.html.
While running patch -p1 --forward --force;
patching file chrome_frame/test/data/chrome_frame_resize.html
Hunk #1 FAILED at 117.
Hunk #2 FAILED at 135.
2 out of 2 hunks FAILED -- saving rejects to file
chrome_frame/test/data/chrome_frame_resize.html.rej

[commit-bot: I haz the power, 2011-12-06 23:42:35]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/grt@chromium.org/8777002/15002

[commit-bot: I haz the power, 2011-12-06 23:42:40]

Can't apply patch for file chrome_frame/test/data/chrome_frame_resize.html.
While running patch -p1 --forward --force;
patching file chrome_frame/test/data/chrome_frame_resize.html
Hunk #1 FAILED at 117.
Hunk #2 FAILED at 135.
2 out of 2 hunks FAILED -- saving rejects to file
chrome_frame/test/data/chrome_frame_resize.html.rej

[Ilya Sherman, 2011-12-07 00:18:40]

I think you got the wrong bug number for this CL, btw ;)

[grt (UTC plus 2), 2011-12-07 01:26:08]

On 2011/12/07 00:18:40, Ilya Sherman wrote:
> I think you got the wrong bug number for this CL, btw ;)

I'm considering 60987 to be a general "the GCF tests are too flaky on
Vista/IE7."  This is the first of four or so CLs that I hope will make the tests
more reliable on Vista.  I thought it would be simpler (and not too confusing)
to associate the CLs with this bug since, in the end, I hope to re-enable all
tests that have been disabled and assigned to this bug.  I'll try to make this
more clear in the description of the next CL.  Let me know if you have any other
suggestions.

Good lookin' out, by the way!

[Ilya Sherman, 2011-12-07 01:34:21]

On 2011/12/07 01:26:08, grt wrote:
> On 2011/12/07 00:18:40, Ilya Sherman wrote:
> > I think you got the wrong bug number for this CL, btw ;)
> 
> I'm considering 60987 to be a general "the GCF tests are too flaky on
> Vista/IE7."  This is the first of four or so CLs that I hope will make the
tests
> more reliable on Vista.  I thought it would be simpler (and not too confusing)
> to associate the CLs with this bug since, in the end, I hope to re-enable all
> tests that have been disabled and assigned to this bug.  I'll try to make this
> more clear in the description of the next CL.  Let me know if you have any
other
> suggestions.
> 
> Good lookin' out, by the way!

Oh, hmm.  What I actually meant was that
http://code.google.com/p/chromium/issues/detail?id=2006#c10 looked unrelated to
this CL.  I'm not sure how bugdroid settled upon that bug #, though.  I guess
it's because you listed the year "2006" in the bug description???  Silly
bugdroid.