LiquidPlanner Classic Forum

*Posted on behalf of a customer. Original posting date 2015-01-19.* In speaking with a newer client, they are intending to have updates to estimates on many JIRA sub-tasks all feed into / roll up to a single LiquidPlanner task. They would anticipate having many groups of sub-tasks feeding to many tasks. I was having trouble conceptualizing how this might work via API - would it be possible for sub tasks to be updated with some kind of ID field that would allow them to all associate with one task? Might this group be forced to consider an architecture where JIRA subtasks have a 1 to 1 relationship with a LiquidPlanner task?

*Posting on behalf of a customer. Original posting date 2014-11-21.* We have spent about 4 weeks working on integrating JIRA with LiquidPlanner and still are probably still a few weeks away from getting basic functionality to work. I feel like this has been done before by someone/somewhere/somehow. If anyone has any code they are willing to share it would be appreciated. We are using webhooks in both platforms with the API and are trying to do a two way sync. We have stumbled and punted on issues like how comment edits are handled and workflow/card sync. So if there is full blown file that has been executed in the past (with sensitive data removed) it would be greatly appreciated. PS. I understand that every implementation is different but having an example of how someone has built it out would be super helpful.

*Posted on behalf of Thibaud. Original posting date 2014-10-27.* Hi, We use LiquidPlanner (LP) to manage projects (mostly for planning) but JIRA for the time tracking. We developped an application that synchronises JIRA to LP (one-way). So we can create new tasks, update info (owner/assignee, low/high estimate, description, etc.). We don't mind much about work log (who does what) we just want to make sure the time left, assignee and done status of tasks in JIRA are properly updated in LP. With the new changes, the part that I don't quite get is the multiple assignments. As there is no way to test it before it's too late, this scares me a bit because having the synchronization work became critical in our organization. When all I want is to make sure the owner of the task reflects the assignee in JIRA and that the remaining estimates (low and high) are up to date (we use the track_time service to update the difference between JIRA and LP), what are the precautions I have to take ? Because I read that assigning the same person to a task will throw an error. Do I use the track_time service the same way I used it in the past (but adding the person_id) for time tracking ? Do I have to delete extra assignments ? If we want to have only one assignee, do we simply update an assignment based on its ID ? My plan is to do this: 1- update all task info except estimates and ownership 2- update the owner: 2.a. if the task.assignments[0].person_id == <owner id> => do nothing (is there a risk of having multiple assignements?) 2.b. otherwise, send a request to /update_assignment with {"assignment_id": task.assignments[0].id, "person_id": <owner id>} 3- update the estimates (I use the same request to /time_tracking with the difference in estimates in JIRA and LP and the person_id being the <owner id>) Is that OK ? I have a feeling that there's a risk of having assignments with person_id: 0 (Unassigned user) or worse, if people for a reason or another change ownership (and log work??) in LP, there will be multiple assignments. Sorry if my questions aren't clear enough, if you need any clarification, don't hesitate to ask. Thanks for the help. - Thibaud -