Fix sched-a by-employer,by-occupation and schedule_b/by_recipient fulltext and n/a search issue

[fec-jli]

Summary

This PR will fix fulltext search and N/A search issue in these three endpoints
ScheduleAByEmployerView
ScheduleAByOccupationView
ScheduleBByRecipientView

• Resolves determine if dsc_sched_a_aggregate_employer aggregates as intended #3885

• Related Add tsvector column to schedule A and schedule B aggregate data #5790

• Create follow up openFEC ticket to fix N/A issue on schedule-a and schedule-b endpoints
  cms ticket: N/A for schedule A datatable not filtering correctly from canonical page fec-cms#5220

Required reviewers

1-2 devs

Impacted three endpoints

"/schedules/schedule_a/by_employer/"
"/schedules/schedule_a/by_occupation/"
"/schedules/schedule_b/by_recipient/"

How to test

• Check out branch
• pytest
• test urls for three endpoints

—[1]ScheduleAByEmployerView
Test 1-1) when employer=amazon.com, return two rows:
employer=amazon.com
employer=amazon com
http://127.0.0.1:5000/v1/schedules/schedule_a/by_employer/?cycle=2020&employer=amazon.com&committee_id=C00003418

When employer=-amazon.com (exclude amazon.com), return 127487 rows
http://127.0.0.1:5000/v1/schedules/schedule_a/by_employer/?cycle=2020&employer=-amazon.com&committee_id=C00003418

Without employer parameter pass, return 2+ 127487= 127489 rows
http://127.0.0.1:5000/v1/schedules/schedule_a/by_employer/?cycle=2020&committee_id=C00003418

Compare with Prod: return 1 (only employer=amazon com, no return for employer=amazon.com)
https://api.open.fec.gov/v1/schedules/schedule_a/by_employer/?cycle=2020&committee_id=C00003418&api_key=DEMO_KEY&employer=amazon.com

Test 1-2)
when employer=N/A, return 1 rows
http://127.0.0.1:5000/v1/schedules/schedule_a/by_employer/?cycle=2020&employer=N/A&committee_id=C00003418

when employer=-N/A, return 127488 rows
http://127.0.0.1:5000/v1/schedules/schedule_a/by_employer/?cycle=2020&employer=-N/A&committee_id=C00003418

Without employer parameter pass, return 1+ 127488= 127489 rows
http://127.0.0.1:5000/v1/schedules/schedule_a/by_employer/?cycle=2020&committee_id=C00003418

Compare with Prod: return 962 rows (contain: n, a…)
https://api.open.fec.gov/v1/schedules/schedule_a/by_employer/?cycle=2020&committee_id=C00003418&api_key=DEMO_KEY&employer=N/A

—[2]ScheduleAByOccupationView
Test 2-1)
when occupation=C.E.O, return 46 rows
http://127.0.0.1:5000/v1/schedules/schedule_a/by_occupation/?cycle=2022&committee_id=C00003418&occupation=C.E.O

when occupation=-C.E.O, return 19665 rows
http://127.0.0.1:5000/v1/schedules/schedule_a/by_occupation/?cycle=2022&committee_id=C00003418&occupation=-C.E.O

Without occupation parameter pass, return 46+ 19665 =19711 rows
http://127.0.0.1:5000/v1/schedules/schedule_a/by_occupation/?cycle=2022&committee_id=C00003418

Compare with Prod: return 29 (occupation=c…e…o)
https://api.open.fec.gov/v1/schedules/schedule_a/by_occupation/?cycle=2022&committee_id=C00003418&occupation=C.E.O&api_key=DEMO_KEY

Test 2-2)
when occupation=N/A, return 1 rows
http://127.0.0.1:5000/v1/schedules/schedule_a/by_occupation/?cycle=2020&committee_id=C00003418&occupation=N/A

when occupation=-N/A, return 39455 rows
http://127.0.0.1:5000/v1/schedules/schedule_a/by_occupation/?cycle=2020&committee_id=C00003418&occupation=-N/A

Without occupation parameter pass, return 1+ 39455 = 39456 rows
http://127.0.0.1:5000/v1/schedules/schedule_a/by_occupation/?cycle=2020&committee_id=C00003418

Compare with Prod: return 193 rows (contain: n, a…)
https://api.open.fec.gov/v1/schedules/schedule_a/by_occupation/?cycle=2020&committee_id=C00003418&occupation=N/A&api_key=DEMO_KEY

—[3]ScheduleBByRecipientView
Test 3-1)
When recipient_name=zoom.com, return 1 row
http://127.0.0.1:5000/v1/schedules/schedule_b/by_recipient/?cycle=2022&committee_id=C00160937&recipient_name=zoom.com

When recipient_name=-zoom.com, return 126 rows
http://127.0.0.1:5000/v1/schedules/schedule_b/by_recipient/?cycle=2022&committee_id=C00160937&recipient_name=-zoom.com

Without recipient_name parameter pass, return 1+126=127 rows
http://127.0.0.1:5000/v1/schedules/schedule_b/by_recipient/?cycle=2022&committee_id=C00160937

Compare prod, return 0 (cannot find recipient_name=zoom.com)
https://api.open.fec.gov/v1/schedules/schedule_b/by_recipient/?cycle=2022&committee_id=C00160937&recipient_name=zoom.com&api_key=DEMO_KEY

Test 3-2)
when recipient_name=N/A, return 1 rows
http://127.0.0.1:5000/v1/schedules/schedule_b/by_recipient/?cycle=2018&committee_id=C00586768&recipient_name=N/A

when recipient_name=-N/A, return 17 rows
http://127.0.0.1:5000/v1/schedules/schedule_b/by_recipient/?cycle=2018&committee_id=C00586768&recipient_name=-N/A

Without recipient_name parameter pass, return 1+17=18 rows
http://127.0.0.1:5000/v1/schedules/schedule_b/by_recipient/?cycle=2018&committee_id=C00586768

Compare prod, return 0 (cannot find recipient_name=N/A)
https://api.open.fec.gov/v1/schedules/schedule_b/by_recipient/?cycle=2018&committee_id=C00586768&recipient_name=N/A&api_key=DEMO_KEY

Replace PR: #5773

[codecov]

Codecov Report

Attention: Patch coverage is 78.37838% with 8 lines in your changes are missing coverage. Please review.

Project coverage is 85.77%. Comparing base (49e8553) to head (f5cd046).
Report is 4 commits behind head on develop.

Files	Patch %	Lines
webservices/filters.py	63.63%	8 Missing ⚠️

Additional details and impacted files

@@             Coverage Diff             @@
##           develop    #5809      +/-   ##
===========================================
- Coverage    85.81%   85.77%   -0.05%     
===========================================
  Files           81       81              
  Lines         8594     8624      +30     
===========================================
+ Hits          7375     7397      +22     
- Misses        1219     1227       +8     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[pkfec]

@fec-jli Thanks for addressing the longstanding text search bug. Nice work.

[cnlucas]

@fec-jli great job on this ticket!
Could we either add testing for the new filter or create a follow-up ticket to add an automated test for the new filter?

[hcaofec]

Great work!
A follow-up ticket to add an automated test for the new filter will be created by @fec-jli