Lobbying for -p1 diffs

[Matthew D. Fuller]

On Thu, Mar 02, 2006 at 12:01:33PM +1100 I heard the voice of
Michael Ellerman, and lo! it spake thus:
> 
> They _are_ different files, a patch describes the differences
> between two files. Try generating a -p0 diff of the same file
> without using a VCS.

Well, _yes_, but no; I'm not trying to compare two different files,
I'm looking at changes in one file.  Anyway, I use a VCS so I don't
have to do that; I've imported things into CVS many times just to make
a single-shot change and be able to generate a decent diff, then rm'd
the module.  ;)


> You're probably right. I've seen lots of bug reports against
> quilt/git/mercurial about the "gloop" in their diffs. Why are we
> still talking about this! :)

Well, it's that or get real work done.  And I did that LAST week!

Really, I think my main issue boils down to: There are cases where -p1
may be useful, so it should be possible to create, but most of the
time it's just noise.  And even if it's noise that's fairly easy to
mentally eliminate, noise is bad.


-- 
Matthew Fuller     (MF4839)   |  fullermd at over-yonder.net
Systems/Network Administrator |  http://www.over-yonder.net/~fullermd/
           On the Internet, nobody can hear you scream.