feat(transform-runtime): add more type hints for instance methods #10068
Conversation
|
Build successful! You can test your changes in the REPL here: https://babeljs.io/repl/build/10920/ |
|
Build successful! You can test your changes in the REPL here: https://babeljs.io/repl/build/10910/ |
devinus
force-pushed
the
feature/more-instance-method-hints
branch
from
f21a95c
to
f0db990
Compare
packages/babel-plugin-transform-runtime/src/runtime-corejs3-definitions.js
Outdated
Show resolved
Hide resolved
packages/babel-plugin-transform-runtime/src/runtime-corejs3-definitions.js
Show resolved
Hide resolved
packages/babel-plugin-transform-runtime/src/runtime-corejs3-definitions.js
Outdated
Show resolved
Hide resolved
devinus
force-pushed
the
feature/more-instance-method-hints
branch
from
f0db990
to
7764056
Compare
devinus
force-pushed
the
feature/more-instance-method-hints
branch
from
7764056
to
922ce97
Compare
| flat: { stable: true, path: "flat", types: ["array"] }, | ||
| forEach: { stable: true, path: "for-each", types: ["array"] }, | ||
| includes: { stable: true, path: "includes", types: ["array", "string"] }, | ||
| indexOf: { stable: true, path: "index-of", types: ["array"] }, |
There was a problem hiding this comment.
❗ I think that this is the only actually needed line in this PR, since we don't need to load the polyfill for "".indexOf.
All the other additions only prevent us from adding polyfills for code that will throw at runtime.
There was a problem hiding this comment.
Doesn't it also prevent polyfilling for literals that may have had their prototypes monkey-patched? Which is really dumb but also possible.
There was a problem hiding this comment.
Seems, for better effect, we should extend typeAnnotationToString for work with more classes / types.
| flags: { stable: true, path: "flags", types: ["regexp"] }, | ||
| flatMap: { stable: true, path: "flat-map", types: ["array"] }, | ||
| flat: { stable: true, path: "flat", types: ["array"] }, | ||
| forEach: { stable: true, path: "for-each", types: ["array"] }, |
There was a problem hiding this comment.
This is also for map/set, but I think that Babel handles them as "unknown" and thus adds the polyfill anyway?
There was a problem hiding this comment.
I think that Babel handles them as "unknown" and thus adds the polyfill anyway?
Correct.
There was a problem hiding this comment.
Could you add a test? I don't want to introduce bugs in case we decide to optimize the logic also for map/set other than array/etc.
There was a problem hiding this comment.
The problem with Map and Set that even if we will add naive detection of them to typeAnnotationToString to prevent adding unnecessary .forEach polyfill, Map anyway could be transpiled to something like _Map by runtime transform.
packages/babel-plugin-transform-runtime/src/runtime-corejs3-definitions.js
Show resolved
Hide resolved
Continues the work from #9754 and adds more hints to the corejs3 definition type hints.