This is an automatically generated e-mail. To reply, visit: http://reviewboard-openlmi.rhcloud.com/r/837/ |
On September 20th, 2013, 12:56 p.m. UTC, Ales Ledvinka wrote:
src/journald/LMI_JournalMessageLogProvider.c (Diff revision 2) 33 #define CIM_MESSAGELOG_ITERATOR_RESULT_SUCCESS 0
This should map to the konkret's "OK" code rather then to exact number. If you disagree then see the PositionAtRecord method comment regarding returning of result.
As discussed face-to-face, this point should about CMPIStatus, not the result.
On September 20th, 2013, 12:56 p.m. UTC, Ales Ledvinka wrote:
src/journald/LMI_JournalMessageLogProvider.c (Diff revision 2) 49 #if 0
remove or is this future work?
Removed the block and added comment for journal_iters_destroy(). Leaving cleanup on CIMOM itself.
On September 20th, 2013, 12:56 p.m. UTC, Ales Ledvinka wrote:
src/journald/LMI_JournalMessageLogProvider.c (Diff revision 2) KUint32 LMI_JournalMessageLog_PositionAtRecord(236 KSetStatus(status, ERR_NOT_SUPPORTED);260 if (IterationIdentifier->null || ! IterationIdentifier->exists ||duverujem cimmomu/konkretu za kazdych okolnosti nebo radsi pridat: NULL == IterationIdentifier || ... NULL == MoveAbsolute NULL == RecordNumber ...
This is a konkret's job. If it fails to provide consistent arguments, we should fix it there at a first place.
On September 20th, 2013, 12:56 p.m. UTC, Ales Ledvinka wrote:
src/journald/LMI_JournalMessageLogProvider.c (Diff revision 2) KUint32 LMI_JournalMessageLog_PositionAtRecord(264 KUint32_Set(&result, CIM_MESSAGELOG_ITERATOR_RESULT_FAILED);This is useless since the status is not konkret's ok. See instead of filenameProvider.c see the filename.h the _Dispatch_Method() for more details on result processing of invoked method.
Fixed, revised status values completely. To be clear, result values (mapping to CMPIResult in konkret-generated code) stay the same as this is what clients get as a method call return value. Only CMPIStatus have been changed to reflect method call status.
On September 20th, 2013, 12:56 p.m. UTC, Ales Ledvinka wrote:
src/journald/LMI_JournalMessageLogProvider.c (Diff revision 2) KUint32 LMI_JournalMessageLog_PositionAtRecord(277 KUint32_Set(&result, CIM_MESSAGELOG_ITERATOR_RESULT_FAILED);missing status or status from previous operation. depending on the operation status if it's other then KOkay(status) match the line returning result might be useless depending on the status code returned
Again, this was meant for status, not the result.
- Tomáš
On September 20th, 2013, 10:05 a.m. UTC, Tomáš Bžatek wrote:
Review request for OpenLMI Developers.
By Tomáš Bžatek.
Updated Sept. 20, 2013, 10:05 a.m.
Repository:
openlmi-providers
Description
Diffs
|