dev/translation#58 - Fix regression. Partially revert schema for civicrm_group.title

[totten]

Overview

This aims to address a recent regression in 5.31-rc. It partially reverts a change where #18599 marked the civicrm_group.title field as <required>true</required>.

It fixes a regression and failing unit-test that can be observed in bknix-min.

See also: https://lab.civicrm.org/dev/translation/-/issues/58

Before and After

• 5.30: From a SQL POV, civicrm_group.title is not required (DEFAULT NULL)
• 5.31 (pre-patch): From a SQL POV, civicrm_group.title is required (NOT NULL)
• 5.31 (post-patch): From a SQL POV, civicrm_group.title is not required (DEFAULT NULL)

Comments

I'm not certain if a revert is workable - but don't see an obvious drawback. It's simple and worth trial.

Even with this schema-level patch, the UI (civicrm/group/add?reset=1) still treats the field as required.

@artfulrobot @seamuslee001 - in the discussion of #18599, was there a specific reason for flagging the field required (perhaps something discussed elsewhere)? From the Github note, it doesn't sound too important?

@bgm - any experience with similar bugs before? thoughts?

[civibot]

(Standard links)

• If this is your first pull-request for CiviCRM, please browse CONTRIBUTING.md for information about the development and testing processes.
• If you are reviewing this pull-request, you may wish to consult the test sites and the Review Standards (long template, short template).

[artfulrobot]

@totten in my comment I suggested the title should be not null, and @seamuslee001 suggested to make it required.

I'd said not null because I didn't think we needed to differentiate between NULL and ''; I tend to see NULL-able columns as an optional feature for when you need that, rather than a default. So I was suggesting that it could be NOT NULL (and default to ''), which is different to suggesting it's required. But I probably did think it should be required - what use is a group without a title? (hmmm. "hidden smart groups" for search results that become mailings maybe...)

Anwyay, as far as I'm concerned, I have no reason to think it must be required; so I'm happy to revert it.

[seamuslee001]

I'm fine to revert as well

[totten]

Thanks, @artfulrobot @seamuslee001.

I think you're right that the NOT NULL makes general sense from SQL modeling point-of-view -- so it was worth trying This interaction is a bit mysterious / hard-to-foresee, but it's helpful that we can back it out for now.