Would it be possible to extract repository information for a set of tasks

[Keith Goodlip]

I would like to export the repository name for tasks which have some form of development linked to said task.

Currently this is only done one at a time and it is very time consuming. It would be nice if this was possible.

Product: Jira Cloud

[Andy Heinzer]

Hi Keith,
I understand that you are using Jira Cloud and want to be able to export the linked repository name for any development related data to these Jira issues.  This is not currently possible to do, but I will try to explain the different ways I have tried to solve this so far.

Jira does store some of this development data from other tools (such as Bitbucket, Github, etc).  And you can add the issue field called 'development' to see some of this.  However when you export this data via CSV, you will see that the json payload in that development field does not actually contain the repo name anywhere.  Hence there does not seem to be a way to do this today via that export.

Granted you can click the development link on the Jira issue and the window that appears will show you that repo information including the repo name, but that information is not fully accessible from the export.   If you were using Jira Server, then it might be possible to extract this data directly from the SQL database.  There is a KB guide to do this over in How to extract information related to the GitHub DVCS connector from the database.  However that is not a viable option here, because Jira Cloud sites do not have access to the SQL database to make this kind of query.  It might be possible to figure out this connection if you took a backup of your data in Cloud and imported it to a Jira Server site, but that is potentially a lot of work just to get access to the SQL database for 1 or 2 SQL queries to make.  So I don't really see this as a great solution.

I also looked to see if there was a means to do this via automation.  My thought was perhaps we can setup a rule to populate that repo name into a custom field of Jira. But the page on Smart values – development again stops this idea because it too does not have the ability to 'see' that repo name.  I tried using {{branch.url}}, which I was hoping would contain the repo name in the URL, but I found that at least Bitbucket is not showing the actual repo name in the URL, but rather a repo id number instead.

I also took a look through our existing feature requests on https://jira.atlassian.com/issues/ for both the JRACLOUD and JSWCLOUD projects, but I did not yet find an existing request here that I think matches your use case.  I would be interested to see if we can create a new feature request here for this.  Could you tell me more about what problem this would solve for you?  I just want to make sure I understand why this is a problem, and how that info helps you so that I can craft a new request properly.

Sorry I don't have a solution yet, but please let me know.

Andy

[Keith Goodlip]

Hi Andy,

Thank you for taking the time to reply at length to me on the task at hand.

In regards to your question I'm using this during a change management audit. I'm exporting the requests together with some data that I require. My JQL is as follows (redacted some confidential info)

project = project-x AND type in (task, bug, story) AND (status changed to Done during (2019-06-01, 2020-05-12) OR status changed to Closed during (2019-06-01, 2020-05-12) ) AND status in (Done, Closed) AND "Brand(s)" in (brand-a, brand-b, brand-c, brand-d, EMPTY) ORDER BY statusCategoryChangedDate ASC

The export contains 

Key, Status Category Changed, Summary, Assignee, Reporter, Development, Prio, Status and some custom fields

This is were this feature comes in handy. From all the issues I've exported only those issues that use a handful of repositories would be in scope. Currently this involves our teams to go through each issue to see if merged requests have used any particular repository. This is very time consuming and takes several days to go through. To be honest there are other factors that filter in, but having this particular feature would shorten the time needed to go through all issues considerably.

Let me know should you have any further questions regarding the task I'm trying to solve.

[Andy Heinzer]

Thanks for explaining in more detail here, Keith.  I did some additional searching and I came across an existing request over in JSWCLOUD-18873 : Use the repo name instead of the ID when visualising a PR link from Bitbucket in the dev panel.  However this is not exactly what I think you are looking for.

In your case you would really need to be able to see this repo name in the csv export itself that Jira creates, not just the development panel itself in order to be useful to you here.   To that end I created this new feature request in JSWCLOUD-18966 : Exporting Jira issues via CSV should also include linked development information such as repository name.

Sorry I don't have a good work-around for this at this time.  But I would recommend watching this issue.  If there is a better solution found, I would expect that issue to be updated with some other means to do this.

Andy