2009/5/30 Martitza <email address hidden>:
> If I can find some time to do a proper job I may look at this. I can
> imagine that the separator may be hardcoded or cannonicalized to the unix
> form in very many places. Just finding them might take a while.
>
> And besides passing the existing tests, should there be any new tests?
If you add a new option you'll need new tests. Otherwise, the
existing tests may need to be parameterized to check paths using an
abstracted function.
> Also I can imagine people have written scripts which depend on existing
> behavior and might want a bzr.config option to control how output is formed:
>
> path_sep_for_output = mode
>
> with mode being "slash" or "backslash" or "native" where "native" would be
> the default to use os.sep
2009/5/30 Martitza <email address hidden>:
> If I can find some time to do a proper job I may look at this. I can
> imagine that the separator may be hardcoded or cannonicalized to the unix
> form in very many places. Just finding them might take a while.
>
> And besides passing the existing tests, should there be any new tests?
If you add a new option you'll need new tests. Otherwise, the
existing tests may need to be parameterized to check paths using an
abstracted function.
> Also I can imagine people have written scripts which depend on existing
> behavior and might want a bzr.config option to control how output is formed:
>
> path_sep_for_output = mode
>
> with mode being "slash" or "backslash" or "native" where "native" would be
> the default to use os.sep
That would be nice. launchpad. net/~mbp/>
--
Martin <http://