[PATCH 0/4] clean-up store_updated_refs

[PATCH 0/4] clean-up store_updated_refs

[Vasyl' Vavrychuk]

Vasyl' Vavrychuk (4):
  Reduce url_len variable scope
  Extract function trim_url and optimize calls of it.
  Dont use the same variable for different things
  Replace pointer arithmetic with strbuf

 builtin/fetch.c |   63 ++++++++++++++++++++++++++++++++----------------------
 1 files changed, 37 insertions(+), 26 deletions(-)

[PATCH 1/4] Reduce url_len variable scope

[Vasyl' Vavrychuk]

Signed-off-by: Vasyl' Vavrychuk <vvavrychuk@gmail.com>
---
 builtin/fetch.c |    6 +++---
 1 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/builtin/fetch.c b/builtin/fetch.c
index 357f3cd..2b0b11e 100644
--- a/builtin/fetch.c
+++ b/builtin/fetch.c
@@ -385,6 +385,7 @@ static int store_updated_refs(const char *raw_url, const char *remote_name,
 		url_len = i + 1;
 		if (4 < i && !strncmp(".git", url + i - 3, 4))
 			url_len = i - 3;
+		url[url_len] = '\0';
 
 		note_len = 0;
 		if (*what) {
@@ -399,7 +400,7 @@ static int store_updated_refs(const char *raw_url, const char *remote_name,
 				    rm->old_sha1),
 			rm->merge ? "" : "not-for-merge",
 			note);
-		for (i = 0; i < url_len; ++i)
+		for (i = 0; url[i]; ++i)
 			if ('\n' == url[i])
 				fputs("\\n", fp);
 			else
@@ -415,8 +416,7 @@ static int store_updated_refs(const char *raw_url, const char *remote_name,
 				 REFCOL_WIDTH, *what ? what : "HEAD");
 		if (*note) {
 			if (verbosity >= 0 && !shown_url) {
-				fprintf(stderr, "From %.*s\n",
-						url_len, url);
+				fprintf(stderr, "From %s\n", url);
 				shown_url = 1;
 			}
 			if (verbosity >= 0)
-- 
1.7.1

Re: [PATCH 1/4] Reduce url_len variable scope

[Junio C Hamano]

Vasyl' Vavrychuk <vvavrychuk@gmail.com> writes:

> Subject: Re: [PATCH 1/4] Reduce url_len variable scope

Please mark the patch in such a way that the readers can tell what
file/function this patch is about by reading "git shortlog" output
without anything else, i.e.

	Subject: builtin/fetch.c: shorten lifetime of the url_len variable

Hmph.  While this does not bring in any new bug to the code, what benefit
do we gain from it?

The scope is not reduced (even though the variable's lifespan is shortened
in the scope).  Are we so short of registers that this change matters?

>  builtin/fetch.c |    6 +++---
>  1 files changed, 3 insertions(+), 3 deletions(-)
>
> diff --git a/builtin/fetch.c b/builtin/fetch.c
> index 357f3cd..2b0b11e 100644
> --- a/builtin/fetch.c
> +++ b/builtin/fetch.c
> @@ -385,6 +385,7 @@ static int store_updated_refs(const char *raw_url, const char *remote_name,
>  		url_len = i + 1;
>  		if (4 < i && !strncmp(".git", url + i - 3, 4))
>  			url_len = i - 3;
> +		url[url_len] = '\0';
>  
>  		note_len = 0;
>  		if (*what) {
> @@ -399,7 +400,7 @@ static int store_updated_refs(const char *raw_url, const char *remote_name,
>  				    rm->old_sha1),
>  			rm->merge ? "" : "not-for-merge",
>  			note);
> -		for (i = 0; i < url_len; ++i)
> +		for (i = 0; url[i]; ++i)
>  			if ('\n' == url[i])
>  				fputs("\\n", fp);
>  			else
> @@ -415,8 +416,7 @@ static int store_updated_refs(const char *raw_url, const char *remote_name,
>  				 REFCOL_WIDTH, *what ? what : "HEAD");
>  		if (*note) {
>  			if (verbosity >= 0 && !shown_url) {
> -				fprintf(stderr, "From %.*s\n",
> -						url_len, url);
> +				fprintf(stderr, "From %s\n", url);
>  				shown_url = 1;
>  			}
>  			if (verbosity >= 0)
> -- 
> 1.7.1

[PATCH 2/4] Extract function trim_url and optimize calls of it.

[Vasyl' Vavrychuk]

Extract compact code into trim_url. Dont call it every iteration in the loop since no reason.

Signed-off-by: Vasyl' Vavrychuk <vvavrychuk@gmail.com>
---
 builtin/fetch.c |   29 ++++++++++++++++++-----------
 1 files changed, 18 insertions(+), 11 deletions(-)

diff --git a/builtin/fetch.c b/builtin/fetch.c
index 2b0b11e..46d8fd6 100644
--- a/builtin/fetch.c
+++ b/builtin/fetch.c
@@ -89,6 +89,19 @@ static void unlock_pack_on_signal(int signo)
 	raise(signo);
 }
 
+static void trim_url(char *url)
+{
+	int i, url_len;
+
+	url_len = strlen(url);
+	for (i = url_len - 1; url[i] == '/' && 0 <= i; i--)
+		;
+	url_len = i + 1;
+	if (4 < i && !strncmp(".git", url + i - 3, 4))
+		url_len = i - 3;
+	url[url_len] = '\0';
+}
+
 static void add_merge_config(struct ref **head,
 			   const struct ref *remote_refs,
 		           struct branch *branch,
@@ -329,7 +342,7 @@ static int store_updated_refs(const char *raw_url, const char *remote_name,
 {
 	FILE *fp;
 	struct commit *commit;
-	int url_len, i, note_len, shown_url = 0, rc = 0;
+	int i, note_len, shown_url = 0, rc = 0;
 	char note[1024];
 	const char *what, *kind;
 	struct ref *rm;
@@ -339,10 +352,12 @@ static int store_updated_refs(const char *raw_url, const char *remote_name,
 	if (!fp)
 		return error("cannot open %s: %s\n", filename, strerror(errno));
 
-	if (raw_url)
+	if (raw_url) {
 		url = transport_anonymize_url(raw_url);
-	else
+		trim_url(url);
+	} else {
 		url = xstrdup("foreign");
+	}
 	for (rm = ref_map; rm; rm = rm->next) {
 		struct ref *ref = NULL;
 
@@ -379,14 +394,6 @@ static int store_updated_refs(const char *raw_url, const char *remote_name,
 			what = rm->name;
 		}
 
-		url_len = strlen(url);
-		for (i = url_len - 1; url[i] == '/' && 0 <= i; i--)
-			;
-		url_len = i + 1;
-		if (4 < i && !strncmp(".git", url + i - 3, 4))
-			url_len = i - 3;
-		url[url_len] = '\0';
-
 		note_len = 0;
 		if (*what) {
 			if (*kind)
-- 
1.7.1

Re: [PATCH 2/4] Extract function trim_url and optimize calls of it.

[Junio C Hamano]

Vasyl' Vavrychuk <vvavrychuk@gmail.com> writes:

> Extract compact code into trim_url. Dont call it every iteration in the loop since no reason.

Too long a line; just drop " since no reason".

> @@ -379,14 +394,6 @@ static int store_updated_refs(const char *raw_url, const char *remote_name,
>  			what = rm->name;
>  		}
>  
> -		url_len = strlen(url);
> -		for (i = url_len - 1; url[i] == '/' && 0 <= i; i--)
> -			;
> -		url_len = i + 1;
> -		if (4 < i && !strncmp(".git", url + i - 3, 4))
> -			url_len = i - 3;
> -		url[url_len] = '\0';
> -
>  		note_len = 0;
>  		if (*what) {
>  			if (*kind)

We repeatedly called strlen(url) for each entry in the ref-map when we
know we do not have any more thing to do; silly.  This is probably a good
thing to do.

[PATCH 3/4] Dont use the same variable for different things

[Vasyl' Vavrychuk]

Signed-off-by: Vasyl' Vavrychuk <vvavrychuk@gmail.com>
---
 builtin/fetch.c |   13 +++++++------
 1 files changed, 7 insertions(+), 6 deletions(-)

diff --git a/builtin/fetch.c b/builtin/fetch.c
index 46d8fd6..ddb40c6 100644
--- a/builtin/fetch.c
+++ b/builtin/fetch.c
@@ -343,7 +343,7 @@ static int store_updated_refs(const char *raw_url, const char *remote_name,
 	FILE *fp;
 	struct commit *commit;
 	int i, note_len, shown_url = 0, rc = 0;
-	char note[1024];
+	char note[1024], display[1024];
 	const char *what, *kind;
 	struct ref *rm;
 	char *url, *filename = dry_run ? "/dev/null" : git_path("FETCH_HEAD");
@@ -415,19 +415,20 @@ static int store_updated_refs(const char *raw_url, const char *remote_name,
 		fputc('\n', fp);
 
 		if (ref) {
-			rc |= update_local_ref(ref, what, note);
+			rc |= update_local_ref(ref, what, display);
 			free(ref);
-		} else
-			sprintf(note, "* %-*s %-*s -> FETCH_HEAD",
+		} else {
+			sprintf(display, "* %-*s %-*s -> FETCH_HEAD",
 				TRANSPORT_SUMMARY_WIDTH, *kind ? kind : "branch",
 				 REFCOL_WIDTH, *what ? what : "HEAD");
-		if (*note) {
+		}
+		if (*display) {
 			if (verbosity >= 0 && !shown_url) {
 				fprintf(stderr, "From %s\n", url);
 				shown_url = 1;
 			}
 			if (verbosity >= 0)
-				fprintf(stderr, " %s\n", note);
+				fprintf(stderr, " %s\n", display);
 		}
 	}
 	free(url);
-- 
1.7.1

Re: [PATCH 3/4] Dont use the same variable for different things

[Junio C Hamano]

Vasyl' Vavrychuk <vvavrychuk@gmail.com> writes:

> Subject: Re: [PATCH 3/4] Dont use the same variable for different things

	Subject: builtin/fetch.c: do not use ... different things

But more important question is why?

What benefit are we getting by almost doubling the stack footprint of this
function?  What problem are you fixing?

Re: [PATCH 3/4] Dont use the same variable for different things

[João P. Sampaio]

On Wed, Feb 16, 2011 at 23:44, Junio C Hamano <gitster@pobox.com> wrote:
> But more important question is why?
>
> What benefit are we getting by almost doubling the stack footprint of this
> function?  What problem are you fixing?

Well, in matters of code readability and maintainability, it's best to
use different variables for different matters...

Nowadays, memory isn't really an issue anymore, is it? I think it's
valid to make code more readable, if memory cost isn't absurd.

-- 
João Paulo Melo de Sampaio
Computer Engineering Student @ UFSCar
Website: http://www.jpmelos.com
Twitter: http://twitter.com/jpmelos (@jpmelos)

[PATCH 4/4] Replace pointer arithmetic with strbuf

[Vasyl' Vavrychuk]

Signed-off-by: Vasyl' Vavrychuk <vvavrychuk@gmail.com>
---
 builtin/fetch.c |   21 ++++++++++++---------
 1 files changed, 12 insertions(+), 9 deletions(-)

diff --git a/builtin/fetch.c b/builtin/fetch.c
index ddb40c6..eaea475 100644
--- a/builtin/fetch.c
+++ b/builtin/fetch.c
@@ -342,8 +342,9 @@ static int store_updated_refs(const char *raw_url, const char *remote_name,
 {
 	FILE *fp;
 	struct commit *commit;
-	int i, note_len, shown_url = 0, rc = 0;
-	char note[1024], display[1024];
+	int i, shown_url = 0, rc = 0;
+	struct strbuf note = STRBUF_INIT;
+	char display[1024];
 	const char *what, *kind;
 	struct ref *rm;
 	char *url, *filename = dry_run ? "/dev/null" : git_path("FETCH_HEAD");
@@ -394,19 +395,20 @@ static int store_updated_refs(const char *raw_url, const char *remote_name,
 			what = rm->name;
 		}
 
-		note_len = 0;
 		if (*what) {
-			if (*kind)
-				note_len += sprintf(note + note_len, "%s ",
-						    kind);
-			note_len += sprintf(note + note_len, "'%s' of ", what);
+			if (*kind) {
+				strbuf_addstr(&note, kind);
+				strbuf_addch(&note, ' ');
+			}
+			strbuf_addf(&note, "'%s' of ", what);
 		}
-		note[note_len] = '\0';
 		fprintf(fp, "%s\t%s\t%s",
 			sha1_to_hex(commit ? commit->object.sha1 :
 				    rm->old_sha1),
 			rm->merge ? "" : "not-for-merge",
-			note);
+			note.buf);
+		strbuf_reset(&note);
+
 		for (i = 0; url[i]; ++i)
 			if ('\n' == url[i])
 				fputs("\\n", fp);
@@ -431,6 +433,7 @@ static int store_updated_refs(const char *raw_url, const char *remote_name,
 				fprintf(stderr, " %s\n", display);
 		}
 	}
+	strbuf_release(&note);
 	free(url);
 	fclose(fp);
 	if (rc & STORE_REF_ERROR_DF_CONFLICT)
-- 
1.7.1

Re: [PATCH 0/4] clean-up store_updated_refs

[Sverre Rabbelier]

Heya,

On Sat, Feb 12, 2011 at 23:38, Vasyl' Vavrychuk <vvavrychuk@gmail.com> wrote:
> Vasyl' Vavrychuk (4):

Neither the patches, nor this cover letter explain _why_ this is being
done. Motivation needed :).

-- 
Cheers,

Sverre Rabbelier

Re: [PATCH 0/4] clean-up store_updated_refs

[Vasyl']

Hi,

I've needed this as part of bigger change which is collecting all
.git/FETCH_HEAD related code across different files to
fetch-head.[ch]. Which in its case I do as part of git-pull.sh to C
porting. So thats motivation chain :)

On Sun, Feb 13, 2011 at 2:09 AM, Sverre Rabbelier <srabbelier@gmail.com> wrote:
> Heya,
>
> On Sat, Feb 12, 2011 at 23:38, Vasyl' Vavrychuk <vvavrychuk@gmail.com> wrote:
>> Vasyl' Vavrychuk (4):
>
> Neither the patches, nor this cover letter explain _why_ this is being
> done. Motivation needed :).
>
> --
> Cheers,
>
> Sverre Rabbelier
>

Re: [PATCH 0/4] clean-up store_updated_refs

[Sverre Rabbelier]

Heya,

On Sun, Feb 13, 2011 at 19:37, Vasyl' <vvavrychuk@gmail.com> wrote:
> I've needed this as part of bigger change which is collecting all
> .git/FETCH_HEAD related code across different files to
> fetch-head.[ch]. Which in its case I do as part of git-pull.sh to C
> porting. So thats motivation chain :)

Thanks, that's the kind of stuff that should go into the cover letter
/ the patch description.

-- 
Cheers,

Sverre Rabbelier