Re: [PATCH] pack-objects: Allow missing base objects when creating thin packs

From: Shawn O. Pearce <spearce@spearce.org>
Date: Aug 12, 2008, 18:18 UTC
Message-ID: <20080812181843.GD31092@spearce.org>
In-Reply-To: <alpine.LFD.1.10.0808121402440.22892@xanadu.home>

Nicolas Pitre <nico@cam.org> wrote:

Show 17 quoted lines

> On Tue, 12 Aug 2008, Shawn O. Pearce wrote:
> > +# Clone patch_clone indirectly by cloning base and fetching.
> > +#
> > +test_expect_success \
> > +    'indirectly clone patch_clone' \
> > +    '(mkdir user_clone &&
> > +      cd user_clone &&
> > +      git init &&
> > +      git pull ../.git &&
> > +      test $(git rev-parse HEAD) = $B
> > +
> > +      git pull ../patch_clone/.git &&
> > +      test $(git rev-parse HEAD) = $C
> > +     )
> > +    '
> 
> What if the first test command fails?  Won't its result be ignored?

Isn't the exit status of the subshell the exit status of the last command in the subshell?

I just changed the test line to compare to "x$C" instead of $C and it correctly detected the error condition:

$ git diff

diff --git a/t/t5306-pack-nobase.sh b/t/t5306-pack-nobase.sh
index 503e9d4..7c55e9e 100755
--- a/t/t5306-pack-nobase.sh
+++ b/t/t5306-pack-nobase.sh
@@ -62,7 +62,7 @@ test_expect_success \
       test $(git rev-parse HEAD) = $B

       git pull ../patch_clone/.git &&
-      test $(git rev-parse HEAD) = $C
+      test $(git rev-parse HEAD) = x$C
      )
     '

$ ./t5306-pack-nobase.sh
*   ok 1: setup base
*   ok 2: setup patch_clone
* FAIL 3: indirectly clone patch_clone
        (mkdir user_clone &&
              cd user_clone &&
              git init &&
              git pull ../.git &&
              test $(git rev-parse HEAD) = $B

              git pull ../patch_clone/.git &&
              test $(git rev-parse HEAD) = x$C
             )

*   ok 4: clone of patch_clone is incomplete
* failed 1 among 4 test(s)

-- 
Shawn.