Bug 60027 – webkit-patch post-commits HEAD^.. throws an exception

• Component: Tools / Tests
• Status: RESOLVED FIXED

[Adam Roben (:aroben)]

webkit-patch post-commits HEAD^.. throws an exception:

[Adam Roben@macpro8 OpenSource]\$ webkit-patch post-commits HEAD^..
Traceback (most recent call last):
  File "/home/Adam Roben/dev/WebKit/OpenSource/Tools/Scripts/webkit-patch", line 71, in <module>
    main()
  File "/home/Adam Roben/dev/WebKit/OpenSource/Tools/Scripts/webkit-patch", line 66, in main
    WebKitPatch(__file__).main()
  File "/home/Adam Roben/dev/WebKit/OpenSource/Tools/Scripts/webkitpy/tool/multicommandtool.py", line 308, in main
    result = command.check_arguments_and_execute(options, args, self)
  File "/home/Adam Roben/dev/WebKit/OpenSource/Tools/Scripts/webkitpy/tool/multicommandtool.py", line 117, in check_arguments_and_execute
    return self.execute(options, args, tool) or 0
  File "/home/Adam Roben/dev/WebKit/OpenSource/Tools/Scripts/webkitpy/tool/commands/upload.py", line 326, in execute
    commit_message = tool.scm().commit_message_for_local_commit(commit_id)
  File "/home/Adam Roben/dev/WebKit/OpenSource/Tools/Scripts/webkitpy/common/checkout/scm/git.py", line 442, in commit_message_for_local_commit
    return CommitMessage(commit_lines[first_line_after_headers:])
NameError: global name 'CommitMessage' is not defined

I don't know when this started happening.

[Adam Roben (:aroben)]

Created attachment 92076 [details]
Patch

[Adam Roben (:aroben)]

Comment on attachment 92076 [details]
Patch

This is the patch I was trying to upload.

[Eric Seidel (no email)]

We need to import CommitMessage into git.py.  Sorry. :(

[Eric Seidel (no email)]

Created attachment 92085 [details]
Patch

[Adam Roben (:aroben)]

Comment on attachment 92085 [details]
Patch

Any way to write a test?

[Eric Seidel (no email)]

There certainly are.  I just wrote this quickly on my train ride to work.  The scm tests take like 30m to run... (I'm working on fixing that), so I didn't bother.

[Adam Barth]

Eric, do you have a plan to fix all these regressions?  There have been a handful of them at this point.

[Eric Seidel (no email)]

Yes, I fixed this one this morning.  I went looking for other bugs on the train but didn't find them.  I think we have one other on file?  Do you know of more?  I'm certainly happy to fix these imports and add associated unit tests so we don't hit them again.  Much of this pain is due to the scm.py tests being too slow to run (and thus they've slowly gotten worse and worse coverage over time and fail to catch these sorts of things).

[Adam Barth]

https://bugs.webkit.org/show_bug.cgi?id=59980 for example.

Lack of testing and slowness of testing is certainly one reason you caused these regressions.  However, it would be useful to read all the code carefully to make sure you're importing everything that you need.  Otherwise, we'll get a long tail of these sorts of bug reports.

[Eric Seidel (no email)]

(In reply to comment #9)
> https://bugs.webkit.org/show_bug.cgi?id=59980 for example.
> 
> Lack of testing and slowness of testing is certainly one reason you caused these regressions.  However, it would be useful to read all the code carefully to make sure you're importing everything that you need.  Otherwise, we'll get a long tail of these sorts of bug reports.

My apologies for the regressions.

[WebKit Commit Bot]

Comment on attachment 92085 [details]
Patch

Clearing flags on attachment: 92085

Committed r85672: <http://trac.webkit.org/changeset/85672>

[WebKit Commit Bot]

All reviewed patches have been landed.  Closing bug.