Bug 53462 – Using Control-C to cancel webkit-patch or other python tools causes unpleasant traceback console spew

• Component: Tools / Tests
• Status: RESOLVED FIXED

[Darin Adler]

For example, I hit Control-C when webkit-patch was asking me about a diff

Was that diff correct? [Y/n]: Traceback (most recent call last):
  File "/Users/darin/Safari/OpenSource/Tools/Scripts/webkit-patch", line 70, in <module>
    main()
  File "/Users/darin/Safari/OpenSource/Tools/Scripts/webkit-patch", line 65, in main
    WebKitPatch(__file__).main()
  File "/Volumes/Home/darin/Safari/OpenSource/Tools/Scripts/webkitpy/tool/multicommandtool.py", line 308, in main
    result = command.check_arguments_and_execute(options, args, self)
  File "/Volumes/Home/darin/Safari/OpenSource/Tools/Scripts/webkitpy/tool/multicommandtool.py", line 117, in check_arguments_and_execute
    return self.execute(options, args, tool) or 0
  File "/Volumes/Home/darin/Safari/OpenSource/Tools/Scripts/webkitpy/tool/commands/abstractsequencedcommand.py", line 51, in execute
    self._sequence.run_and_handle_errors(tool, options, state)
  File "/Volumes/Home/darin/Safari/OpenSource/Tools/Scripts/webkitpy/tool/commands/stepsequence.py", line 70, in run_and_handle_errors
    self._run(tool, options, state)
  File "/Volumes/Home/darin/Safari/OpenSource/Tools/Scripts/webkitpy/tool/commands/stepsequence.py", line 64, in _run
    step(tool, options).run(state)
  File "/Volumes/Home/darin/Safari/OpenSource/Tools/Scripts/webkitpy/tool/steps/confirmdiff.py", line 73, in run
    diff_correct = self._tool.user.confirm("Was that diff correct?")
  File "/Volumes/Home/darin/Safari/OpenSource/Tools/Scripts/webkitpy/common/system/user.py", line 128, in confirm
    response = raw_input("%s [%s]: " % (message, choice))
KeyboardInterrupt

Yuck!

Who else should I cc on this bug?

[Eric Seidel (no email)]

We could catch these, sure.  For now we've enjoyed the default handler's print-the-stacktrace behavior.

[Darin Adler]

Since I’m using the scripts, not developing them, the traceback is never useful to me.

[Eric Seidel (no email)]

Yes, I think you are correct that these scripts have moved from a development stage to being deployed and used and we should get rid of debug messages like these.  I agree with this bug completely.

[Adam Barth]

Created attachment 80793 [details]
Patch

[Mihai Parparita]

Comment on attachment 80793 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=80793&action=review

> Tools/Scripts/webkit-patch:72
> +        pass

Exiting with a non-0 exit code seems better (see http://trac.webkit.org/browser/trunk/Tools/Scripts/webkitpy/layout_tests/run_webkit_tests.py#L440)

[Adam Barth]

Created attachment 80797 [details]
Patch

[Eric Seidel (no email)]

Comment on attachment 80797 [details]
Patch

I would hav elogged "user interupted".  I might even put this in the multicommandtool.py instead.

Also, isn't there a python helper for the 128 + bit?

I don't think this should exit 0 though, so you're right in returning the signal code.

[WebKit Commit Bot]

Comment on attachment 80797 [details]
Patch

Clearing flags on attachment: 80797

Committed r77275: <http://trac.webkit.org/changeset/77275>

[WebKit Commit Bot]

All reviewed patches have been landed.  Closing bug.