Re: [PATCH v4 02/26] refs: make repack_without_refs and is_branch public

On Fri, 2015-10-16 at 08:34 +0200, Michael Haggerty wrote:

On 10/15/2015 09:46 PM, David Turner wrote:

quoted

is_branch was already non-static, but this patch declares it in the
header.

The commit message no longer reflects the patch.

quoted

Signed-off-by: Ronnie Sahlberg <redacted>
Signed-off-by: David Turner <redacted>
Signed-off-by: Junio C Hamano <redacted>
---
 refs.c | 5 +++--
 refs.h | 2 ++
 2 files changed, 5 insertions(+), 2 deletions(-)

diff --git a/refs.c b/refs.c
index fe71ea0..84abc82 100644
--- a/refs.c
+++ b/refs.c

@@ -2816,8 +2816,9 @@ int pack_refs(unsigned int flags)
 
 /*
  * Rewrite the packed-refs file, omitting any refs listed in
- * 'refnames'. On error, leave packed-refs unchanged, write an error
- * message to 'err', and return a nonzero value.
+ * 'refnames'. On error, packed-refs will be unchanged, the return
+ * value is nonzero, and a message about the error is written to the
+ * 'err' strbuf.

^^^ ?

It is preferable for docstrings to be written in imperative form, so in
my opinion this is a step backwards...

...literally. Your "new" version comes from an older version of Git; it
was changed in

    79e4d8a9b8 repack_without_refs(): make function private (2015-06-22)

to the imperative form.

Assuming you are using `git-format-patch` to prepare your patches, it is
always a good idea to read over the prepared email files before sending
them to the ML, to check for bloopers like this.

Sorry about that one.  It's hard to keep track of what all of these
patches do -- especially the ones that were rebases of Ronnie's.  I've
fixed that and 03/26 as well.

Do you have comments on any of the rest before I re-roll?