User tests: Successful: Unsuccessful:
Pull Request for Issue #27307
This is reverting of #22757 which may cause loss of user data in specific cases, because of duplicate field names (group of fields).
Please look #27307 for detail
You do not have duplicated field names
You have duplicated field names
Is there a possibility to fix the issues that #22757 dealt with in another way?
Unfortunately there no easy way to do it, and practically there no much reason to spend a time for it.
I do not take that as a bug. it just a wish of perfectionists :)
In general, random row numbers does not lead to any functional issue.
Anyway it does not possible to keep "perfect numbering" after all. Add/remove/sort rows will break numbering.
I preparing similar pull request for j4, there may be will be possible 0,1,2.. numbering, but only until first remove/move.
I would mark that issues as "won't fix".
Regardless of further discussions. Please(!) don't hesitate merging this pr in 3.9.15 as-is and solve the other problem with another pr then.
This pr fixes a severe bug (data loss) and is already tested successfully.
|Status||Pending||⇒||Ready to Commit|
|Status||Ready to Commit||⇒||Fixed in Code Base|
|Closed_Date||0000-00-00 00:00:00||⇒||2019-12-26 22:28:23|
It's been merged so no :) Release blocker tags are a way for me and harald to track open issues/pr's that block us making a release