[webkit-reviews] review denied: [Bug 85368] [webkit-patch] Add new chrome-channels to track down current chrome release channels for a committed bug. : [Attachment 139793] Patch

[bugzilla-daemon at webkit.org]

Dirk Pranke <dpranke at chromium.org> has denied Gavin Peters
<gavinp at chromium.org>'s request for review:
Bug 85368: [webkit-patch] Add new chrome-channels to track down current chrome
release channels for a committed bug.
https://bugs.webkit.org/show_bug.cgi?id=85368

Attachment 139793: Patch
https://bugs.webkit.org/attachment.cgi?id=139793&action=review

------- Additional Comments from Dirk Pranke <dpranke at chromium.org>
View in context: https://bugs.webkit.org/attachment.cgi?id=139793&action=review


We typically require at least *some* form of test for anything that gets
checked in, even if it is a "scripting hack" :)

> Tools/Scripts/webkitpy/common/net/omahaproxy.py:67
> +		       row = (int(version['base_webkit_revision']),
version['channel'], platform['os'], version['date'])

Should this be a dict or an object instead of a tuple?

> Tools/Scripts/webkitpy/tool/commands/chromechannels.py:38
> +    help_text = "Lists which chrome channels the commit associated with a
bug is included in."

I'm actually unclear on this ... is it figuring out when a bugfix is
introduced, when a bug is introduced, or is really just querying revision
numbers regardless of what the change in the revision was for?

Also, it looks like the usage is "webkit-patch chrome-channels bug" ... can we
mention that bugno is a required argument somewhere and whether it is a bug
number, or a URL to the bug, or ... ?

What happens if there were multiple commits for a single bug number? It's rare,
but it certainly happens ... do you look for channels that have any of the
commits? all of the commits? etc.