artifact counts in RRDI
Has anyone experienced the following?
I have run across a behavior of RRDI [which i consider a defect] related to counting artifacts of type requirement [it may apply to other types as well].
The behavior is that when RRDI is asked to count the artifacts of type requirement, it counts the number of unique instances of the data warehouse field "requirement ID", rather than the data warehouse field "reference ID". Reference ID is the unique identifier for artifact, while "requirement ID" is unique identifier of EACH INSTANCE of the artifact, i.e. if you use the artifact in multiple locations in RM, e.g. module, collection, etc., RRDI returns the count of each time it is used rather than the true unique number of requirements. This behavior causes my reports to return the wrong number [count] for some reports I am running.
One answer
Could you provide the full URL of the defect? I'm not sure which repository the record is in. This would help for others looking for this issue.
I just want to note that the jazz.wiki page Reporting Quick Tips has a video on Counting the requirements using the identifier field. Maybe this would help.
Thanks.
Comments
Defect 80062 Requirement count metrics show inaccurate results
https://jazz.net/jazz03/resource/itemName/com.ibm.team.workitem.WorkItem/80062
Thanks Francesco.
https://jazz.net/jazz03/web/projects/Requirements%20Management#action=com.ibm.team.workitem.viewWorkItem&id=80062
I have [finally] viewed Petru's video of creating a report to count reference IDs. thank you. It looks like the report described in the video will provide the data I am needing.............however, I need that data vs time, days weeks, months. and the "out of the box RRDI report is still wrong [per my definition of reqs counting], and while Petru's custom report will work, it takes a sophisticated report engineer to build it. I will also note that, IMHO, the term "requirements type" in the video is Very misleading...........I believe a better term is "artifact type". In any case I look forward to resolution of this issue.
Comments
ross jones
Feb 24 '14, 1:05 p.m.this is now being worked as Defect 80062
1 vote