feat(driver-adapters): fix bigint handling on creation and filter

[jkomyno]

This PR closes https://github.com/prisma/team-orm/issues/656 and https://github.com/prisma/team-orm/issues/836.

Please read this comment.

[github-actions]

WASM Size

Engine	This PR	Base branch	Diff
WASM	2.338MiB	2.338MiB	2.000B
WASM (gzip)	907.885KiB	907.788KiB	100.000B

[github-actions]

✅ WASM query-engine: no benchmarks have regressed

Full benchmark report

DATABASE_URL="postgresql://postgres:postgres@localhost:5432/bench?schema=imdb_bench&sslmode=disable" \
node --experimental-wasm-modules query-engine/driver-adapters/executor/dist/bench.mjs

After changes in 57c5acb

[codspeed-hq]

CodSpeed Performance Report

Merging #4648 will not alter performance

_{Comparing fix/driver-adapters-bigint (57c5acb) with main (9c3f205)}

Summary

✅ 11 untouched benchmarks

[jkomyno]

Note: planetscale (wasm) is failing for reasons seemingly independent of this PR.

[jkomyno]

Note: planetscale (wasm) is failing for reasons seemingly independent of this PR.

I understood the problem, for which I don't have yet a solution: on PlanetScale, bigint values are placed into queries as strings, resulting in SQL snippets like

LIMIT '1' OFFSET '0'

which obviously trigger syntax errors.

[SevInf]

Note: planetscale (wasm) is failing for reasons seemingly independent of this PR.

I understood the problem, for which I don't have yet a solution: on PlanetScale, bigint values are placed into queries as strings, resulting in SQL snippets like

LIMIT '1' OFFSET '0'

which obviously trigger syntax errors.

Interesting. Should we report it to planetscale maybe?

[SevInf]

LGTM when tests are green.
Maybe, skip Plantescale tests back and open follow-up issue?

[jkomyno]

LGTM when tests are green. Maybe, skip Plantescale tests back and open follow-up issue?

Yeah, I'm thinking the same. I've created https://github.com/prisma/team-orm/issues/836, and excluded PlanetScale. I'll let the CI run, and merge afterwards if it's all good

[jkomyno]

LGTM when tests are green. Maybe, skip Plantescale tests back and open follow-up issue?

Yeah, I'm thinking the same. I've created prisma/team-orm#836, and excluded PlanetScale. I'll let the CI run, and merge afterwards if it's all good

PlanetScale still failing, I'll keep exploring why.

[jkomyno]

I've opened a PR on PlanetScale to fix this: planetscale/database-js#159.