[PATCH/RFC] branch.c: Replace `git_config` with `git_config_get_string`

[PATCH/RFC] branch.c: Replace `git_config` with `git_config_get_string`

[Tanay Abhra]

Original implementation uses a callback based approach which has some
deficiencies like a convoluted control flow and redundant variables.
Use git_config_get_string instead of git_config to take advantage of
the config hash-table.

---

This patch builds on top of patch series[1]. It passes all the tests, and
this in first of series of patches that aim to replace all git_config
calls scattered arund the code base with appropriate non-callback based
calls.

There are total 111 calls in total in all of git codebase. How should I send
the patches, alphabetically or otherwise?

[1] http://thread.gmane.org/gmane.comp.version-control.git/251704

Cheers,
Tanay Abhra.

 branch.c | 25 +++++++++----------------
 1 file changed, 9 insertions(+), 16 deletions(-)

diff --git a/branch.c b/branch.c
index 660097b..257b1bf 100644
--- a/branch.c
+++ b/branch.c
@@ -140,33 +140,26 @@ static int setup_tracking(const char *new_ref, const char *orig_ref,
 	return 0;
 }
 
-struct branch_desc_cb {
+struct branch_desc {
 	const char *config_name;
 	const char *value;
 };
 
-static int read_branch_desc_cb(const char *var, const char *value, void *cb)
-{
-	struct branch_desc_cb *desc = cb;
-	if (strcmp(desc->config_name, var))
-		return 0;
-	free((char *)desc->value);
-	return git_config_string(&desc->value, var, value);
-}
-
 int read_branch_desc(struct strbuf *buf, const char *branch_name)
 {
-	struct branch_desc_cb cb;
+	const char *value;
+	struct branch_desc desc;
 	struct strbuf name = STRBUF_INIT;
 	strbuf_addf(&name, "branch.%s.description", branch_name);
-	cb.config_name = name.buf;
-	cb.value = NULL;
-	if (git_config(read_branch_desc_cb, &cb) < 0) {
+	desc.config_name = name.buf;
+	desc.value = NULL;
+	value = git_config_get_string(desc.config_name);
+	git_config_string(&desc.value, desc.config_name, value);
+	if (!desc.value) {
 		strbuf_release(&name);
 		return -1;
 	}
-	if (cb.value)
-		strbuf_addstr(buf, cb.value);
+	strbuf_addstr(buf, desc.value);
 	strbuf_release(&name);
 	return 0;
 }
-- 
1.9.0.GIT

Re: [PATCH/RFC] branch.c: Replace `git_config` with `git_config_get_string`

[Matthieu Moy]

> Subject: Re: [PATCH/RFC] branch.c: Replace `git_config` with `git_config_get_string`

Here and elsewhere: usually, no capital after :.

Tanay Abhra <tanayabh@gmail.com> writes:

> Original implementation uses a callback based approach which has some
> deficiencies like a convoluted control flow and redundant variables.

"deficiencies" might be a bit strong (the code did work).

> There are total 111 calls in total in all of git codebase. How should I send
> the patches, alphabetically or otherwise?

My advice would be: try as much as possible to split according to the
complexity of the patch.

As a reviewer, I find it rather easy to review a large number of trivial
and similar changes, but I hate having to switch back to "wow, the
author did something tricky, let's try to understand this" in the middle
of a trivial series.

(we had this discussion about `...` Vs $(...) and test -a Vs test ... &&
series, which were essentially very trivial changes, but with subtle
bugs introduced and hidden by the volume of trivial changes).

>  branch.c | 25 +++++++++----------------
>  1 file changed, 9 insertions(+), 16 deletions(-)

Removing more lines than it adds. I like the patch already ;-).

> diff --git a/branch.c b/branch.c
> index 660097b..257b1bf 100644
> --- a/branch.c
> +++ b/branch.c
[...]
>  int read_branch_desc(struct strbuf *buf, const char *branch_name)
>  {
> -	struct branch_desc_cb cb;
> +	const char *value;
> +	struct branch_desc desc;
>  	struct strbuf name = STRBUF_INIT;
>  	strbuf_addf(&name, "branch.%s.description", branch_name);
> -	cb.config_name = name.buf;
> -	cb.value = NULL;
> -	if (git_config(read_branch_desc_cb, &cb) < 0) {
> +	desc.config_name = name.buf;
> +	desc.value = NULL;
> +	value = git_config_get_string(desc.config_name);
> +	git_config_string(&desc.value, desc.config_name, value);

You're ignoring the return value of git_config_string, which is an error
code. It shouldn't harm, because the code is non-zero iff desc.value is
set to non-NULL, but you may want to write the code as

if (git_config_string(...)) {
	strbuf_release(...);
	return -1;
}

In any case, the patch sounds good to me.

-- 
Matthieu Moy
http://www-verimag.imag.fr/~moy/

Re: [PATCH/RFC] branch.c: Replace `git_config` with `git_config_get_string`

[Tanay Abhra]

On 06/16/2014 10:29 AM, Matthieu Moy wrote:
>> Subject: Re: [PATCH/RFC] branch.c: Replace `git_config` with `git_config_get_string`
> 
> Here and elsewhere: usually, no capital after :.
> 

Noted.

> Tanay Abhra <tanayabh@gmail.com> writes:
> 
>> Original implementation uses a callback based approach which has some
>> deficiencies like a convoluted control flow and redundant variables.
> 
> "deficiencies" might be a bit strong (the code did work).
> 

Hehe, I did spell out what the "deficiencies" were, nevertheless will revise it in
next iteration.

>> There are total 111 calls in total in all of git codebase. How should I send
>> the patches, alphabetically or otherwise?
> 
> My advice would be: try as much as possible to split according to the
> complexity of the patch.
> 
> As a reviewer, I find it rather easy to review a large number of trivial
> and similar changes, but I hate having to switch back to "wow, the
> author did something tricky, let's try to understand this" in the middle
> of a trivial series.
> 
> (we had this discussion about `...` Vs $(...) and test -a Vs test ... &&
> series, which were essentially very trivial changes, but with subtle
> bugs introduced and hidden by the volume of trivial changes).
> 

Noted.

>>  branch.c | 25 +++++++++----------------
>>  1 file changed, 9 insertions(+), 16 deletions(-)
> 
> Removing more lines than it adds. I like the patch already ;-).
> 
>> diff --git a/branch.c b/branch.c
>> index 660097b..257b1bf 100644
>> --- a/branch.c
>> +++ b/branch.c
> [...]
>>  int read_branch_desc(struct strbuf *buf, const char *branch_name)
>>  {
>> -	struct branch_desc_cb cb;
>> +	const char *value;
>> +	struct branch_desc desc;
>>  	struct strbuf name = STRBUF_INIT;
>>  	strbuf_addf(&name, "branch.%s.description", branch_name);
>> -	cb.config_name = name.buf;
>> -	cb.value = NULL;
>> -	if (git_config(read_branch_desc_cb, &cb) < 0) {
>> +	desc.config_name = name.buf;
>> +	desc.value = NULL;
>> +	value = git_config_get_string(desc.config_name);
>> +	git_config_string(&desc.value, desc.config_name, value);
> 
> You're ignoring the return value of git_config_string, which is an error
> code. It shouldn't harm, because the code is non-zero iff desc.value is
> set to non-NULL, but you may want to write the code as
> 
> if (git_config_string(...)) {
> 	strbuf_release(...);
> 	return -1;
> }
> 
> In any case, the patch sounds good to me.
> 

Yes, for clarity sake, will rewrite the section like that.
Thanks for the review.