No permission in permission.perm_list that give permission to override MAX_HOLDS

[Blake GH]

When a patron is at max holds and someone is attempting to add another hold this error occurs:

"The patron has reached the maximum number of holds"

Currently the only way to override this is with the permission "Everything". The system needs MAX_HOLDS.override defined in permission.perm_list.

This is the string required as defined by PermitHold.pm line 211 where the LEGACY_HOLD_EVENT_MAP hash map defines 'config.hold_matrix_test.max_holds' => 'MAX_HOLDS'

I fixed this in our consortium:

insert into permission.perm_list(code,description)
values('MAX_HOLDS.override','Allows user to override the "The patron has reached the maximum number of holds"');

[Mike Rylander (mrylander)]

FWIW, for sites that lack but need this permission, you should be able to add it through the staff client under Admin -> Server Admin -> Permissions.

[Blake GH (bmagic)]

http://git.evergreen-ils.org/?p=working/Evergreen.git;a=commit;h=8b00406ae17d884b2db936a041bb4973b2fe1a06

[Ben Shum (bshum)]

Hmm, reviewed Blake's commit and have some thoughts...

1) In the upgrade script provided, he uses a hardcoded ID of 555 (but we've already gone past this number in stock seed data)

2) We need a change to the seed data file (Open-ILS/src/sql/Pg/950.data.seed-values.sql) to add the permission for fresh system installs. At which point in there, we'll define a proper ID for the new permission.

3) Do we need this permission applied to any staff groups by default (circ administrator, etc.?)

For the upgrade script, while I'm generally reluctant to specify an exact ID for the permissions, since these can and tend to be wildly variable in actual production systems, the expectation is that any custom permissions live above ID 1000, so anything under that is safe to insert. That said, we should add some protection in the upgrade script to fail peacefully if the permission was already added manually by the consortium in the past.

Marking back to "In Progress" and assigning to Blake for now to give this some more poking, but he may seek further assistance from folks with more DB upgrade script writing experience. And also, does this quality for a test plan and possible PGTap test?

[Blake GH (bmagic)]

Ben,

Good point. I reworked it allowing it to get the next available ID number below 1000. I created the pgTAP test as well.

http://git.evergreen-ils.org/?p=working/Evergreen.git;a=shortlog;h=refs/heads/user/blake/LP1320301