CRM-21225 Allow for display profile alternate - schema changes only

[seamuslee001]

Overview

this picks up just the schema changes from https://github.com/civicrm/civicrm-core/pull/11370/files this is in the hope to try to get this part merged through

Before

No diplay_title column on civicrm_uf_group

After

display_title added

ping @eileenmcnaughton @agileware @JoeMurray

---

• CRM-21225: CIVICRM-506 Add a new field, Profile Title so that the Profile Name does not serve a dual purpose of labelling the Profile and displaying the Title for the Profile fields publicly

[eileenmcnaughton]

This seems good to me - do we have a precedent anywhere for naming conventions for front end title fields vs back office title fields?

[seamuslee001]

@eileenmcnaughton @totten @jusfreeman I have now renamed the column frontend_title as per discussion on MM. I'll work on the main code once the schema change has been merged

[seamuslee001]

Jenkins re test this please

[eileenmcnaughton]

We discussed this in chat. The situation is that there is a need for 3 'names'

a symbolic, unchanging machine name
a backend/administrative title (used to browse/select among profiles)
a frontend/page title (like the '<h1'> on a public page)

We have a standard of name for the first & title kinda covers 2 & 3. After discussion we preferred adding 'frontend_title' to display_title (display when?)

Happy with this change now - merge on pass - it could be extended to other entities like custom data group etc

[gah242s]

This is breaking multilingual + logging upgrades. The select statement generates FROM civicrm_uf_group_en_US, which I believe should be civicrm_uf_group. Here's where the field name shows up in the error: [nativecode=1054 ** Unknown column 'a.frontend_title' in 'field list']. Going with a very uneducated guess, I'd think that the error may lie in line 493 of CRM/Upgrade/Incremental/php/FourSeven.php. Does there need to be another FALSE argument at the end?
(Edited to replace installs with upgrades.)

[seamuslee001]

@gah242s are you able to put the backtrace into a gist or similar? there could well be something going strange it could also be that the multilingual views haven't properly re-generated

[gah242s]

https://gist.github.com/gah242s/2fbebbb638b8a2065de9eeca7e2a989f
Here's a gist, also posted in MM. Hopefully it's useful?