On Mon, Jul 31, 2006 at 01:36:37PM +0200, Marc Haber wrote: > On Mon, Jul 31, 2006 at 01:35:40PM +0200, Martin Quinson wrote: > > Thanks. Got it. I do some more checks, and I upload a fixed version. The new > > behaviour is to fail with an informative message when it happens. Does it > > fit your expectations ? > > Absolutely. I do not have a problem to mount /proc in a new chroot, I > only hate to debug for hours if I forget ;)
[patch-wrapper.test]
Use of uninitialized value in undef operator at ./run line 39.
Use of uninitialized value in undef operator at ./run line 39.
Use of uninitialized value in undef operator at ./run line 39.
[1] $ rm -rf d -- ok
[2] $ mkdir -p d/somewhere -- ok
[3] $ cd d -- ok
[5] $ cat > foo.orig -- ok
[11] $ sed -e 's/3/3a/' foo.orig > foo -- ok
[12] $ diff -u foo.orig foo > foo.diff -- ok
[13] $ mv foo.orig foo -- ok
[15] $ patch-wrapper -s -p0 < foo.diff -- failed
patch-wrapper: /proc not mounted! != ~
[16] $ quilt pop -q -- failed
No patch removed != Removing patch patches/foo.diff
~ != No patches applied
[19] $ rm -rf ${QUILT_PATCHES:-patches} ${QUILT_PC:-.pc} -- ok
[21] $ patch-wrapper --backup -B xxx/ -s -p0 < foo.diff -- failed
patch-wrapper: /proc not mounted! != ~
[22] $ find xxx -type f -- failed
find: xxx: No such file or directory != xxx/foo
[...]
Rah. It works (ie, the first error message is informative). I upload it.
The uninitialized values are really harmless, this is just because there is
no environment variable. I could fix it, but that would be yet another patch
against upstream for a really tiny issue so I'm not sure.
Bye, Mt.
--
A language that doesn't affect the way you think about programming, is not
worth knowing. -- "Epigrams in Programming", by Alan J. Perlis
signature.asc
Description: Digital signature
