[svn.haxx.se] · SVN Dev · SVN Users · SVN Org · TSVN Dev · TSVN Users · Subclipse Dev · Subclipse Users · this month's index

Re: Patch command execution

From: Daniel Berlin <dan_at_dberlin.org>
Date: 2002-02-07 00:34:53 CET

On Wed, 6 Feb 2002, Garrett Rooney wrote:

> On Wed, Feb 06, 2002 at 02:41:12PM -0600, Karl Fogel wrote:
> > Daniel Berlin <dan@dberlin.org> writes:
> > > Shouldn't we be using the -N option so we don't try to apply reversed
> > > /already applied patches (-N = forward patch only, ignore reversed/already
> > > applied)
> > > Otherwise, if the change already exists in your local tree, it'll get a
> > > reject.
> > >
> > > Of course, patch isn't perfect (i don't think it'll let you ignore just
> > > portions of an already applied patch), so maybe this is deliberate?
> >
> > I think we should pass -N too, but let's give it a day to see if
> > anyone else realizes some evil consequences of doing so...
>
> someone mentioned this on the list a while back, and even tried it out
> and posted a unit test for this behavior, but i believe they mentioned
> that it didn't work the way they wanted it to for some reason. i
> believe they posted something about it in the issue about patch
> erroring out if the changes already exist.
There's no way around that, actually.
It auto-skips the rest of the patch.
Ideally, what you want it to do, is match hunks that exist, and ignore
them if they are the same, but apply those hunks that don't exist.

But of course, patch, on detecting one reversed hunk, will skip the rest
of the patch.

> -garrett
>
>

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Received on Sat Oct 21 14:37:05 2006

This is an archived mail posted to the Subversion Dev mailing list.

This site is subject to the Apache Privacy Policy and the Apache Public Forum Archive Policy.