threads / patch / 49120

patchFix a recently-introduced compile warning

Subject: [PATCH 0/1] Fix a recently-introduced compile warning

## tl;dr

3 messages between Aug 14, 2018 and Aug 14, 2018. Diffs are folded; open one to read it.

replies: 2people: 2as markdown or json

Johannes Schindelin via GitGitGadget· Aug 14, 2018, 11:02 UTC · lore

With the relatively frequent breakages of pu recently, I had trouble staying on top of the compile errors/test failures, sorry.

This one exists since Sunday, and it is a compile error only with DEVELOPER=1, which is, however, the recommended way to build in Git for Windows' SDK.

Note: it is based on nd/clone-case-smashing-warning.

Johannes Schindelin (1):
  mark_colliding_entries(): fix incorrect #if...#endif guard

 entry.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

base-commit: f80218bf4e65ccc06cc9173c0ac5a5520d380f36
Published-As: https://github.com/gitgitgadget/git/releases/tags/pr-18%2Fdscho%2Fclone-case-smashing-warning-v1
Fetch-It-Via: git fetch https://github.com/gitgitgadget/git pr-18/dscho/clone-case-smashing-warning-v1
Pull-Request: https://github.com/gitgitgadget/git/pull/18

-- 
gitgitgadget

Johannes Schindelin via GitGitGadget· Aug 14, 2018, 11:02 UTC · re: Johannes Schindelin via GitGitGadget · lore

[PATCH 1/1] mark_colliding_entries(): fix incorrect #if...#endif guard

From: Johannes Schindelin <johannes.schindelin@gmx.de>

The way the guard was put, the code was declaring an unused variable on Windows. No need to do that, so let's fix it.

Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
---
 entry.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

Show changes to entry.c +3 −3

diff --git a/entry.c b/entry.c
index c70340df8..2bce13352 100644
--- a/entry.c
+++ b/entry.c
@@ -402,11 +402,9 @@ static int check_path(const char *path, int len, struct stat *st, int skiplen)
 static void mark_colliding_entries(const struct checkout *state,
 				   struct cache_entry *ce, struct stat *st)
 {
+#if !defined(GIT_WINDOWS_NATIVE) /* inode is always zero on Windows */
 	int i;
 
-	ce->ce_flags |= CE_MATCHED;
-
-#if !defined(GIT_WINDOWS_NATIVE) /* inode is always zero on Windows */
 	for (i = 0; i < state->istate->cache_nr; i++) {
 		struct cache_entry *dup = state->istate->cache[i];
 
@@ -422,6 +420,8 @@ static void mark_colliding_entries(const struct checkout *state,
 		}
 	}
 #endif
+
+	ce->ce_flags |= CE_MATCHED;
 }
 
 /*

-- 
gitgitgadget

Junio C Hamano· Aug 14, 2018, 15:08 UTC · re: Johannes Schindelin via GitGitGadget · lore

Re: [PATCH 1/1] mark_colliding_entries(): fix incorrect #if...#endif guard

"Johannes Schindelin via GitGitGadget" <gitgitgadget@gmail.com> writes:

Show 34 quoted lines

> From: Johannes Schindelin <johannes.schindelin@gmx.de>
>
> The way the guard was put, the code was declaring an unused variable on
> Windows. No need to do that, so let's fix it.
>
> Signed-off-by: Johannes Schindelin <johannes.schindelin@gmx.de>
> ---
>  entry.c | 6 +++---
>  1 file changed, 3 insertions(+), 3 deletions(-)
>
> diff --git a/entry.c b/entry.c
> index c70340df8..2bce13352 100644
> --- a/entry.c
> +++ b/entry.c
> @@ -402,11 +402,9 @@ static int check_path(const char *path, int len, struct stat *st, int skiplen)
>  static void mark_colliding_entries(const struct checkout *state,
>  				   struct cache_entry *ce, struct stat *st)
>  {
> +#if !defined(GIT_WINDOWS_NATIVE) /* inode is always zero on Windows */
>  	int i;
>  
> -	ce->ce_flags |= CE_MATCHED;
> -
> -#if !defined(GIT_WINDOWS_NATIVE) /* inode is always zero on Windows */
>  	for (i = 0; i < state->istate->cache_nr; i++) {
>  		struct cache_entry *dup = state->istate->cache[i];
>  
> @@ -422,6 +420,8 @@ static void mark_colliding_entries(const struct checkout *state,
>  		}
>  	}
>  #endif
> +
> +	ce->ce_flags |= CE_MATCHED;
>  }

Even though it looks a bit odd to smudge 'ce' itself after the loop, I think this would not break anything, simply because the loop does treat the 'ce' as special and stops the iteration once it sees it, even before its MATCHED bit is set.

Duy, if you are going to send a new version updated for other reasons, please squash this in. In the meantime, I'll queue it on top of what we have right now.

Thanks, both.