On Oct 28, 2009, at 7:28 PM, Colin Barrett wrote:
On Oct 28, 2009, at 5:21 PM, Matthew wrote:
On Wed, Oct 28, 2009 at 17:27, Colin Barrett <[email protected]
> wrote:
I decided to do this update after getting two or three emails
about patches
the past couple weeks about patches and I began to wonder why we
were still
instructing people to use .diff files in this glorious new age of
DVCS ;)
I've been encouraged to submit diffs because they are easier to see
within trac and therefore are more likely to be reviewed. Considering
that this is the same person (mostly) who has actually reviewed and
applied the patches I've submitted, I have to admit that I'm inclined
to continue submitting diffs and wonder if moving away from diffs
will
result in less patches being applied.
I'm happy to do either; I'll do what I can to get the patches
reviewed
and applied. So far my anecdotal evidence suggests that providing
diffs helps.
If it's a simple .diff, it really makes no difference. However many
patches require updates to NIBs and currently that requires
downloading a zip, unzipping it, and applying the patch. Pretty big
PITA. Furthermore it's pretty simple in the Bitbucket UI to view the
diff of an particular commit.
Not if they use hg diff --git ; that includes changes to binary files.
Augie
Your'e still welcome to submit diffs; it's just not the only way to
do things.
-Colin
