From mboxrd@z Thu Jan  1 00:00:00 1970
From: andy at warmcat.com (Andy Green)
Date: Sun, 17 Jun 2018 07:17:34 +0800
Subject: [PATCH 07/11] ui-blame: free read_sha1_file() buffer after use
In-Reply-To: <20180616142303.GT1922@john.keeping.me.uk>
References: <152885510454.7253.3542488576272033383.stgit@mail.warmcat.com>
 <152885532551.7253.921107086774301587.stgit@mail.warmcat.com>
 <20180616142303.GT1922@john.keeping.me.uk>
Message-ID: <86926277-4f64-f91b-e8ee-f92ffbaaca03@warmcat.com>



On 06/16/2018 10:23 PM, John Keeping wrote:
> On Wed, Jun 13, 2018 at 10:02:05AM +0800, Andy Green wrote:
>> Signed-off-by: Andy Green <andy at warmcat.com>
>> ---
>>   ui-blame.c |    3 +++
>>   1 file changed, 3 insertions(+)
>>
>> diff --git a/ui-blame.c b/ui-blame.c
>> index 17e2d60..a5c7d69 100644
>> --- a/ui-blame.c
>> +++ b/ui-blame.c
>> @@ -206,6 +206,9 @@ static void print_object(const unsigned char *sha1, const char *path,
> 
> There's an early return above here after allocating buf, I'm not sure if
> we care about freeing buf on that path as well.
> 
> If we do, I guess we push free(buf) to the end and add a free_buf label
> for the error path to jump to.

Yes, I missed it thanks.  I adapted it with the label.

>>   	} else {
>>   		html_txt(buf);
>>   	}
>> +
>> +	free(buf);
>> +
>>   	html("</code></pre>");
>>   
>>   	html("</div></td>\n");
>>