[Koha-bugs] [Bug 23681] Patron restrictions should be user definable

[bugzilla-daemon at bugs.koha-community.org]

https://bugs.koha-community.org/bugzilla3/show_bug.cgi?id=23681

--- Comment #57 from Andrew Isherwood <andrew.isherwood at ptfs-europe.com> ---
(In reply to Jonathan Druart from comment #54)
> (In reply to Andrew Isherwood from comment #1)
> > It has become increasingly apparent that using authorised values for this is
> > not the most appropriate way forward. The number of modifications and
> > additions to AV was increasing the scope of the work too much, it was clear
> > that AV was not the correct place for this.
> 
> Why? Can you explain?
> That was my first reaction reading the patches.
> Using AV will make the feature translatable in the future (bug 20307).

Clearly AV would have been the preferable route, which is why I spent some
considerable time trying to implement using AV before reluctantly abandoning
it, it certainly wasn't abandoned lightly. There were too many requirements of
this work that were not provided by AV and working around them was becoming
increasingly cludgy and unmaintainable.

This work was done a few months ago, so my memory of it isn't what it was, but
examples would be:
- the ability to specify a default restriction type, which any restrictions
would fall back to if the restriction type they were using was deleted
- the ability to prevent certain restriction types from being edited
- the ability to prevent certain restriction types from being deleted

> So you are adding a regression, the restrictions are no longer translatable.

They were previously held in borrower_debarments.type, they are now in
debarment_types, so I'm guessing adding translation support for the default
values is not an insurmountable problem, I am happy to add whatever work is
required to make that happen to this bug. Any additional values are added by
the user anyway, so are by their nature translatable.

>Is the "default" flag really useful? We could imagine a new "internal" flag (to replace ronly) for AVs.

I believe so. Restrictions need a type, even if one is not explicitly chosen.
Also, we need an ability to say "if a restriction type is deleted, any
restrictions that use this type should fall back to type X", the default flag
indicates which type is X.

-- 
You are receiving this mail because:
You are watching all bug changes.