Re: [PATCHv3 0/2] Fix relative path issues in recursive submodules.

From: Ramsay Jones <ramsay@ramsayjones.plus.com>
Date: Apr 13, 2016, 19:21 UTC
Message-ID: <570E9C46.6080905@ramsayjones.plus.com>
In-Reply-To: <CAGZ79kYE3E7stNc3Xbzr-qvncLGt4qRPMbv897k5h0B50-Rogg@mail.gmail.com>

On 12/04/16 16:58, Stefan Beller wrote:

> On Fri, Apr 1, 2016 at 7:41 AM, Ramsay Jones
> <ramsay@ramsayjones.plus.com> wrote:
>>

[snip[

Show 27 quoted lines

>>> -     }
>>> +     sm_gitdir_rel = strbuf_detach(&sb, NULL);
>>
>> ... this is good, but ...
>>
>>> +     sm_gitdir = absolute_path(sm_gitdir_rel);
>>>
>>>       if (!is_absolute_path(path)) {
>>> -             /*
>>> -              * TODO: add prefix here once we allow calling from non root
>>> -              * directory?
>>> -              */
>>> -             strbuf_addf(&sb, "%s/%s",
>>> -                         get_git_work_tree(),
>>> -                         path);
>>> +             strbuf_addf(&sb, "%s/%s", get_git_work_tree(), path);
>>>               path = strbuf_detach(&sb, 0);
>>
>> ... can you please fix this up.
>>
>> Thanks!
>>
>> ATB,
>> Ramsay Jones
> 
> Looking at the current code of origin/sb/submodule-helper-clone-regression-fix
> we do not have this issue there, but I'll keep it in mind for a resend.

Hmm, actually, the above change wasn't the original culprit (as I thought), but a different instance of the same fault. :-D

I've lost track of which version is now in 'pu' (currently @ 45a4edc "Merge branch 'sb/submodule-init' into pu"), but sparse is still warning:

      SP submodule.c
  submodule.c:256:43: warning: Using plain integer as NULL pointer

So, the fix looks like:

  diff --git a/submodule.c b/submodule.c
  index 5d1238a..4cc1c27 100644
  --- a/submodule.c
  +++ b/submodule.c
  @@ -253,7 +253,7 @@ const char *submodule_strategy_to_string(const struct submodule_update_strategy
                  return NULL;
          case SM_UPDATE_COMMAND:
                  strbuf_addf(&sb, "!%s", s->command);
  -               return strbuf_detach(&sb, 0);
  +               return strbuf_detach(&sb, NULL);
          }
          return NULL;
   }

Also, I note that t7406-submodule-update.sh test #4 is failing. (looks like absolute vs relative paths)

ATB, Ramsay Jones