Skip to content
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.

Sign up for GitHub

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

Jump to bottom

Allow narrowing {} originating in unknown using instanceof #51014

Closed
wants to merge 3 commits into from
Closed

Allow narrowing {} originating in unknown using instanceof #51014

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
20826ca
Allow narrowing `{}` originating in `unknown` using `instanceof`
Andarist Sep 30, 2022
2b0f6ff
Merge remote-tracking branch 'origin/main' into fix/51007
Andarist Nov 11, 2022
89872e6
Add extra test case
Andarist Nov 11, 2022
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Jump to
Jump to file
Failed to load files.
Diff view
Diff view
4 changes: 3 additions & 1 deletion src/compiler/checker.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -25721,7 +25721,9 @@ namespace ts {
const nonConstructorTypeInUnion = find((rightType as UnionType).types, (t) => !isConstructorType(t));
if (!nonConstructorTypeInUnion) return type;
}

if (assumeTrue && declaredType === unknownType && isEmptyAnonymousObjectType(type)) {
Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Perhaps this condition should be relaxed somehow or maybe it should be pushed further down the road to getNarrowedType. This is the version that I'm most confident with though ;p

The added check is roughly based on the check recently introduced here: https://github.com/microsoft/TypeScript/pull/50610/files#diff-d9ab6589e714c71e657f601cf30ff51dfc607fc98419bf72e04f6b0fa92cc4b8R25153

return targetType;
}
return getNarrowedType(type, targetType, assumeTrue, /*checkDerived*/ true);
}

Expand Down
6 changes: 6 additions & 0 deletions tests/baselines/reference/inKeywordAndUnknown.errors.txt
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -20,4 +20,10 @@ tests/cases/compiler/inKeywordAndUnknown.ts(12,18): error TS2638: Type '{}' may
}
y; // {}
}


// repro #51007
function isHTMLTable(table: unknown): boolean {
return !!table && table instanceof Object && 'html' in table;
}

10 changes: 10 additions & 0 deletions tests/baselines/reference/inKeywordAndUnknown.js
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -15,6 +15,12 @@ function f(x: {}, y: unknown) {
}
y; // {}
}


// repro #51007
function isHTMLTable(table: unknown): boolean {
return !!table && table instanceof Object && 'html' in table;
}


//// [inKeywordAndUnknown.js]
Expand All @@ -34,3 +40,7 @@ function f(x, y) {
}
y; // {}
}
// repro #51007
function isHTMLTable(table) {
return !!table && table instanceof Object && 'html' in table;
}
13 changes: 13 additions & 0 deletions tests/baselines/reference/inKeywordAndUnknown.symbols
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -31,3 +31,16 @@ function f(x: {}, y: unknown) {
>y : Symbol(y, Decl(inKeywordAndUnknown.ts, 2, 17))
}


// repro #51007
function isHTMLTable(table: unknown): boolean {
>isHTMLTable : Symbol(isHTMLTable, Decl(inKeywordAndUnknown.ts, 15, 1))
>table : Symbol(table, Decl(inKeywordAndUnknown.ts, 19, 21))

return !!table && table instanceof Object && 'html' in table;
>table : Symbol(table, Decl(inKeywordAndUnknown.ts, 19, 21))
>table : Symbol(table, Decl(inKeywordAndUnknown.ts, 19, 21))
>Object : Symbol(Object, Decl(lib.es5.d.ts, --, --), Decl(lib.es5.d.ts, --, --))
>table : Symbol(table, Decl(inKeywordAndUnknown.ts, 19, 21))
}

20 changes: 20 additions & 0 deletions tests/baselines/reference/inKeywordAndUnknown.types
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -40,3 +40,23 @@ function f(x: {}, y: unknown) {
>y : Record<"a", unknown>
}


// repro #51007
function isHTMLTable(table: unknown): boolean {
>isHTMLTable : (table: unknown) => boolean
>table : unknown

return !!table && table instanceof Object && 'html' in table;
>!!table && table instanceof Object && 'html' in table : boolean
>!!table && table instanceof Object : boolean
>!!table : boolean
>!table : boolean
>table : unknown
>table instanceof Object : boolean
>table : {}
>Object : ObjectConstructor
>'html' in table : boolean
>'html' : "html"
>table : Object
}

6 changes: 6 additions & 0 deletions tests/cases/compiler/inKeywordAndUnknown.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -16,3 +16,9 @@ function f(x: {}, y: unknown) {
}
y; // {}
}


// repro #51007
function isHTMLTable(table: unknown): boolean {
return !!table && table instanceof Object && 'html' in table;
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I think the test here should actually test for an access of table.html in some capacity - e.g.

function tryGetHtmlPropFromTable(table: unknown) {
    if (!!table && table instanceof Object && 'html' in table && table.html instanceof Object) {
        return table.html;
    }
    return undefined;
}

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

hm, what would it change here? the reported error was that in couldn't be used at all in this situation - we could add more "assertions" here than what was originally reported but I'm not sure what this particular snippet would bring to the table here. What kind of nuance am I missing? 😅

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

The original test is fine; having another test that actually tries to narrow and use the newly-tested property corresponds to something closer to real-world usage.

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Ok, I've pushed out the proposed test case - please take a look if this is what you had in mind here.

}