[PATCH/RFC 0/3] merge-recursive: Avoid diagnostic on empty work tree

From: Brad King <brad.king@kitware.com>
Date: Jan 24, 2014, 15:01 UTC
Message-ID: <cover.1390574980.git.brad.king@kitware.com>
In-Reply-To: <CABPp-BGAsrrjcZxVirzKU_VEyUM1U=4TFj18CieKKE7==c7v2A@mail.gmail.com>

On 01/23/2014 07:24 PM, Elijah Newren wrote:

Show 5 quoted lines

> Two options are just doing a stat to determine whether the file
> is present (which means we'll be stat'ing the file multiple times
> in these cases, which feels wasteful), or perhaps writing a
> modified make_cache_entry() with the behavior we want
> (seems like ugly code duplication).  Suggestions?

Perhaps we can thread enough information through the make_cache_entry signature to allow the caller to know when lstat reported ENOENT. Here is a series that takes such an approach.

* Patch 1 is the original test case from $gmane/240853.

* Patch 2 extends the make_cache_entry signature to return lstat errno.

* Patch 3 uses this information to silence the add_cacheinfo diagnostic

-Brad

Brad King (3):
  t3030-merge-recursive: Test known breakage with empty work tree
  read-cache.c: Thread lstat error through make_cache_entry signature
  merge-recursive: Tolerate missing file when HEAD is up to date

 builtin/apply.c            |  2 +-
 builtin/checkout.c         |  2 +-
 builtin/reset.c            |  2 +-
 cache.h                    |  2 +-
 merge-recursive.c          | 22 ++++++++++++++--------
 read-cache.c               | 12 +++++++-----
 resolve-undo.c             |  2 +-
 t/t3030-merge-recursive.sh | 47 ++++++++++++++++++++++++++++++++++++++++++++++
 8 files changed, 73 insertions(+), 18 deletions(-)

-- 
1.8.5.2