1. TortoiseHg
  2. TortoiseHg
  3. thg
  4. Issues
Issue #202 resolved

Revision history Python exception

Anonymous created an issue

{{{

!python

Please report this bug to http://bitbucket.org/tortoisehg/thg/issues Mercurial version (1.8). TortoiseHg version (2.0) Command: --nofork log --listfile C:\DOCUME~1\neilk\LOCALS~1\Temp\THGA.tmp CWD: C:\hgsource\SKDIERemoteHandler Extensions loaded: fixfrozenexts Python version: 2.6.6 (r266:84297, Aug 24 2010, 18:46:32) [MSC v.1500 32 bit (Intel)] Windows version: (5, 1, 2600, 2, 'Service Pack 3') Processor architecture: x86 ** Qt-4.7.1 PyQt-4.8.2 Traceback (most recent call last): File "tortoisehg\hgqt\run.pyo", line 457, in call File "tortoisehg\hgqt\workbench.pyo", line 707, in run File "tortoisehg\hgqt\filedialogs.pyo", line 105, in init File "tortoisehg\hgqt\filedialogs.pyo", line 118, in _readSettings File "tortoisehg\hgqt\csinfo.pyo", line 421, in set_expanded File "tortoisehg\hgqt\csinfo.pyo", line 405, in update File "tortoisehg\hgqt\csinfo.pyo", line 318, in get_markup File "tortoisehg\hgqt\csinfo.pyo", line 264, in get_markup File "tortoisehg\hgqt\csinfo.pyo", line 211, in get_data File "tortoisehg\hgqt\csinfo.pyo", line 141, in preset_func File "mercurial\context.pyo", line 693, in user File "mercurial\util.pyo", line 169, in get File "mercurial\context.pyo", line 660, in _user File "mercurial\ui.pyo", line 315, in username Abort: no username supplied (see "hg help config")

}}}

Comments (6)

  1. Adrian Buehlmann
    • changed status to open

    That's interesting.

    There's a changeset with an empty username *and* no username has been defined in the settings.

    What's puzzling for me is: why is mercurial/context.py, line 692, returning the config's username for a committed changeset with an empty username:

        def user(self):
            return self._user or self._repo.ui.username()
    

    I'll ponder this and probably file a bug on mercurial or ask on mercurial-devel.

  2. Adrian Buehlmann

    Weird. hgqt/csinfo.py tries to create a csinfo object for the working context, which fails if no username has been configured.

    This happens on "thg log <filename>" (as per the traceback reported and I can repro it if the csinfo panel is expanded in the FileLogDialog).

    We probably shouldn't allow to have csinfo objects for the working context.

    Sure, a username needs to be configured, but only until the user commits. Exploring history should be possible without setting a username.

  3. Log in to comment