jh granarolo wrote:
Here I would tend to think that the file META1 is first moved to
META.in, then another file named META1 is added to the repository, and
this is what I do when I build the tree.
But " darcs get --to-match 'hash the-hash-of-this-patch' " disagrees
with me : the copy of the repository obtained only contains the file
META.in.
I tried to reproduce this kind of changelog on a test repository,
without success : I only obtain the same output when a file "META1" is
actually added to the repository after the first one has been moved.
Any suggestion on how to handle this ? (or maybe on a better way to
build the file tree for a given patch?)
Perhaps this is an amend-record? (That is, META1 was added, recorded,
moved and then amend-recorded.) Sometimes darcs creates subtly different
patches than you would expect in such situations.
Moves are tricky no matter what, particularly because I think that
--xml-output sometimes does reorder the primitive patches into nice sort
orders that also make it tougher to follow the actual operations. Have
you compared ``darcs annotate -p 'The name of the patch in question'``
(or the 'p'ager operation in ``darcs changes -i``) to see if the order
of primitive patches makes more sense there? If that is the case it may
be useful to check the bug tracker for an issue on this reordering (I
think there might be one) or otherwise file a new one.
As for a better way to do what you are doing: if I knew the context of
what you are doing I might suggest you take some other approach or just
handle gracefully any error conditions you are dealing with. I've got a
feeling that what you are trying to do is not necessarily what you want
to be doing...
--
--Max Battcher--
http://worldmaker.net
_______________________________________________
darcs-users mailing list
[email protected]
http://lists.osuosl.org/mailman/listinfo/darcs-users
