[PATCH stable v2] parsers.c: fix a couple of memory leaks

[Siddharth Agarwal]

On 07/03/2014 09:25 PM, Kevin Bullock wrote:
> On Jun 12, 2014, at 6:08 PM, danek.duvall at oracle.com wrote:
>
>> # HG changeset patch
>> # User Danek Duvall <danek.duvall at oracle.com>
>> # Date 1402525864 25200
>> #      Wed Jun 11 15:31:04 2014 -0700
>> # Branch stable
>> # Node ID ed93ca72e5f22b29624394a484668c153020882a
>> # Parent  b35f8c487e396487e89f98e92da57ac5eb9833af
>> parsers.c: fix a couple of memory leaks
> Appears to have been queued without response as 8da100383dc3. (Hi patchbot!)

My fault -- I didn't hit reply all.

>
> pacem in terris / мир / शान्ति / ‎‫سَلاَم‬ / 平和
> Kevin R. Bullock
>
>> diff --git a/mercurial/parsers.c b/mercurial/parsers.c
>> --- a/mercurial/parsers.c
>> +++ b/mercurial/parsers.c
>> @@ -1299,7 +1299,7 @@ static PyObject *find_deepest(indexObjec
>>   	static const Py_ssize_t capacity = 24;
>>   	int *depth, *interesting = NULL;
>>   	int i, j, v, ninteresting;
>> -	PyObject *dict = NULL, *keys;
>> +	PyObject *dict = NULL, *keys = NULL;
>>   	long *seen = NULL;
>>   	int maxrev = -1;
>>   	long final;
>> @@ -1403,8 +1403,10 @@ static PyObject *find_deepest(indexObjec
>>   		final |= i;
>>   		j -= 1;
>>   	}
>> -	if (final == 0)
>> -		return PyList_New(0);
>> +	if (final == 0) {
>> +		keys = PyList_New(0);
>> +		goto bail;
>> +	}
>>
>>   	dict = PyDict_New();
>>   	if (dict == NULL)
>> @@ -1428,19 +1430,13 @@ static PyObject *find_deepest(indexObjec
>>
>>   	keys = PyDict_Keys(dict);
>>
>> -	free(depth);
>> -	free(seen);
>> -	free(interesting);
>> -	Py_DECREF(dict);
>> -
>> -	return keys;
>> bail:
>>   	free(depth);
>>   	free(seen);
>>   	free(interesting);
>>   	Py_XDECREF(dict);
>>
>> -	return NULL;
>> +	return keys;
>> }
>>
>> /*
>> _______________________________________________
>> Mercurial-devel mailing list
>> Mercurial-devel at selenic.com
>> http://selenic.com/mailman/listinfo/mercurial-devel
> _______________________________________________
> Mercurial-devel mailing list
> Mercurial-devel at selenic.com
> http://selenic.com/mailman/listinfo/mercurial-devel