Check accrualPeriodicity against list of valid frequency values.

[beeyayjay]

Checks the validity of the accrualPeriodicity field in prepareRow().

User story

If a json harvest has an accrualPeriodicity value, that value gets saved to the frequency field, even if the value isn't valid for the field.

The pr adds a check to prepare row. If the value is a valid key value for field_frequency, there is no change. If it is a valid display value, it is set to the key value. If it is neither, it is set to null.

How to reproduce

1. Run a harvest against a harvest cache with an invalid value for accrualPeriodicity.
2. Browse to the dataset page. The invalid value will be displayed as the frequency.
3. Edit the page. The select value for the frequency field will be set to '- None -'.

QA Steps

1. Create/modify three harvest cache files with accrualPeriodicity fields.
   a. Give one a valid Frequency key (e.g. R/P1D).
   b. Give another a valid Frequency value (e.g. Daily).
   c. Give the third an invalid value
2. Run the harvest.
3. Check the harvested datasets. Frequency should appear with its correct value for the first two on both the dataset page and its edit form. Frequency should not appear on the dataset page for the third.

[beeyayjay]

Wait on this. I just realized the code probably belongs in dkan_migrate_base_data_json.inc instead.

[janette]

@beeyayjay since this is specific to POD endpoints could you put the code in here https://github.com/GetDKAN/dkan/blob/7.x-1.x/modules/dkan/dkan_harvest/modules/dkan_harvest_datajson/dkan_harvest_datajson.migrate.inc#L81 and add a message for the log when an invalid value is detected and ignored.