Erik, thanks of course but please, use the ML tailor related material, I
feel much more comfortable that way.
Erik Schnetter wrote:
Lele,
when I try to create CVS or SVN repositories from darcs repositories, I
still have problems. I see error messages like
darcs failed: user error (Error applying patch to working dir:
./CarpetExtra/WaveToyF77/test/test_rad: removeDirectory: unsatisified
constraints (Directory not empty))
This may have left your working directory an inconsistent
but recoverable state. If you had no un-recorded changes
by using 'darcs revert' you should be able to make your
working directory consistent again.
which occur if darcs wants to delete a directory, but this directory is
not empty because it still contains a "CVS" or a ".svn" directory.
Uhm, I know CVS has this bad habit, but has svn really inherited that
too? For CVS, there is cvsps.py::__maybeDeleteDirectory() that should
handle the case, while there's nothing equivalent for svn...
Any chance of composing a simple test that demonstrate this?
On another account, I'm trying to create a darcs repository from a CVS
repository. There I had the problem that one of the CVS commit
messages contained the character "a0", which is similar to a space
character, but has bit 8 set. tailor failed because it could not
interpret this as UTF-8. I have "corrected" this by forcing CVS
repositories to be encoded as ISO-8859-1. I sent you this patch.
There may be a better way to implement this.
Uhm, it's difficult to take such a wide assumption: are really all CVS
repositories encoded in latin1? Isn't the per-repository option enough?
thanks again,
bye, lele.
_______________________________________________
Tailor mailing list
[email protected]
http://lists.zooko.com/mailman/listinfo/tailor
