Question : Why do the build for test case count is changing.

1.8 related questions and discussions.
Please upgrade to LATEST 1.9.x.
No more fixes for 1.8.

Moderators: Amaradana, TurboPT, TL Developers

Question : Why do the build for test case count is changing.

Postby Amaradana » Wed Mar 25, 2009 3:06 pm

Hi
This is a small scenario where we need to consider to implement this if not yet implement.

Work Flow
1. Created a test plan and 100 test cases were assigned and executed in build 1.
2. So in build 2, a new future is developed and 50 test cases were written, before assigning these to test plan i closed the build 1 and inactivated. now after assigning these 50 test cases to test plan my results for build 1 also changed since in the reports test case count was increased. and the overall status is changed for pass, fail, blocked which is not supposed to happen.

rather than showing this behaviour we need to freeze the build when it is closed and inactivated like when the test cases were added to test plan for the further buils the count should not be changed.

Please let me know if you do not understand this scenario, i am sure for working with you to try to make you understand.

Thanks,
Amardeep Maradana
http://amartesterworld.blogspot.com
Amaradana
Member of TestLink Community
 
Posts: 398
Joined: Mon Feb 16, 2009 11:19 am

Postby havlatm » Tue Mar 31, 2009 9:20 am

There is basic rule: One Test Plan has just one set of Test cases. So you can add during testing more new ones. And this affect metrics of older builds. you can two possibilities now:
- Create a new Test Plan for the second round.
- Use Keywords to recognize phases of testing in one Test Plan.
havlatm
Member of TestLink Community
 
Posts: 940
Joined: Mon Oct 31, 2005 1:24 am
Location: Czech

Postby Amaradana » Wed Apr 01, 2009 1:07 pm

Yes partly i figured out for my solution as you said the first case seems more resanable, but in organizational level while generating the report, rather than taking each of the reports ( 1 test plan and contains one build)and clubing, i feel that, it would be nice if the test cases were actually allocated to builds rather than test plan, since if we create in assumption that a test plan as a release, and the builds in it, and accordingly the results will be generated, This is one method and the other one if the build count varies also it is great since some organization look overalll status/progress for the release(as for our current functioianlity).

Let me know your thoughts Havlatm. we will discuss more about the functionality since i feel that there is a gap in the current functionality what we have. I am very glad to discuss in imporving any of the functionality

Let me know your thoughts.
Amaradana
Member of TestLink Community
 
Posts: 398
Joined: Mon Feb 16, 2009 11:19 am

Postby havlatm » Thu Apr 02, 2009 9:56 am

We had discussed it. We do not plan set relation between builds and TCs for performance and costs reason.
There is easier solution: trace time for these objects and generate reports with respect to it. Unfortunately, time information is not logged yet. So it needs DB enhancement and code enhancement to allow it.

It's not in current scope for 1.9 version as much as I remember. However it can change if you or anybody other contribute. Anyway, my feeling is that it's matter of time.
havlatm
Member of TestLink Community
 
Posts: 940
Joined: Mon Oct 31, 2005 1:24 am
Location: Czech


Return to TestLink 1.8 (OBSOLETE => EOL)[LOCKED]



Who is online

Users browsing this forum: No registered users and 1 guest