The `parser_napi` pseudo-benchmark does not depend on any Oxc crates, so
remove all dependencies for this benchmark. This reduces build time of
this benchmark on CI by ~1 minute.
After #2780, NAPI parser benchmark sometimes took the longest of all the
benchmarks, so making it faster reduces time to complete the benchmarks
overall.
Export `SourcemapVisualizer` from codegen, it will be used oxc and
rolldown sourcemap test, so it support multiply source print, it will
using sourcemap `sourcesContent` as original source.
Add NodeJS parser to benchmarks.
Previous attempt #2724 did not work due CodSpeed producing very
inaccurate results (https://github.com/CodSpeedHQ/action/issues/96).
This version runs the actual benchmarks without CodSpeed's
instrumentation. Then another faux-benchmark runs within Codspeed's
instrumented action and just performs meaningless calculations in a loop
for as long as is required to take same amount of time as the original
uninstrumented benchmarks took.
It's unfortunate that we therefore don't get flame graphs on CodSpeed,
but this seems to be the best we can do for now.
#2751 contained a mistake, which was pointed out by Adrian @ CodSpeed on
Discord.
For PRs from forks, `CODSPEED_TOKEN` is not provided, and the submission
to CodSpeed is "tokenless". #2751 wrongly assumed all runs are submitted
with a token. This PR fixes that.
Follow-on from #2751. Further shards linter benchmarks so each fixture runs in its own job.
This reduces total time to run benchmarks by another ~75 secs. So approx 2.5 mins shaved off in total.
This PR shards benchmarks when running on CI. Each benchmark (parser, minifier etc) runs as a separate job, and then a final job combines the results and uploads to Codspeed.
A bit of a hacky implementation. Uses a small NodeJS HTTP server to intercept the results from `codspeed-runner`, and then another NodeJS script to combine them all together, and upload to CodSpeed.
I will submit PRs on Codspeed's runner + action to do it properly, but as I imagine it'll be a slow process getting that merged upstream, I wanted to see if it worked first. We can replace this once it's supported upstream.
Sharding only reduces total time to run the benchmarks by about 70 secs at present, because linter benchmark takes 6 mins alone and holds up the whole process (all the rest are done in ~2 mins). If we can split up the linter benchmark, we can likely get total run time down to around 3 mins. I'll try that in a follow-on PR.
I guess the other upside is we can now add as many benchmarks as we like with impunity - they'll run in parallel, and so won't slow things down overall.
This PR merges the previous confusing features `serde` and `wasm` into a
single `serialize` feature.
We'll eventually do serialize + type information for both wasm and napi
targets.
`oxc_macros` is removed from `oxc_ast`'s dependency because it requires
`syn` and friends, which goes against our policy ["Third-party
dependencies should be
minimal."](https://oxc-project.github.io/docs/contribute/rules.html#development-policy)
