[MERGE] bzr rm should delete the working file (Bug #82602)

[Marius Kruger]

hi,
On 4/19/07, John Arbash Meinel <john at arbash-meinel.com> wrote:
>
> The only part I don't like is the "magic" for taking
>
> self.run_bzr('command to run')
>
> There are a few issues... The biggest being that shlex.split() doesn't
> support Unicode arguments.


all the tests pass, and if someone have a unique test and we cant fix it
by encoding/decoding (like we did in command .get_alias ) they can
just fall back to self.run_bzr('command', 'to', 'run')

I'm not sure if the others caught that you changed this, but it is
> something that we haven't done (in favor of being very explicit about
> what arguments are being passed by run_bzr).

 Obviously, I don't think its that a bad an idea.

I would rather something like this wasn't "snuck-in" without discussion
> on a merge for "bzr rm".
>
(I'm not strictly opposed to the change, though I'm a little negative on
> it. I am opposed to it being merged without discussion).


agreed :)
The reason I made it like this is because runbzr did it, and I found
simplified my tests, but you didn't like me using it.
If you would like I can move this magic to my TestCase
and I can resubmit moving it into run_bzr later for more descussion.




-- 

I don't trust a Revision Control System with less than 5900 unit tests.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://lists.ubuntu.com/archives/bazaar/attachments/20070420/13722dd6/attachment.htm