On 10/28/2009 10:44 AM, Daniel P. Berrange wrote:
> On Wed, Oct 28, 2009 at 03:32:32PM +0100, Chris Lalancette wrote:
>> Cole Robinson wrote:
>>
>>>
>>> static int umlListDomains(virConnectPtr conn, int *ids, int nids) {
>>> diff --git a/src/util/util.c b/src/util/util.c
>>> index 98f8a14..49eac6d 100644
>>> --- a/src/util/util.c
>>> +++ b/src/util/util.c
>>> @@ -1804,30 +1804,42 @@ int virDiskNameToIndex(const char *name) {
>>> #define AI_CANONIDN 0
>>> #endif
>>>
>>> -char *virGetHostname(void)
>>> +char *virGetHostname(virConnectPtr conn)
>>
>> To be honest, I'm not sure we even need to add the "conn" parameter here. I
>> seem to remember DanB saying that it was only needed in very few circumstances
>> anymore. Can't we just call "virReportSystemError(NULL)", etc? That keeps this
>> function signature a bit simpler. (It's not a deal-breaker in any case, just
>> nice to not have to require this parameter)
>
> That's correct, we should avoid adding virConnectPtr to any new /existing
> internal APIs where-ever possible. The only real exception is if the
> internal function needs to invoke one of the other public driver APis
> directly.
>
Ahh, I wasn't aware. I'll respin the patch with Chris' recommendations.
- Cole
--
Libvir-list mailing list
Libvir-list@xxxxxxxxxx
https://www.redhat.com/mailman/listinfo/libvir-list