"Max Bowsher" <maxb@ukf.net> writes:
> > (Config._prep_maps): Postponed the removal of map section names from
> > self._groups until all map specifications have been processed.
>
> Why is this necessary, given that self._groups is never tested during
> _prep_maps() ?
From Servatius's email:
When the [maps] section specifies the same mapping specification to be
used by more than one option (e.g. reply_to = [mail-addresses] and
to_addr = [mail-addresses]), the script fails. I postponed the removal
of a map section name from the self._groups list in Config._prep_maps()
until all map specifications have been processed.
Indeed:
% Traceback (most recent call last):
File "/home/cmpilato/projects/subversion/tools/hook-scripts/mailer/mailer.py", line 1271, in ?
sys.argv[3:3+expected_args])
File "/usr/local/lib/svn-python/svn/core.py", line 40, in run_app
return apply(func, (pool,) + args, kw)
File "/home/cmpilato/projects/subversion/tools/hook-scripts/mailer/mailer.py", line 55, in main
cfg = Config(config_fname, repos, { 'author' : repos.author })
File "/home/cmpilato/projects/subversion/tools/hook-scripts/mailer/mailer.py", line 1042, in __init__
self._prep_maps()
File "/home/cmpilato/projects/subversion/tools/hook-scripts/mailer/mailer.py", line 1108, in _prep_maps
self._groups.remove(sectname)
ValueError: list.remove(x): x not in list
> > (Config.get): Apply mappings on a per-word basis.
>
> Sure, this is fine for email addresses, but is this the right thing to
> do for *all* possible parameter types?
> And what about compatibility?
Oops. Yeah, that's badness on both points. I'll have to be reverting
that part of the change. Like... now.
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Received on Fri Jun 24 20:01:04 2005