This is an automatically generated e-mail. To reply, visit: http://reviewboard-openlmi.rhcloud.com/r/1570/ |
On April 2nd, 2014, 8:25 a.m. CEST, Michal Minar wrote:
src/software-dbus/LMI_MemberOfSoftwareCollectionProvider.c (Diff revision 2) static CMPIStatus LMI_MemberOfSoftwareCollectionAssociators(259 ci = _cb->bft->getInstance(_cb, cc, o, properties, &st);This does not speed up associators call much. There is no reason to call GetInstance() to obtain the instance we already have. I'd move GetInstance's body of SoftwareIdentity to some utility function so we could translate SwPackage to CMPIInstance here and anywhere else.
I've disabled enumerating the instances here, but I've implemented asoc. and ref. names functions.
On April 2nd, 2014, 8:25 a.m. CEST, Michal Minar wrote:
src/software-dbus/sw-utils.h (Diff revision 2) short check_and_create_error_msg(PkResults *results, GError *gerror,123 * @return 0 if cm_class is type of type, 1 if not,
124 * negative value in case of an error
125 */
126 short cm_class_is_a(const CMPIBroker *cb, const char *ns, const char *cm_class,127 const char *type);*is* in cm_class_is_a() suggests that the result is a boolean. Function has inverse characteristic though. Please consider using *cmp* in name instead or swap return values. I would ignore error case (-1) all together.
I've updated the return status.
- Peter
On April 3rd, 2014, 6:33 p.m. CEST, Peter Schiffer wrote:
Review request for OpenLMI Developers.
By Peter Schiffer.
Updated April 3, 2014, 6:33 p.m.
Repository:
openlmi-providers
Description
Diffs
|