Stefan Sperling <stsp@stsp.name> wrote:
> Have you taken a look at Subversion's patch.c: choose_target_filename()?

I forgot to reply to this.  I'm going a bit out of topic here but i
wanted to explain this somewhere because it's an idea I followed when
implementing the patch command that I haven't explained elsewhere.

Larry' patch (and svn' patch.c too judging from the code, i never used
subversion) make use of some heuristics to determine what is the "old
file" and what is the "new" one (like choosing the path with the smaller
number of components.)

While this is perfectly fine for patch(1), i'd like (if possible) for
got to not do this.  One important advantage got patch has over patch(1)
is that we _know_ which files are part of the repository and which
don't.  We don't have to guess.  Or at least, I'd like if we can manage
not to resort to guesses and heuristics :)

(That's the idea behind the "check file status before applying patch"
diff that i sent previously and that now i have to adjust.)

It's not perfect yet, in particular this means that at the moment `got
patch' fails on partial checkouts (i.e. `got co -p') and diff for
relative paths (the one you generate with cvs in a subdirectory for
instance), but I think we can fix these issues in a different way.