Re: [PATCH v2 2/4] use strbuf_getcwd() to get the current working directory without fixed-sized buffers

["René Scharfe" <l.s.r@web.de>]

Am 21.07.2014 04:33, schrieb Jeff King:
> On Sun, Jul 20, 2014 at 06:49:54PM +0200, René Scharfe wrote:
>
>> diff --git a/builtin/init-db.c b/builtin/init-db.c
>> index 56f85e2..c4958b6 100644
>> --- a/builtin/init-db.c
>> +++ b/builtin/init-db.c
>> @@ -535,10 +535,10 @@ int cmd_init_db(int argc, const char **argv, const char *prefix)
>>   		usage(init_db_usage[0]);
>>   	}
>>   	if (is_bare_repository_cfg == 1) {
>> -		static char git_dir[PATH_MAX+1];
>> -
>> -		setenv(GIT_DIR_ENVIRONMENT,
>> -			getcwd(git_dir, sizeof(git_dir)), argc > 0);
>> +		struct strbuf cwd = STRBUF_INIT;
>> +		strbuf_getcwd(&cwd);
>> +		setenv(GIT_DIR_ENVIRONMENT, cwd.buf, argc > 0);
>> +		strbuf_release(&cwd);
>
> Hmm. You are not making anything worse here, as we already do not check
> the return value of getcwd. But what happens if it fails? Looks like we
> currently get a segfault, and the new code will silently set the
> variable to the empty string. Neither is particularly helpful.
>
> Should we be using the xgetcwd helper that you add in the next patch?

Probably.  And I was so glad to have found an example case for getcwd 
without dying and without touching the get-there-and-back cases. :) 
Guess I'll have to look closer at setup.c and perhaps unix-socket.c for 
a replacement.

By the way: Simply setting $GIT_DIR to "." probably won't work in the 
two cases, I guess?

>
>> -			setenv(GIT_DIR_ENVIRONMENT, getcwd(git_dir, sizeof(git_dir)), 0);
>> +			strbuf_getcwd(&cwd);
>> +			setenv(GIT_DIR_ENVIRONMENT, cwd.buf, 0);
>> +			strbuf_release(&cwd);
>
> Ditto here.
>
> -Peff
>