[Webkit-unassigned] [Bug 39624] --squash should go away and become the default

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=39624





--- Comment #36 from Ojan Vafai <ojan at chromium.org>  2010-07-09 12:15:32 PST ---
(In reply to comment #35)
> (From update of attachment 60408 [details])
> Can we be more user friendly during the transition?  Maybe we should understand --no-squash and print a message that explains the new way is --git-commit=HEAD.. ?  Another option is to automatically re-write --no-squash into --git-commit=HEAD..  Some minor comments below.

Good idea. I put in --squash and --no-squash. I made them error out with a descriptive message. I'd like to remove these in a couple weeks once people have adjusted.

> WebKitTools/Scripts/webkitpy/tool/steps/checkstyle_unittest.py:38
>  +      pass
> Should we delete this file if there aren't any tests in it?

Done.

> WebKitTools/Scripts/webkitpy/tool/steps/commit.py:68
>  +              username = User.prompt("%s login: " % e.server_host, repeat=5)
> You should use self._tool.user here.

Done.

> WebKitTools/Scripts/webkitpy/tool/steps/commit.py:71
>  +              self._commit(username=username)
> What if this throws an AmbiguousCommitError?  I think you want to use some sort of loop with these things as state variables.  You can put a retry limit in the loop if you're worried about going infinite.

This isn't possible right now. Only the SVN impl of scm fires AuthenticationError and only the git impl fires AmbiguousCommitError. The former could definitely change though. Added a try/except.

> WebKitTools/ChangeLog:9
>  +          need to be met by a wrapper script around webkit-patch.
> This isn't really true anymore, right?  I can use --git-commit=HEAD..

No, there are still people who want to do setup local commits and have the script act on all the commits at once (e.g. commit them all with a single command). Made the description a bit more accurate.

-- 
Configure bugmail: https://bugs.webkit.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.