[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Quilt-dev] dropping a patch
From: |
Randy.Dunlap |
Subject: |
Re: [Quilt-dev] dropping a patch |
Date: |
Thu, 13 Apr 2006 14:29:18 -0700 |
On Thu, 13 Apr 2006 22:51:01 +0200 Jean Delvare wrote:
> Hi Randy,
>
> [Randy Dunlap]
> > > > Am I missing some shortcut to drop a patch (that may contain
> > > > patches to multiple files)?
> > > >
> > > > Do I actually have to go thru:
> > > >
> > > > quilt push # make sure that it is applied since 'remove' seems to
> > > > # work only on applied patches
> > > > quilt files # list all of them since I forgot what they are
> > > > quilt remove file1 file2 file3 file4 file5...
> > > > quilt delete drop-this.patch # top patch filename
>
> [Jean Delvare]
> > > I doubt this would work. "quilt remove" restores the files as they were
> > > at the time they were added to the patch, so the above sequence doesn't
> > > actually do anything (besides deleting the patch itself.)
>
> [Randy Dunlap]
> > 1/ The patch was bad/wrong, so I _want_ quilt remove to restore
> > the files to their previous state. That's what I want drop to do
> > (or maybe I should call it "revert").
>
> You should call it "delete" ;) The behavior you describe is exactly what
> "quilt delete" does. The sequence in your example above (push, remove
> files, delete) is equivalent to "quilt delete -n".
Darn, you're right. I didn't realize that 'delete' would do the
'remove(s)' too.
Hm, sometimes it doesn't work for me, so I'll look into why that is.
It just says:
No patch removed
Patch <patch-filename> is currently applied
Hm again. This is quilt 0.42. I'll upgrade and try again
(since it works on a different machine with 0.44)...
> > 2/ The sequence above does work for what I want to do, it's just
> > cumbersome... OK, I'll write a script for it.
> > And check on Josh's drop patches.
>
> You will probably not like them if what you call "drop" is what we call
> "delete". Josh's "drop" function does not discard the changes, it's
> quite the opposite: it removes them from quilt's control by making them
> part of the base code.
OK, thanks. Could still be useful.
---
~Randy