No blaming! I actually thought what person A had done was right, and
I almost did the same thing a day ago (Darin noticed and suggested I
use a default argument on a function to make it so the code would work
both with and without my patch).
Do we have docs on what the proper thing to do is?
In composing this email I found "Landing Two-sided Patches" in this document:
http://dev.chromium.org/developers/how-tos/webkit-merge-1
but I'd never read it because I haven't ever done a WebKit merge.
Is that the rule everyone who commits to WebKit should have known about?
--
Chromium Developers mailing list: chromi...@chromium.org
View archives, change email options, or unsubscribe:
http://groups.google.com/a/chromium.org/group/chromium-dev
On Fri, Feb 19, 2010 at 10:46 AM, Stephen White
<senor...@chromium.org> wrote:
> I think the main thing is, if you're landing a two-sided patch, give the
> WebKit gardener a heads-up. Then there should be no unexpected reverting.
> One thing I've done in the past is to hide the WebKit changes behind an
> #ifdef, so that it works either with or without the Chrome-side change.
> Then once the merge brings it to Chrome, land the Chrome-side and #define
> the symbol in the same CL. Then, if all the bots are happy, land a WebKit
> change to remove the old code and the #ifdef. That way, if there are
> problems, you can revert just the Chrome-side change without needing an
> emergency WebKit merge.
> Granted, this is more work, and is only practical for small to medium-sized
> changes. But it's probably the safest.
Where do you end up defining the macro in Chrome?
You have to click on an individual calendar entry and see the tiny
text where it says who is invited to that meeting.
Brett
--