We have a custom phabricator extension to handle train blocker tasks.
Entering a non-conforming task number causes an exception see code in ReleaseDetailsCustomField.php
Description
Description
Details
Details
Related Changes in GitLab:
| Title | Reference | Author | Source Branch | Dest Branch | |
|---|---|---|---|---|---|
| Refactor handling of version ranges by ReleaseDetailsCustomField | repos/phabricator/extensions!25 | dduvall | review/T304540 | wmf/stable |
Related Objects
Related Objects
- Mentioned In
- T362689: "Undefined offset: 5" error when visiting https://phabricator.wikimedia.org/T361395
rPHEX1b5c4a44369d: Fix resolution of next version in the series for release details - Mentioned Here
- rPHEXce33d8fa9ca8: Avoid linking to wrong previous train task in train blocker tasks
T362689: "Undefined offset: 5" error when visiting https://phabricator.wikimedia.org/T361395
T330797: Uninstall Differential (Phabricator application)
Event Timeline
Comment Actions
Mentioned in SAL (#wikimedia-releng) [2022-03-23T17:35:25Z] <brennen> restarting phabricator for T304540, brief downtime expected
Comment Actions
@dduvall: Would you be willing and have time to transfer your patch in D1200 into a MR in https://gitlab.wikimedia.org/repos/phabricator/extensions ?
It's the last open patch we have in Differential which we'd like to switch off per T330797. If not I can do so - just don't want to steal credits from you. :) TIA!
(However does so in the end, please do abandon the patch in D1200 afterwards. Thanks.)
Comment Actions
dduvall opened https://gitlab.wikimedia.org/repos/phabricator/extensions/-/merge_requests/25
Refactor handling of version ranges by ReleaseDetailsCustomField
Comment Actions
brennen merged https://gitlab.wikimedia.org/repos/phabricator/extensions/-/merge_requests/25
Refactor handling of version ranges by ReleaseDetailsCustomField
Comment Actions
Entering a non-conforming task number causes an exception see code in ReleaseDetailsCustomField.php
Did you mean version string (usually sth like 1.99.99-wmf.99) instead of task number?
Does anyone have an exception text / stacktrace?
I assume this is much less likely to happen after rPHEXce33d8fa9ca8582a428fcfa58ad6f99ad5bc3b89.
Comment Actions
No reply; closing for now as there is no exception stacktrace and per my last comment.
Please reopen once there are clearer steps to reproduce, a stacktrace, and if this still happens.
PS: Also, if anyone ever wanted to experience the variety of random stuff put into the Release Version freetext field,
SELECT CONCAT('T',t.id), ff.rawCorpus FROM phabricator_maniphest.maniphest_task_ffield ff INNER JOIN phabricator_maniphest.maniphest_task_fdocument fd ON fd.id = ff.documentId INNER JOIN phabricator_maniphest.maniphest_task t ON t.phid = fd.objectPHID WHERE ff.fieldKey = "!jdY";
provides some great inspiration.