[darcs-users] darcs patch: Do not remove temporary directories if D... (and 1 more)

David Roundy droundy at darcs.net
Sat Oct 11 11:45:21 UTC 2008


On Sat, Oct 11, 2008 at 03:35:45PM +0400, Dmitry Kurochkin wrote:
> On Sat, Oct 11, 2008 at 3:24 PM, David Roundy <droundy at darcs.net> wrote:
> > On Sat, Oct 11, 2008 at 04:16:08AM -0700, Dmitry Kurochkin wrote:
> >> Sat Oct 11 14:58:08 MSD 2008  Dmitry Kurochkin <dmitry.kurochkin at gmail.com>
> >>   * Do not remove temporary directories if DARCS_KEEP_TMPDIR environment variable is set.
> >>
> >> Sat Oct 11 15:06:39 MSD 2008  Dmitry Kurochkin <dmitry.kurochkin at gmail.com>
> >>   * Simplify Darcs.Lock.withDir.
> >
> > These look good.  The only catch is that there is no documentation patch to
> > let folks know that they  can make use of it.  And for the same reason, I'm
> > not clear as to what the use case for this feature is.  I'll wait for an
> > explanation (ideally in the form of a documentation patch) before applying
> > this one.
> 
> Documentation patch is attached. I am looking at issue1139 and it is
> useful to keep temporary directories used for diff.

Argh.  Our way of building docs is unclear.  You've managed to not get your
documentation into a place that users can read it, since Darcs/Lock.lhs
isn't part of the manual.  :( Which is not your fault, since it's got some
latex docs.  But there's another file (somewhere...  I'd have to grep for
it, and maybe read the manual) that we document what environment variables
darcs accepts as input, and this should be documented there.  If it's a
debugging feature that only developers are doing to use, I think we
shouldn't be using an environment variable.  Developers, after all, can
pretty easily modify the code to change its behavior, but environment
variables have the potential to cause behavior that's confusing to users
("it does A on my account, but does B on his account").  So it's your
call.  If you think it's worth mentioning in the actual manual, you can
write something up, otherwise I'll leave it out.

David


More information about the darcs-users mailing list