RESOLVED FIXED 91560
"webkit-patch rebaseline <test>" fails with exit_code 1
https://bugs.webkit.org/show_bug.cgi?id=91560
Summary "webkit-patch rebaseline <test>" fails with exit_code 1
Emil A Eklund
Reported 2012-07-17 17:15:06 PDT
$ ./Tools/Scripts/webkit-patch rebaseline css2.1/t100801-c544-valgn-03-d-agi.html Which builder to pull results from: build.webkit.org bots 1. Apple Lion Debug WK1 (Tests) 2. Apple Lion Debug WK2 (Tests) 3. Apple Lion Release WK1 (Tests) 4. Apple Lion Release WK2 (Tests) 5. Apple Win 7 Release (Tests) 6. Apple Win XP Debug (Tests) 7. EFL Linux 64-bit Debug 8. EFL Linux 64-bit Release 9. GTK Linux 32-bit Release 10. GTK Linux 64-bit Debug 11. GTK Linux 64-bit Release 12. GTK Linux 64-bit Release WK2 (Tests) 13. Qt Linux Release build.chromium.org bots 14. Webkit Linux 15. Webkit Linux (dbg) 16. Webkit Linux 32 17. Webkit Mac10.6 18. Webkit Mac10.6 (dbg) 19. Webkit Mac10.7 20. Webkit Win 21. Webkit Win (dbg)(1) 22. Webkit Win (dbg)(2) 23. Webkit Win7 Enter one or more numbers (comma-separated) or ranges (e.g. 3-7), or "all": 18 Traceback (most recent call last): File "./Tools/Scripts/webkit-patch", line 84, in <module> main() File "./Tools/Scripts/webkit-patch", line 79, in main WebKitPatch(os.path.abspath(__file__)).main() File "/usr/local/google/projects/webkit3/Tools/Scripts/webkitpy/tool/multicommandtool.py", line 311, in main result = command.check_arguments_and_execute(options, args, self) File "/usr/local/google/projects/webkit3/Tools/Scripts/webkitpy/tool/multicommandtool.py", line 120, in check_arguments_and_execute return self.execute(options, args, tool) or 0 File "/usr/local/google/projects/webkit3/Tools/Scripts/webkitpy/tool/commands/rebaseline.py", line 426, in execute self._rebaseline(options, test_list) File "/usr/local/google/projects/webkit3/Tools/Scripts/webkitpy/tool/commands/rebaseline.py", line 305, in _rebaseline self._tool.scm().add_list(list(files_to_add)) File "/usr/local/google/projects/webkit3/Tools/Scripts/webkitpy/common/checkout/scm/svn.py", line 180, in add_list return self._run_svn(["add"] + paths, return_exit_code=return_exit_code) File "/usr/local/google/projects/webkit3/Tools/Scripts/webkitpy/common/checkout/scm/svn.py", line 131, in _run_svn return self.run([self.executable_name] + args, **kwargs) File "/usr/local/google/projects/webkit3/Tools/Scripts/webkitpy/common/checkout/scm/scm.py", line 76, in run decode_output=decode_output) File "/usr/local/google/projects/webkit3/Tools/Scripts/webkitpy/common/system/executive.py", line 418, in run_command (error_handler or self.default_error_handler)(script_error) File "/usr/local/google/projects/webkit3/Tools/Scripts/webkitpy/common/system/executive.py", line 335, in default_error_handler raise error webkitpy.common.system.executive.ScriptError: Failed to run "[u'svn', u'add']" exit_code: 1
Attachments
Patch (3.21 KB, patch)
2012-07-27 17:10 PDT, Dirk Pranke
abarth: review+
Ojan Vafai
Comment 1 2012-07-17 18:16:41 PDT
Are you running this on Windows? Wouldn't surprise me if this just didn't work on Windows.
Emil A Eklund
Comment 2 2012-07-18 09:54:13 PDT
No, linux. Worked up until a week or two ago.
Ojan Vafai
Comment 3 2012-07-18 10:28:25 PDT
Can you run with -v? That should print out the rebaseline-json command it's using.
Emil A Eklund
Comment 4 2012-07-18 10:30:27 PDT
rebaseline-json: {'css2.1/t100801-c544-valgn-03-d-agi.html': {'Webkit Mac10.6 (dbg)': ['wav', 'txt', 'png']}} webkitpy.tool.commands.rebaseline: [DEBUG] "" is not a JSON object, ignoring webkitpy.common.system.executive: [DEBUG] "svn add" took 0.03s Traceback (most recent call last): ... same stack trace as before ... webkitpy.common.system.executive.ScriptError: Failed to run "[u'svn', u'add']" exit_code: 1
Ojan Vafai
Comment 5 2012-07-18 11:25:17 PDT
Works for me on Linux. :( Are you on precise perhaps?
Emil A Eklund
Comment 6 2012-07-18 11:34:36 PDT
Nope, ghardy.
Ojan Vafai
Comment 7 2012-07-18 11:50:24 PDT
I'm on Lucid. Maybe it's your python version? I'm on 2.6.5.
Emil A Eklund
Comment 8 2012-07-18 11:55:32 PDT
$ more /proc/version Linux version 2.6.38.8-gg868 $ python --version Python 2.6.5 This all worked up until a week or two ago and I haven't upgraded or changed any part of my system config since.
Emil A Eklund
Comment 9 2012-07-18 11:59:27 PDT
Is there any way to see what command the script tries to execute when it encounters the error?
Dirk Pranke
Comment 10 2012-07-18 13:01:42 PDT
(In reply to comment #9) > Is there any way to see what command the script tries to execute when it encounters the error? webkit-patch --verbose should give you a little more info. Also, this is probably a dumb question, but you are in a svn checkout and not a git checkout, right?
Emil A Eklund
Comment 11 2012-07-18 13:07:13 PDT
Yeah, webkit svn checkout with the chromium data (update-webkit --chromium).
Dirk Pranke
Comment 12 2012-07-18 13:10:29 PDT
(In reply to comment #11) > Yeah, webkit svn checkout with the chromium data (update-webkit --chromium). turns out I no longer had an svn checkout on linux, so I'm recreating one now and will attempt to reproduce once I'm there ...
Dirk Pranke
Comment 13 2012-07-18 13:11:59 PDT
oop, nevermind, I reproduced this on mac. Will investigate ...
Dirk Pranke
Comment 14 2012-07-18 13:21:07 PDT
yeah, that's what I thought. We're trying to add nothing (an empty list of files), which is why all you're seeing is "svn add" :). It looks like, at least as of now, that test isn't failing anywhere, so we have no work to do, and we don't trap that properly. It's a simple fix to not try to add nothing; however, at a higher level, it would be nice if the tool was reporting that there was nothing to do as the test wasn't failing. Also, it's not clear to me that I'm reproducing what you were seeing either, since maybe the test was failing for you at the time? Lastly, I'm concerned that we're listing the debug bots as well as the release bots; we shouldn't be, since we don't want to add the debug versions if there are release versions.
Emil A Eklund
Comment 15 2012-07-18 13:28:27 PDT
Probably a different failure now than before then as the test was failing at the time. Will report back next time I encounter this error with a test that is failing. Thanks!
Dirk Pranke
Comment 16 2012-07-18 13:34:16 PDT
(In reply to comment #15) > Probably a different failure now than before then as the test was failing at the time. Will report back next time I encounter this error with a test that is failing. > > Thanks! My guess is that it had something to do with you trying to use a debug bot and us ending up throwing away the baseline so you still weren't changing anything. until we figure out how to strip the debug bots out of this, don't try to rebaseline from a debug bot if there is a release equivalent :).
Dirk Pranke
Comment 17 2012-07-27 17:10:00 PDT
Dirk Pranke
Comment 18 2012-07-27 17:11:13 PDT
*** Bug 92559 has been marked as a duplicate of this bug. ***
Dirk Pranke
Comment 19 2012-07-27 17:35:04 PDT
Note You need to log in before you can comment on or make changes to this bug.