[chromium-dev] Re: Chrome Keyboard Access, opinions?

2009-10-09 Thread Jay Campan
> +1.  To a beginner, left and right arrow might be more intuitive and an > opportunity for us to innovate.  But millions of people use screenreaders, > have trouble using the mouse, or are just power users who love keyboard > shortcuts, and we're just frustrating them by not letting them use stan

[chromium-dev] Re: Chrome Keyboard Access, opinions?

2009-10-09 Thread Jay Campan
On Thu, Oct 8, 2009 at 6:32 PM, Finnur Thorarinsson wrote: > Maybe it's just me, but I don't see the point in making a toolbar as a > whole focusable. The keyboard shortcut should put focus on the first element > in the toolbar and tab should cycle focus from there. > The goal was to have a mode w

[chromium-dev] Build problems on Windows: native client missing headers

2009-09-30 Thread Jay Campan
If you get build errors on Windows about some native client generated header file missing, make sure to run the sync command from cmd.exe, not from cygwin bash. (or just run "gclient runhook --force" from cmd.exe) There is a bug (http://code.google.com/p/nativeclient/issues/detail?id=105) when ge

[chromium-dev] Re: browser_tests hang on Win7

2009-09-28 Thread Jay Campan
That OleInitialize/Uninitialize business is there to prevent unbalanced Ole init calls in some cases. I did not realize DllMain was called with the loader lock held. Looks like it could cause deadlocks. I'll make move that code to a separate function invoked before the DLL is unloaded . Jay On M

[chromium-dev] Re: Dashboard to show which UI/unit/etc (non-webkit) tests are slowest

2009-09-17 Thread Jay Campan
> I don't think the focus tests need as much a delay between tab > presses, do they Jay? I have actually a CL almost ready to go that uses a notification instead of waiting for the tab press to trigger a focus change. Takes the test time from 20s to 1s. That should fix both focus related tests. J

[chromium-dev] Re: How do you perform tab traversal?

2009-09-12 Thread Jay Campan
Your code looks good. Could it be that you are not running the message loop with an AcceleratorHandler instance? (you did not include the code that does that) The AcceleratorHandler deals with accelerator but is also intercepting tab key messages to handle tab traversal. Something like: Accelerat

[chromium-dev] Re: how to induce flakiness, episode 1326: misusing IN_PROC_BROWSER_TEST

2009-09-02 Thread Jay Campan
(resending as it was not sent to chromium-dev) I think we can have a unique launcher for the interactive ui tests, the browser tests and the sync integration tests. The launcher would run each test isolated, by way of process fork (Linux and Mac) or DLL unload/reload (Windows), just as it is done