-
Notifications
You must be signed in to change notification settings - Fork 4
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Add Babel scenario #17
base: main
Are you sure you want to change the base?
Conversation
Thanks for the PR; the repo is in flux and benchmarking resources are limited, so I'm not sure I'm going to take this quite yet. Is there something special about Babel that warrants its addition besides it just being a popular library? All of the current benchmarks are "special" in that they test codebases that are slow or often become slow for one reason or another. It's likely that the list from microsoft/TypeScript#44033 will come first. |
Because there are all JS/TS/FLOW AST type definitions in So personally it seems to me that Babel might be a suitable typical example to measure this kind of usage. :) Of course, if you think Babel is not suitable for adding to this repository, I completely understand. The project itself is amazing, and it’s great to see it coming! |
I've done a bunch of refactoring and so this PR is no longer up to date with how to add a benchmark; a better example is #7, if you wanted to update this PR. |
a51f28f
to
ae3e498
Compare
I didn't add it to |
It shouldn't, other than picking an agent for the baseline to run on long term, and knowing roughly how long it takes to do a single compile (which I could measure for you). |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Until we figure out how to safely install the repo, I unfortunately cannot merge this.
Additionally, the scripts need to be chmod -x
'd.
FWIW I'm waiting on this one until the babel repo's layout is finalized. There's also a concern that the benchmark takes too long if the new project references hurt so much, that and the fact that build mode means that building projects may cause things to "fail fast" and cause the benchmark to stop working if TS ever adds new errors to babel's TS code. |
The speed problem may be solved by cc @nicolo-ribaudo You may have better ideas! |
Thank you very much for your work on the benchmark!
Babel seems to be a good scenario, a lot of typescript code, and widely used.
BTW: Babel has the
incremental
option enabled, I'm not sure if I need to do anything extra.