Stricter relational comparisons checking valueOf
#52807
Closed
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
typescript-bot
added
Author: Team
For Uncommitted Bug
|
In the self test:
function compareComparableValues(a: string | undefined, b: string | undefined): Comparison;
function compareComparableValues(a: number | undefined, b: number | undefined): Comparison;
function compareComparableValues(a: string | number | undefined, b: string | number | undefined) {
return a === b ? Comparison.EqualTo :
a === undefined ? Comparison.LessThan :
b === undefined ? Comparison.GreaterThan :
a < b ? Comparison.LessThan :
Comparison.GreaterThan;
}Seems like a great catch to me! The entire thing could/should just be: function compareComparableValues<T extends string | number>(a: T | undefined, b: T | undefined) {
// ....
}EDIT: this is wrong, see below |
|
@typescript-bot test this |
|
Heya @RyanCavanaugh, I've started to run the extended test suite on this PR at 08a34e1. You can monitor the build here. |
|
Heya @RyanCavanaugh, I've started to run the diff-based top-repos suite on this PR at 08a34e1. You can monitor the build here. Update: The results are in! |
|
Heya @RyanCavanaugh, I've started to run the parallelized Definitely Typed test suite on this PR at 08a34e1. You can monitor the build here. |
|
Heya @RyanCavanaugh, I've started to run the diff-based user code test suite (tsserver) on this PR at 08a34e1. You can monitor the build here. Update: The results are in! |
|
Heya @RyanCavanaugh, I've started to run the perf test suite on this PR at 08a34e1. You can monitor the build here. Update: The results are in! |
|
Heya @RyanCavanaugh, I've started to run the diff-based user code test suite on this PR at 08a34e1. You can monitor the build here. Update: The results are in! |
|
The case of function compare<T extends string | number>(a: T, b: T) {
return a > b;
}is pretty tough actually, since by our own logic this should be illegal (and is under this PR) but people will probably feel that it ought to be legal. |
|
@RyanCavanaugh Here are the results of running the user test suite comparing Something interesting changed - please have a look. Detailspuppeteer
|
|
@RyanCavanaugh Here are the results of running the user test suite comparing Something interesting changed - please have a look. Details
|
|
Heya @RyanCavanaugh, I've run the RWC suite on this PR - assuming you're on the TS core team, you can view the resulting diff here. |
|
RWC diffs are mostly what you'd expect from this:
|
|
@RyanCavanaugh Here they are:
CompilerComparison Report - main..52807
System
Hosts
Scenarios
TSServerComparison Report - main..52807
System
Hosts
Scenarios
StartupComparison Report - main..52807
System
Hosts
Scenarios
Developer Information: |
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
@RyanCavanaugh Here are the results of running the top-repos suite comparing Something interesting changed - please have a look. Details
|
|
Those vscode bugs look real, in particular the |
|
@typescript-bot pack this |
|
Heya @jakebailey, I've started to run the tarball bundle task on this PR at 8703e3c. You can monitor the build here. |
|
@RyanCavanaugh Here are the results of running the top-repos suite comparing Something interesting changed - please have a look. Detailsangular/angular-cli
|
|
@RyanCavanaugh Here are some more interesting changes from running the top-repos suite Detailsionic-team/ionic-framework
|
|
@RyanCavanaugh Here are some more interesting changes from running the top-repos suite Detailspnpm/pnpm
|
|
@RyanCavanaugh Here are some more interesting changes from running the top-repos suite Detailstrpc/trpc
|
|
Hey @jakebailey, I've packed this into an installable tgz. You can install it for testing by referencing it in your and then running There is also a playground for this build and an npm module you can use via |
|
The other break is regarding this definition export type Length = { _brand: 'Length'; };
if (length1 > length2) {In reality, a export type Length = { _brand: 'Length'; valueOf(): number }; |
|
Would they be better served by |
jakebailey
reviewed
Feb 16, 2023
| function getValueOfResult(type: Type): Type { | ||
| const valueOf = getPropertyOfType(type, "valueOf" as __String); | ||
| if (valueOf) { | ||
| const signatures = getSignaturesOfType(getTypeOfSymbol(valueOf), ts.SignatureKind.Call); |
There was a problem hiding this comment.
FYI annoying auto-import thing here, but should should just be SignatureKind.Call; sometimes the ts prefix appears first in the completion list and gets used (and that's one reason why I'm trying to get us to remove them everywhere).
There was a problem hiding this comment.
Main no longer has this ts import so when you merge from main next, you'll have to fix this (and then never have to worry about again).
|
I watched the design meeting back, and I didn't realize my fix for Given a type variable
If we're going to disallow comparison of these, then I think the error needs to be improved to say why, because without knowing I could have |
|
Kind of funny but this allows |
RyanCavanaugh
commented
Feb 24, 2023
| function getValueOfResult(type: Type): Type { | ||
| const valueOf = getPropertyOfType(type, "valueOf" as __String); | ||
| if (valueOf) { | ||
| const signatures = getSignaturesOfType(getTypeOfSymbol(valueOf), ts.SignatureKind.Call); |
There was a problem hiding this comment.
Suggested change
| const signatures = getSignaturesOfType(getTypeOfSymbol(valueOf), ts.SignatureKind.Call); | |
| const signatures = getSignaturesOfType(getTypeOfSymbol(valueOf), SignatureKind.Call); | |
|
We should include |
weswigham
reviewed
Feb 24, 2023
| if (valueOf) { | ||
| const signatures = getSignaturesOfType(getTypeOfSymbol(valueOf), ts.SignatureKind.Call); | ||
| if (signatures && signatures.length > 0) { | ||
| const returnType = getReturnTypeOfSignature(signatures[0]); |
There was a problem hiding this comment.
This favors the first signature in an overload list, while usually in places where we're overload-blind today we usually favor the last overload in the list. Maybe signatures[signatures.length - 1] for consistency? Or, potentially better yet, rather than picking one overload, getUnionType(map(signatures, getReturnTypeOfSignature)) just to hedge (in theory, only overloads that match an all-undefined/empty argument list would actually apply.... but overloads on valueOf are probably a bit suspicious anyway). I say that, but we'll actually see an overloaded valueOf for every intersected type eg, number & {myBrand} is going to have a valueOf from number's apparent type and object's apparent type intersected together, which'll result in two overloads unless they're merged for being identical.
I think this is basically how everyone intuitively thinks about Footnotes
|
|
This is mostly being addressed by some other PRs |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Per discussion in #52790, this implements the following ruleset for
>,<,>=, and<=:For each operand expressions, let the "operand type" be:
valueOfproperty of the expression's type, if that existsOne of the following must be true:
Otherwise an error is issued