Convert IATS Settings to be standard practice

[seamuslee001]

Iats Settings are currently done in a non standard way in a couple of different ways

1. There is currently no metadata e.g. title etc for the settings
2. All the settings are stored in a single array rather than as individual settings in the database as is the standard way for CiviCRM settings to be

This aims to address this by adding metadata for each individual setting and split up the array into individual settings rather than in the single array.

This function adds in helper function CRM_Iats_Utils::getSettings to handle the two different formats. This also has an upgrade step to perform the conversion.

The other thing this PR aims to do as per https://docs.civicrm.org/dev/en/latest/framework/setting/extension/ switches to using the Generic form for settings rather than a custom one.

[seamuslee001]

@adixon thoughts on this?

[adixon]

Thank you, this looks excellent. Nice when other people clean up my technical debt!

I'll merge into master so it has some time to get tested before the next release.

[KarinG]

And tested! We use iATS master in the WFC tests:

Backtrace is:

The website encountered an unexpected error. Please try again later.

TypeError: max(): Argument #1 ($value) must be of type array, string given in max() (line 256 of sites/default/files/civicrm/ext/com.iatspayments.civicrm/CRM/Core/Payment/iATSService.php).

CRM_Core_Payment_iATSService->doPayment() (Line: 116)
civicrm_api3_payment_processor_pay() (Line: 89)
Civi\API\Provider\MagicFunctionProvider->invoke() (Line: 156)
Civi\API\Kernel->runRequest() (Line: 79)

[adixon]

Thanks - that's a php8.? error unrelated to this pull, but I've just fixed it in master.

85f0a9d

Handy testing!

[KarinG]

Test matrix runs a couple of time per week -> and 4 days ago all was green.

It used PHP 8.1.25 for both for this failed run as well as 4 days ago - so you're right it may not be this particular PR but it's a very recent one.

Will report back after the next run (if I re-run now - it really just re-runs - it has cached the code it pulls in - and don't think we need to run a one off)

PS - since this WFC testing is where we have useful coverage for the iATS extension - I'd like to remove our CiviCRM core tests -> PR = #432

[KarinG]

Re: thanks - that's a php8.? error unrelated to this pull, but I've just fixed it in master

We still see the same error in the WFC test matrix (using iATS master) ->

@seamuslee001 - has another PR in the queue for review.