Add a method for setting the value of an Accessibility object....

Add a method for setting the value of an Accessibility object.
BUG=none
TEST=none

Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=56867

[amit, 2010-08-18 17:28:01]

drive by lgtm

http://codereview.chromium.org/3156026/diff/17001/18004
File chrome_frame/test/ui_test.cc (right):

http://codereview.chromium.org/3156026/diff/17001/18004#newcode446
chrome_frame/test/ui_test.cc:446: TEST_F(ContextMenuTest, FLAKY_CFSaveAs) {
Nice, so this test is ready to loose its flaky stigma?

[kkania, 2010-08-18 17:34:27]

On 2010/08/18 17:28:01, amit wrote:
> drive by lgtm
> 
> http://codereview.chromium.org/3156026/diff/17001/18004
> File chrome_frame/test/ui_test.cc (right):
> 
> http://codereview.chromium.org/3156026/diff/17001/18004#newcode446
> chrome_frame/test/ui_test.cc:446: TEST_F(ContextMenuTest, FLAKY_CFSaveAs) {
> Nice, so this test is ready to loose its flaky stigma?

I think it should remain flaky because it relies on a delay for waiting for the
file to be downloaded. Do you think it is worth it to wait in the loop and
periodically check if the file exists? Or is there some easier way to hook into
being notified when the download is done? I don't get any helpful events from
DWebBrowserEvents...

[stoyan, 2010-08-18 17:38:24]

LGTM

http://codereview.chromium.org/3156026/diff/17001/18001
File chrome_frame/test/chrome_frame_ui_test_utils.cc (right):

http://codereview.chromium.org/3156026/diff/17001/18001#newcode263
chrome_frame/test/chrome_frame_ui_test_utils.cc:263: result == E_INVALIDARG) {
MSDN says E_INVALIDARG may be returned if  varChildID contains VT_EMPTY, but we
have just checked it is VT_I4. So E_INVALIDARG seems like genuine error here.

[amit, 2010-08-18 17:48:23]

On Wed, Aug 18, 2010 at 10:34 AM, <kkania@chromium.org> wrote:

> On 2010/08/18 17:28:01, amit wrote:
>
>> drive by lgtm
>>
>
>  http://codereview.chromium.org/3156026/diff/17001/18004
>> File chrome_frame/test/ui_test.cc (right):
>>
>
>  http://codereview.chromium.org/3156026/diff/17001/18004#newcode446
>> chrome_frame/test/ui_test.cc:446: TEST_F(ContextMenuTest, FLAKY_CFSaveAs)
>> {
>> Nice, so this test is ready to loose its flaky stigma?
>>
>
> I think it should remain flaky because it relies on a delay for waiting for
> the
> file to be downloaded. Do you think it is worth it to wait in the loop and
> periodically check if the file exists? Or is there some easier way to hook
> into
> being notified when the download is done? I don't get any helpful events
> from
> DWebBrowserEvents...


File change notification API's could be used.
http://msdn.microsoft.com/en-us/library/aa365261(VS.85).aspx

>
>
> http://codereview.chromium.org/3156026/show
>

[kkania, 2010-08-19 02:15:27]

Added observer to watch for downloads.

[kkania, 2010-08-19 16:59:21]

On 2010/08/19 02:15:27, kkania wrote:
> Added observer to watch for downloads.

Please review again when time allows.

[amit, 2010-08-19 21:41:03]

lgtm

http://codereview.chromium.org/3156026/diff/35001/36001
File chrome_frame/test/chrome_frame_test_utils.cc (right):

http://codereview.chromium.org/3156026/diff/35001/36001#newcode984
chrome_frame/test/chrome_frame_test_utils.cc:984: while
(!file_util::GetFileSize(download, &downloaded_size) ||
instead of implementing a loop you could use RegisterWaitForSingleObject here. 
BTW, stoyan's idea of waiting for a second is OK since this is mainly for URLs
already in wininet cache.