Alexandrshy opened a new pull request, #9377:
URL: https://github.com/apache/devlake/pull/9377

   ### Summary
   
   Since #8931, Jira Cloud issues are collected via 
`/rest/api/3/search/jql?expand=changelog`, which embeds only the newest 40 
changelog histories per issue (the agile endpoint used before returned up to 
100). `collectIssueChangelogs` still fetched the full changelog only for issues 
with `changelog_total > 100`, so issues with 41–100 histories silently lost 
everything older than the newest 40, including their first status transitions.
   
   This PR lowers the threshold to 40.
   
   ### Does this close any open issues?
   
   None.
   
   ### Screenshots
   
   N/A
   
   ### Other Information
   
   - Verified locally with a throwaway test (not part of this PR) that runs the 
real `extractIssues` → `collectIssueChangelogs` → `extractIssueChangelogs` 
chain against a fake Jira Cloud server on MySQL 8.4. Before the fix, issues 
with 41, 60 and 100 histories kept 40; after it, all histories are stored, also 
on an incremental run. Issues with up to 40 or more than 100 histories, and 
epics, behave as before.
   - Jira plugin unit and e2e tests pass.
   - Extra cost: one API request per updated issue with 41–100 histories.
   - Histories lost before the fix come back after a one-off `fullSync` run of 
`collectIssueChangelogs` + `extractIssueChangelogs`.


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]

Reply via email to