fix(api): return total count on related endpoint

[villebro]

SUMMARY

While reviewing/testing #16144 , I came across a TODO related to the /related endpoints not returning total object count, but rather the count of results on the requested page.

This PR makes the following changes:

• When include_ids is undefined, always return the total number of objects as the count parameter as opposed to the actual number of objects in result.
• When include_ids is defined, behavior is unchanged, i.e. count equals the length of result.
• When include_ids is defined and page > 0, a 400 is returned due to pagination of mixed results not being possible.

BEFORE

Currently the count parameter in the response is equal to the length of the result array (in this example, the total number of objects is in fact two, not just one):

AFTER

Now the count parameter shows the total number of objects, despite the page size being less:

When sending the request with include_ids and page=0, the result is the same as before (notice how page_size only affects the values not included in include_ids):

Now requesting other than the first page with defined include_ids will result in a 400:

BEFORE/AFTER SCREENSHOTS OR ANIMATED GIF

TESTING INSTRUCTIONS

ADDITIONAL INFORMATION

• Has associated issue:
• Required feature flags:
• Changes UI
• Includes DB Migration (follow approval process in SIP-59)
  • Migration is atomic, supports rollback & is backwards-compatible
  • Confirm DB migration upgrade and downgrade tested
  • Runtime estimates and downtime expectations provided
• Introduces new feature or API
• Removes existing feature or API

[michael-s-molina]

Should we send a 422 instead? https://softwareengineering.stackexchange.com/questions/329229/should-i-return-an-http-400-bad-request-status-if-a-parameter-is-syntactically

[villebro]

That's a good idea - will update 👍

[codecov]

Codecov Report

Merging #16397 (c108922) into master (970d762) will decrease coverage by 0.30%.
The diff coverage is 67.62%.

❗ Current head c108922 differs from pull request most recent head dc8a4b9. Consider uploading reports for the commit dc8a4b9 to get more accurate results

@@            Coverage Diff             @@
##           master   #16397      +/-   ##
==========================================
- Coverage   76.64%   76.33%   -0.31%     
==========================================
  Files        1000     1000              
  Lines       53489    53491       +2     
  Branches     6816     6818       +2     
==========================================
- Hits        40996    40834     -162     
- Misses      12257    12419     +162     
- Partials      236      238       +2     

Flag	Coverage Δ
hive	?
mysql	81.49% <85.59%> (+<0.01%)	⬆️
postgres	81.55% <86.55%> (+<0.01%)	⬆️
presto	?
python	81.64% <86.55%> (-0.42%)	⬇️
sqlite	81.20% <85.59%> (+<0.01%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

Impacted Files	Coverage Δ
...c/views/CRUD/data/database/DatabaseModal/index.tsx	44.24% <0.00%> (-0.12%)	⬇️
superset/charts/commands/importers/dispatcher.py	80.64% <ø> (ø)
superset/cli.py	52.85% <0.00%> (ø)
superset/commands/importers/v1/__init__.py	94.73% <ø> (ø)
superset/commands/importers/v1/examples.py	39.53% <ø> (ø)
superset/common/tags.py	0.00% <0.00%> (ø)
superset/dashboards/api.py	93.03% <ø> (ø)
...perset/dashboards/commands/importers/dispatcher.py	83.87% <ø> (ø)
superset/dashboards/commands/importers/v0.py	89.94% <ø> (ø)
superset/dashboards/schemas.py	99.37% <ø> (ø)
... and 80 more

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 970d762...dc8a4b9. Read the comment docs.

[zhaoyongjie]

Tested in my local environment, LGTM. Thanks for the improvement.