fix: limit/offset/where should be applied before json nested joins otherwise queries get slow. closes #3666 closes #3481 closes #3652.

[mswertz]

closes #3652.
closes #3476

What are the main changes you did:

• made sure filter, limit, offset, order by are applied before json is build

how to test:

• compare speed on this PR vs another PR. See example below

[mswertz]

BEFORE (jit=on)
https://emx2.dev.molgenis.org/CatalogueOntologies/tables/#/MedDRA?_page=4446

BEFORE (jit=off)
https://preview-emx2-pr-3704.dev.molgenis.org/CatalogueOntologies/tables/#/MedDRA?_page=4446

AFTER (jit=on)
https://preview-emx2-pr-3719.dev.molgenis.org/CatalogueOntologies/tables/#/MedDRA?_page=4446
(n.b. other graphql queries now become slowest such as getting the schema)

[mswertz]

go between page 36 and 37

BEFORE (with jit != off)
https://emx2.dev.molgenis.org/catalogue-demo/ssr-catalogue/ATHLETE/variables?page=37&view=harmonization

BEFORE (with jit=off):
https://preview-emx2-pr-3704.dev.molgenis.org/catalogue-demo/ssr-catalogue/ATHLETE/variables?page=36&view=harmonization

AFTER:
https://preview-emx2-pr-3719.dev.molgenis.org/catalogue-demo/ssr-catalogue/ATHLETE/variables?page=37&view=harmonization

[timcadman]

This is a big speed improvement, nice work :) I wasn't sure how to measure the times precisely as you did, but I can confirm that the PR speeds things up significantly.

[sonarcloud]

Quality Gate passed

Issues
0 New issues
0 Accepted issues

Measures
0 Security Hotspots
96.0% Coverage on New Code
0.0% Duplication on New Code

See analysis details on SonarCloud