BDMS-139: Create Parameter table

[ksmuczynski]

Why

This PR addresses the following problem / context:

• Need a new Parameter table to store centralized metadata about each parameter, such as its category (e.g., 'Metals', 'Field Parameter') and its unique chemical identifier (CAS Number).
• The primary purpose of the Parameter table is is to ensure that every observation in the database is linked to a single, authoritative definition of what was measured.
• Need a new Regulatory_Limit table to store all the specific, citable standards and limits, regulatory and health-based, associated with a given parameter.
• The Regulatory_Limit table allows a single Parameter to have multiple different limits set by various agencies (e.g., a federal EPA limit and a state-level NMED limit).

How

Implementation summary - the following was changed / added / removed:

• Created a new Parameter table and populated with appropriate fields and relationships.
• Created a new Regulatory_Limit table and populated with appropriate fields and relationships.
• Updated the lexicon.json file to include valid values for the limit_type field.
• Added Parameter and RegulatoryLimit tables to the __init__.py file

Notes

Any special considerations, workarounds, or follow-up work to note?

• Related validation schema files and test files still need to be created or updated.
• Transfer script(s) needs to be created.

[codecov-commenter]

Codecov Report

❌ Patch coverage is 93.40102% with 13 lines in your changes missing coverage. Please review.
✅ All tests successful. No failed tests found.

Files with missing lines	Patch %	Lines
tests/conftest.py	93.51%	7 Missing ⚠️
core/initializers.py	81.25%	3 Missing ⚠️
tests/test_collabnet.py	0.00%	2 Missing ⚠️
core/app.py	50.00%	1 Missing ⚠️

Files with missing lines	Coverage Δ
api/search.py	97.77% <100.00%> (ø)
db/__init__.py	97.61% <100.00%> (ø)
db/lexicon.py	88.23% <100.00%> (ø)
db/observation.py	100.00% <100.00%> (ø)
db/parameter.py	100.00% <100.00%> (ø)
db/regulatory_limit.py	100.00% <100.00%> (ø)
schemas/observation.py	98.43% <100.00%> (ø)
schemas/parameter.py	100.00% <100.00%> (ø)
tests/__init__.py	100.00% <100.00%> (ø)
tests/test_observation.py	95.65% <100.00%> (ø)
... and 4 more

[jirhiker]

@ksmuczynski please update the lexicon.json file to use the new normalized structure. Then I will merge this PR

[jirhiker]

@ksmuczynski tests are failing. once they are fixed i will merge PR

[jacob-a-brown]

@jirhiker three updates needed to be made to ensure the tests past and that the API is working as expected:

• The scope of all test fixtures needed to be changed to function (the default), otherwise too many database connections occurred and we'd receive the error sqlalchemy.exc.TimeoutError: QueuePool limit of size 5 overflow 10 reached, connection timed out, timeout 30.00. Because the fixtures are now scoped to functions they are torn down after the test is complete. This is to keep the tests independent.
• After changing all of the fixtures the only test that was failing was tests/test_search.py::test_search_api. It would find the well and the spring, but not the contact. This happened because the concatenation of

vector = (
    Contact.search_vector
    | Email.search_vector
    | Phone.search_vector
    | Address.search_vector)

meant that if Contact, Email, Phone, or Address were NULL then the whole vector would be NULL. With the help of AI it was fixed by using

vector = (
        func.coalesce(Contact.search_vector, text("''::tsvector"))
        .op("||")(func.coalesce(Email.search_vector, text("''::tsvector")))
        .op("||")(func.coalesce(Phone.search_vector, text("''::tsvector")))
        .op("||")(func.coalesce(Address.search_vector, text("''::tsvector")))
    )

This error didn't occur previously because after the contact, email, phone, and address fixtures were invoked elsewhere they would persist. Now that invoking the contact fixture only creates a contact its related email, phone, and address attributes are NULL unless those fixtures are also invoked.

• For similar reasons, the search results for contacts should use outerjoin incase one of the attributes does not exist.