Fixed inferred const rest types in signatures

[Mateusz Burzyński (Andarist)]

fixes#55033

[David Blass (ssalbdivad)]

Pretty hyped for this one. I've run into so many issues trying to constrain narrowed array types without breaking inference. Couldn't really tell from the linked issue if this will fix a lot of them, but I have faith 🙏

[Jake Bailey (jakebailey)]

After messing with this same code for #53255, #53258 (note a similar getRestTypeAtPosition change), #53398, I'm like 50% afraid of this 😄

TypeScript Bot (@typescript-bot) test this
TypeScript Bot (@typescript-bot) test top100
TypeScript Bot (@typescript-bot) user test this
TypeScript Bot (@typescript-bot) run dt
TypeScript Bot (@typescript-bot) perf test this faster
TypeScript Bot (@typescript-bot) pack this

[TypeScript Bot (typescript-bot)]

Heya Jake Bailey (@jakebailey), I've started to run the diff-based top-repos suite on this PR at fb32b83. You can monitor the build here.

Update: The results are in!

[TypeScript Bot (typescript-bot)]

Heya Jake Bailey (@jakebailey), I've started to run the diff-based user code test suite on this PR at fb32b83. You can monitor the build here.

Update: The results are in!

[TypeScript Bot (typescript-bot)]

Heya Jake Bailey (@jakebailey), I've started to run the abridged perf test suite on this PR at fb32b83. You can monitor the build here.

Update: The results are in!

[TypeScript Bot (typescript-bot)]

Heya Jake Bailey (@jakebailey), I've started to run the tarball bundle task on this PR at fb32b83. You can monitor the build here.

[TypeScript Bot (typescript-bot)]

Heya Jake Bailey (@jakebailey), I've started to run the extended test suite on this PR at fb32b83. You can monitor the build here.

[TypeScript Bot (typescript-bot)]

Heya Jake Bailey (@jakebailey), I've started to run the parallelized Definitely Typed test suite on this PR at fb32b83. You can monitor the build here.

Update: The results are in!

[TypeScript Bot (typescript-bot)]

Hey Jake Bailey (@jakebailey), I've packed this into an installable tgz. You can install it for testing by referencing it in your package.json like so:

{
"devDependencies": {
"typescript": "https://typescript.visualstudio.com/cf7ac146-d525-443c-b23c-0d58337efebc/_apis/build/builds/155902/artifacts?artifactName=tgz&fileId=FAE8E38594F6A90F7E599497909CC324A2C4C0E435E4CFA4BD738E7304C0DB8F02&fileName=/typescript-5.2.0-insiders.20230718.tgz"
}
}

and then running npm install.

---

There is also a playground for this build and an npm module you can use via "typescript": "npm:@typescript-deploys/pr-build@5.2.0-pr-55034-8".;

[Mateusz Burzyński (Andarist)]

After messing with this same code for #53255, #53258 (note a similar getRestTypeAtPosition change), #53398, I'm like 50% afraid of this 😄

When learning about this issue I immediately opened your PR to check out if it's related anyhow to this issue. I think that at the end of the day, it isn't and that this PR shouldn't actually impact anybody negatively. This only makes those rests readonly when the type variable is const - given that const is pretty new I don't expect any accidental breakage here.

[Jake Bailey (jakebailey)]

After messing with this same code for #53255, #53258 (note a similar getRestTypeAtPosition change), #53398, I'm like 50% afraid of this 😄

When learning about this issue I immediately opened your PR to check out if it's related anyhow to this issue. I think that at the end of the day, it isn't and that this PR shouldn't actually impact anybody negatively. This only makes those rests readonly when the type variable is const - given that const is pretty new I don't expect any accidental breakage here.

Yeah, I'm pretty sure that I was a little overzealous in propagating readonly to the last use (this PR), where my issue was definitely intended to change the other use.

[TypeScript Bot (typescript-bot)]

Jake Bailey (@jakebailey)
The results of the perf run you requested are in!

Here they are:

Comparison Report - main..55034

Metric	main	55034	Delta	Best	Worst	p-value
Angular - node (v16.17.1, x64)
Memory used	366,223k (± 0.01%)	366,214k (± 0.00%)	~	366,200k	366,225k	p=0.423 n=6
Parse Time	3.56s (± 0.65%)	3.58s (± 0.63%)	~	3.56s	3.62s	p=0.168 n=6
Bind Time	1.19s (± 0.63%)	1.20s (± 0.46%)	~	1.19s	1.20s	p=0.137 n=6
Check Time	9.71s (± 0.46%)	9.67s (± 0.42%)	~	9.62s	9.73s	p=0.146 n=6
Emit Time	8.09s (± 0.46%)	8.02s (± 0.77%)	~	7.96s	8.12s	p=0.065 n=6
Total Time	22.54s (± 0.28%)	22.46s (± 0.43%)	~	22.37s	22.63s	p=0.126 n=6
Compiler-Unions - node (v16.17.1, x64)
Memory used	193,442k (± 0.68%)	193,464k (± 0.68%)	~	192,849k	196,148k	p=0.471 n=6
Parse Time	1.59s (± 0.51%)	1.59s (± 0.62%)	~	1.57s	1.60s	p=0.383 n=6
Bind Time	0.83s (± 0.66%)	0.82s (± 0.92%)	~	0.81s	0.83s	p=0.137 n=6
Check Time	10.15s (± 0.76%)	10.18s (± 0.56%)	~	10.11s	10.28s	p=0.378 n=6
Emit Time	3.00s (± 0.53%)	3.01s (± 1.35%)	~	2.98s	3.09s	p=1.000 n=6
Total Time	15.56s (± 0.54%)	15.60s (± 0.54%)	~	15.51s	15.76s	p=0.378 n=6
Monaco - node (v16.17.1, x64)
Memory used	346,183k (± 0.01%)	346,189k (± 0.00%)	~	346,176k	346,200k	p=0.689 n=6
Parse Time	2.77s (± 0.49%)	2.78s (± 0.77%)	~	2.76s	2.82s	p=0.742 n=6
Bind Time	1.07s (± 0.38%)	1.08s (± 0.59%)	+0.01s (+ 1.09%)	1.07s	1.09s	p=0.009 n=6
Check Time	8.01s (± 0.67%)	8.03s (± 0.45%)	~	7.98s	8.07s	p=0.687 n=6
Emit Time	4.49s (± 0.60%)	4.47s (± 0.86%)	~	4.44s	4.54s	p=0.332 n=6
Total Time	16.34s (± 0.50%)	16.35s (± 0.41%)	~	16.27s	16.45s	p=0.872 n=6
TFS - node (v16.17.1, x64)
Memory used	300,246k (± 0.00%)	300,229k (± 0.01%)	~	300,203k	300,252k	p=0.128 n=6
Parse Time	2.21s (± 0.62%)	2.21s (± 0.72%)	~	2.19s	2.23s	p=0.869 n=6
Bind Time	1.20s (± 1.33%)	1.21s (± 0.81%)	~	1.19s	1.22s	p=0.510 n=6
Check Time	7.33s (± 0.45%)	7.34s (± 0.34%)	~	7.30s	7.37s	p=0.618 n=6
Emit Time	4.35s (± 0.52%)	4.33s (± 1.25%)	~	4.26s	4.42s	p=0.145 n=6
Total Time	15.09s (± 0.33%)	15.08s (± 0.37%)	~	15.01s	15.14s	p=0.574 n=6
material-ui - node (v16.17.1, x64)
Memory used	481,657k (± 0.01%)	481,630k (± 0.01%)	~	481,584k	481,695k	p=0.575 n=6
Parse Time	3.25s (± 0.41%)	3.27s (± 0.57%)	+0.02s (+ 0.46%)	3.23s	3.28s	p=0.048 n=6
Bind Time	0.96s (± 0.78%)	0.96s (± 1.26%)	~	0.95s	0.98s	p=0.235 n=6
Check Time	18.45s (± 0.60%)	18.36s (± 0.58%)	~	18.28s	18.57s	p=0.199 n=6
Emit Time	0.00s (± 0.00%)	0.00s (± 0.00%)	~	0.00s	0.00s	p=1.000 n=6
Total Time	22.67s (± 0.46%)	22.58s (± 0.52%)	~	22.47s	22.81s	p=0.199 n=6
xstate - node (v16.17.1, x64)
Memory used	561,367k (± 0.03%)	561,174k (± 0.02%)	-193k (- 0.03%)	561,060k	561,346k	p=0.031 n=6
Parse Time	4.00s (± 0.34%)	4.03s (± 0.43%)	+0.02s (+ 0.54%)	4.01s	4.06s	p=0.042 n=6
Bind Time	1.78s (± 1.05%)	1.79s (± 1.22%)	~	1.77s	1.83s	p=0.505 n=6
Check Time	3.07s (± 0.39%)	3.06s (± 0.68%)	~	3.04s	3.09s	p=0.514 n=6
Emit Time	0.10s (± 5.76%)	0.10s (± 5.76%)	~	0.09s	0.10s	p=1.000 n=6
Total Time	8.94s (± 0.39%)	8.97s (± 0.44%)	~	8.92s	9.03s	p=0.228 n=6

System

Machine Name	ts-ci-ubuntu
Platform	linux 5.4.0-148-generic
Architecture	x64
Available Memory	16 GB
Available Memory	15 GB
CPUs	4 × Intel(R) Core(TM) i7-4770 CPU @ 3.40GHz

Hosts

• node (v16.17.1, x64)

Scenarios

• Angular - node (v16.17.1, x64)
• Compiler-Unions - node (v16.17.1, x64)
• Monaco - node (v16.17.1, x64)
• TFS - node (v16.17.1, x64)
• material-ui - node (v16.17.1, x64)
• xstate - node (v16.17.1, x64)

Benchmark	Name	Iterations
Current	55034	6
Baseline	main	6

Developer Information:

Download Benchmark

[TypeScript Bot (typescript-bot)]

Jake Bailey (@jakebailey) Here are the results of running the user test suite comparing main and refs/pull/55034/merge:

There were infrastructure failures potentially unrelated to your change:

• 1 instance of "Unknown failure"
• 1 instance of "Package install failed"

Otherwise...

Something interesting changed - please have a look.

Details

rxjs-src

/mnt/ts_downloads/rxjs-src/build.sh

• [NEW] error TS2428: All declarations of 'WeakMap' must have identical type parameters.
  • /home/vsts/work/1/s/typescript-55034/lib/lib.es2015.collection.d.ts(63,11)
  • /home/vsts/work/1/s/typescript-55034/lib/lib.es2015.iterable.d.ts(162,11)
  • /home/vsts/work/1/s/typescript-55034/lib/lib.es2015.symbol.wellknown.d.ts(140,11)
  • /home/vsts/work/1/s/typescript-55034/lib/lib.es2015.collection.d.ts(63,11)
  • /home/vsts/work/1/s/typescript-55034/lib/lib.es2015.iterable.d.ts(162,11)
  • /home/vsts/work/1/s/typescript-55034/lib/lib.es2015.symbol.wellknown.d.ts(140,11)
  • /home/vsts/work/1/s/typescript-55034/lib/lib.es2015.collection.d.ts(63,11)
  • /home/vsts/work/1/s/typescript-55034/lib/lib.es2015.iterable.d.ts(162,11)
  • /home/vsts/work/1/s/typescript-55034/lib/lib.es2015.symbol.wellknown.d.ts(140,11)
  • /home/vsts/work/1/s/typescript-55034/lib/lib.es2015.collection.d.ts(63,11)
  • /home/vsts/work/1/s/typescript-55034/lib/lib.es2015.iterable.d.ts(162,11)
  • /home/vsts/work/1/s/typescript-55034/lib/lib.es2015.symbol.wellknown.d.ts(140,11)
  • /home/vsts/work/1/s/typescript-55034/lib/lib.es2015.collection.d.ts(63,11)
  • /home/vsts/work/1/s/typescript-55034/lib/lib.es2015.iterable.d.ts(162,11)
  • /home/vsts/work/1/s/typescript-55034/lib/lib.es2015.symbol.wellknown.d.ts(140,11)
• [MISSING] error TS2428: All declarations of 'WeakMap' must have identical type parameters.
  • /home/vsts/work/1/s/typescript-main/lib/lib.es2015.collection.d.ts(63,11)
  • /home/vsts/work/1/s/typescript-main/lib/lib.es2015.iterable.d.ts(162,11)
  • /home/vsts/work/1/s/typescript-main/lib/lib.es2015.symbol.wellknown.d.ts(140,11)
  • /home/vsts/work/1/s/typescript-main/lib/lib.es2015.collection.d.ts(63,11)
  • /home/vsts/work/1/s/typescript-main/lib/lib.es2015.iterable.d.ts(162,11)
  • /home/vsts/work/1/s/typescript-main/lib/lib.es2015.symbol.wellknown.d.ts(140,11)
  • /home/vsts/work/1/s/typescript-main/lib/lib.es2015.collection.d.ts(63,11)
  • /home/vsts/work/1/s/typescript-main/lib/lib.es2015.iterable.d.ts(162,11)
  • /home/vsts/work/1/s/typescript-main/lib/lib.es2015.symbol.wellknown.d.ts(140,11)
  • /home/vsts/work/1/s/typescript-main/lib/lib.es2015.collection.d.ts(63,11)
  • /home/vsts/work/1/s/typescript-main/lib/lib.es2015.iterable.d.ts(162,11)
  • /home/vsts/work/1/s/typescript-main/lib/lib.es2015.symbol.wellknown.d.ts(140,11)
  • /home/vsts/work/1/s/typescript-main/lib/lib.es2015.collection.d.ts(63,11)
  • /home/vsts/work/1/s/typescript-main/lib/lib.es2015.iterable.d.ts(162,11)
  • /home/vsts/work/1/s/typescript-main/lib/lib.es2015.symbol.wellknown.d.ts(140,11)

[TypeScript Bot (typescript-bot)]

Jake Bailey (@jakebailey) Here are the results of running the top-repos suite comparing main and refs/pull/55034/merge:

Everything looks good!

[TypeScript Bot (typescript-bot)]

Hey Jake Bailey (@jakebailey), the results of running the DT tests are ready.
Everything looks the same!
You can check the log here.

[Nathan Shively-Sanders (sandersn)]

The bug is marked Working As Intended now, so should we close this PR?

[Mateusz Burzyński (Andarist)]

The reported issue had some issues in it (it didnt use a readonly constraint). So im not sure if „working as intended” was applied to only what was oroginally reported or if it was applied to everything that was discussed there. I think that this particular thing that I fixed here still has some merit so i’d appreciate if this would get reviewed

[Anders Hejlsberg (ahejlsberg)]

I like where this is going, but I think we also need the changes in #55229. For example, the following is broken by this PR without the additional changes:

declarefunctioninvoke<constTextendsunknown[]>(f: (...args: T)=>void, ...args: T): T;invoke((a: string,b: number)=>{},"hello",123);

I know this isn't a particularly likely break, but the inference failure because of non-readonly constraints is just confounding to most users and hard to justify (even if I've tried).

[Mateusz Burzyński (Andarist)]

I know this isn't a particularly likely break, but the inference failure because of non-readonly constraints is just confounding to most users and hard to justify (even if I've tried).

ye, this tends to be pretty confusing - would it be worth adding some diagnostics that would prompt users to add the reaodnly modifier to their constraints? Might not be feasible when constraints are defined using aliases but catching common cases when they are declared directly in the signature could still help a lot

[Anders Hejlsberg (ahejlsberg)]

would it be worth adding some diagnostics that would prompt users to add the reaodnly modifier to their constraints?

I'm hoping there's no need for that with #55229 since inference now adjusts itself based on constraints.

[Mateusz Burzyński (Andarist)]

Closing as those commits are now included in #55229