It's all about the answers!

Ask a question

RTC: how to withdraw a changeset?


Hao Wang (111) | asked Mar 08 '13, 5:43 p.m.
The question is simple: after I made some code changes in RTC, I created a new changeset from them. I submitted for review and the review is pending. Now I discovered some problem in my change. How can I withdraw the review and the changeset so I can correct the problem, merge the new change to the changeset and re-submit it for review?

Thank you!

3 answers



permanent link
VK L (8177145159) | answered Mar 09 '13, 3:45 a.m.
 Hi Hao,
              you can discard the changeset and then check-in/deliver your new code. As for the review part, is the workitem sent for a review?

Thanks.

Comments
Hao Wang commented Mar 10 '13, 3:46 a.m. | edited Mar 10 '13, 9:17 p.m.

Hi Valli,


Thank you for the reply.

Yes, the changeset and associated work item has been sent for review. The review was pending. Later I asked the reviewer to reject it. So now it is rejected.

Since I made two changes to one file in the code - change 1 made before creating the changeset and change 2 made after creating the changeset. I believe if I discard the changeset I would lose both change 1 and 2. The file would go back to the state before change 1. Is this the expected behavior? Basically, what I wanted is to have the file keep both change 1 and 2 and the changeset removed. Is there a way to achieve this?


VK L commented Mar 10 '13, 4:25 a.m.

Have you delivered the changeset? changeset is the collection of the 2 changes - so if you just want to remove the identifier  - edit the changeset comment (or) you can relocate the changes to different changesets.

Anyways, after modifying the files, you can associate it to a different changeset as well


Geoffrey Clemm commented Mar 10 '13, 9:36 p.m.
FORUM ADMINISTRATOR / FORUM MODERATOR / JAZZ DEVELOPER

Hao: When you say "I want to keep both changes to the file but remove the change set" ... what did you mean by "remove the change set"?  Every file change is part of a change-set (or stated otherwise, a change set is just a set of changes).  Note that once you complete a change set, you cannot remove a change from it, or add a change to it.  When you deliver a change set, it is automatically completed.


Hao Wang commented Mar 10 '13, 9:37 p.m.

I haven't delivered the changeset. It was rejected by the reviewer. The changeset only contains the first change. Below are the steps I did.


1. Made change 1 to the file.
2. Created a changeset from change 1.
3. Submit the changeset for review.
4. Made change 2 to the same file.
5. The review is rejected.

I would like to go to a state where the file contains cumulative changes from both change 1 and 2. And the changeset is removed. The reason for this is simple: the reviewer or myself found some problem with my change 1 so I want to correct the problem by adding change 2 to the changeset. However, this is not allowed because the changeset is not active in the workspace. What would you do in this case?

Thanks a lot!


permanent link
Geoffrey Clemm (30.1k23035) | answered Mar 10 '13, 10:15 p.m.
FORUM ADMINISTRATOR / FORUM MODERATOR / JAZZ DEVELOPER
The main thing to keep in mind is that the "approval" is on the work item, not the change set.   The change sets are just ways to associate changes to the work item.   So in particular, if a work item has been rejected, then the normal approach would be to add another change set to the work item that fixes the problem identified by the reviewer.   So there is no need to modify or remove the existing change sets ... just associate the new ones with the existing work item, and and add a new review request to the work item.

permanent link
VK L (8177145159) | answered Mar 11 '13, 8:00 a.m.
 If the changeset is not yet delivered, you can continue to make change to the file -> and the change should be auto-associated to the changeset - so you should be able to re-submit it for a review now.

You can verify your workspace - to ensure that the change is associated with the same chnageset.

Thanks.

Comments
Hao Wang commented Mar 22 '13, 12:06 a.m.

 Thanks.

Your answer


Register or to post your answer.