Skip to content

Pull off full code points in template literal inference #63537

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

Draft
jakebailey wants to merge 1 commit into
base: main
Choose a base branch
from
+182 −1
Draft

Pull off full code points in template literal inference #63537

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
5 changes: 4 additions & 1 deletion src/compiler/checker.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -26744,7 +26744,10 @@ export function createTypeChecker(host: TypeCheckerHost): TypeChecker {
pos += delim.length;
}
else if (pos < getSourceText(seg).length) {
addMatch(seg, pos + 1);
// Consume a whole code point (rune) so that surrogate pairs (e.g. emoji)
// are not split in half. A lone surrogate consumes a single code unit.
const ch = getSourceText(seg).codePointAt(pos)!;
addMatch(seg, pos + (ch >= 0x10000 ? 2 : 1));
}
else if (seg < lastSourceIndex) {
addMatch(seg + 1, 0);
Expand Down
76 changes: 76 additions & 0 deletions tests/baselines/reference/templateLiteralTypesInferenceRunes.symbols
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,76 @@
//// [tests/cases/compiler/templateLiteralTypesInferenceRunes.ts] ////

=== templateLiteralTypesInferenceRunes.ts ===
// Template literal type inference should pull off whole code points (runes)
// rather than splitting surrogate pairs (e.g. emoji) into lone surrogates.
// https://github.com/microsoft/TypeScript/issues/63533

type Heads<S> = S extends `${infer C}${infer R}` ? [C, R] : never;
>Heads : Symbol(Heads, Decl(templateLiteralTypesInferenceRunes.ts, 0, 0))
>S : Symbol(S, Decl(templateLiteralTypesInferenceRunes.ts, 4, 11))
>S : Symbol(S, Decl(templateLiteralTypesInferenceRunes.ts, 4, 11))
>C : Symbol(C, Decl(templateLiteralTypesInferenceRunes.ts, 4, 34))
>R : Symbol(R, Decl(templateLiteralTypesInferenceRunes.ts, 4, 44))
>C : Symbol(C, Decl(templateLiteralTypesInferenceRunes.ts, 4, 34))
>R : Symbol(R, Decl(templateLiteralTypesInferenceRunes.ts, 4, 44))

type A = Heads<"😀abc">;
>A : Symbol(A, Decl(templateLiteralTypesInferenceRunes.ts, 4, 66))
>Heads : Symbol(Heads, Decl(templateLiteralTypesInferenceRunes.ts, 0, 0))

declare let a: A;
>a : Symbol(a, Decl(templateLiteralTypesInferenceRunes.ts, 7, 11))
>A : Symbol(A, Decl(templateLiteralTypesInferenceRunes.ts, 4, 66))

const chk: ["😀", "abc"] = a;
>chk : Symbol(chk, Decl(templateLiteralTypesInferenceRunes.ts, 8, 5))
>a : Symbol(a, Decl(templateLiteralTypesInferenceRunes.ts, 7, 11))

// Multiple surrogate-pair characters in a row
type B = Heads<"😀😁">;
>B : Symbol(B, Decl(templateLiteralTypesInferenceRunes.ts, 8, 29))
>Heads : Symbol(Heads, Decl(templateLiteralTypesInferenceRunes.ts, 0, 0))

declare let b: B;
>b : Symbol(b, Decl(templateLiteralTypesInferenceRunes.ts, 12, 11))
>B : Symbol(B, Decl(templateLiteralTypesInferenceRunes.ts, 8, 29))

const chk2: ["😀", "😁"] = b;
>chk2 : Symbol(chk2, Decl(templateLiteralTypesInferenceRunes.ts, 13, 5))
>b : Symbol(b, Decl(templateLiteralTypesInferenceRunes.ts, 12, 11))

// Surrogate pair followed by a BMP character
type Pair<S> = S extends `${infer C1}${infer C2}` ? [C1, C2] : never;
>Pair : Symbol(Pair, Decl(templateLiteralTypesInferenceRunes.ts, 13, 29))
>S : Symbol(S, Decl(templateLiteralTypesInferenceRunes.ts, 16, 10))
>S : Symbol(S, Decl(templateLiteralTypesInferenceRunes.ts, 16, 10))
>C1 : Symbol(C1, Decl(templateLiteralTypesInferenceRunes.ts, 16, 33))
>C2 : Symbol(C2, Decl(templateLiteralTypesInferenceRunes.ts, 16, 44))
>C1 : Symbol(C1, Decl(templateLiteralTypesInferenceRunes.ts, 16, 33))
>C2 : Symbol(C2, Decl(templateLiteralTypesInferenceRunes.ts, 16, 44))

type C = Pair<"😀x">;
>C : Symbol(C, Decl(templateLiteralTypesInferenceRunes.ts, 16, 69))
>Pair : Symbol(Pair, Decl(templateLiteralTypesInferenceRunes.ts, 13, 29))

declare let c: C;
>c : Symbol(c, Decl(templateLiteralTypesInferenceRunes.ts, 18, 11))
>C : Symbol(C, Decl(templateLiteralTypesInferenceRunes.ts, 16, 69))

const chk3: ["😀", "x"] = c;
>chk3 : Symbol(chk3, Decl(templateLiteralTypesInferenceRunes.ts, 19, 5))
>c : Symbol(c, Decl(templateLiteralTypesInferenceRunes.ts, 18, 11))

// A leading BMP character then a surrogate pair
type D = Pair<"x😀">;
>D : Symbol(D, Decl(templateLiteralTypesInferenceRunes.ts, 19, 28))
>Pair : Symbol(Pair, Decl(templateLiteralTypesInferenceRunes.ts, 13, 29))

declare let d: D;
>d : Symbol(d, Decl(templateLiteralTypesInferenceRunes.ts, 23, 11))
>D : Symbol(D, Decl(templateLiteralTypesInferenceRunes.ts, 19, 28))

const chk4: ["x", "😀"] = d;
>chk4 : Symbol(chk4, Decl(templateLiteralTypesInferenceRunes.ts, 24, 5))
>d : Symbol(d, Decl(templateLiteralTypesInferenceRunes.ts, 23, 11))

74 changes: 74 additions & 0 deletions tests/baselines/reference/templateLiteralTypesInferenceRunes.types
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,74 @@
//// [tests/cases/compiler/templateLiteralTypesInferenceRunes.ts] ////

=== templateLiteralTypesInferenceRunes.ts ===
// Template literal type inference should pull off whole code points (runes)
// rather than splitting surrogate pairs (e.g. emoji) into lone surrogates.
// https://github.com/microsoft/TypeScript/issues/63533

type Heads<S> = S extends `${infer C}${infer R}` ? [C, R] : never;
>Heads : Heads<S>
> : ^^^^^^^^

type A = Heads<"😀abc">;
>A : ["😀", "abc"]
> : ^^^^^^^^^^^^^

declare let a: A;
>a : ["😀", "abc"]
> : ^^^^^^^^^^^^^

const chk: ["😀", "abc"] = a;
>chk : ["😀", "abc"]
> : ^^^^^^^^^^^^^
>a : ["😀", "abc"]
> : ^^^^^^^^^^^^^

// Multiple surrogate-pair characters in a row
type B = Heads<"😀😁">;
>B : ["😀", "😁"]
> : ^^^^^^^^^^^^

declare let b: B;
>b : ["😀", "😁"]
> : ^^^^^^^^^^^^

const chk2: ["😀", "😁"] = b;
>chk2 : ["😀", "😁"]
> : ^^^^^^^^^^^^
>b : ["😀", "😁"]
> : ^^^^^^^^^^^^

// Surrogate pair followed by a BMP character
type Pair<S> = S extends `${infer C1}${infer C2}` ? [C1, C2] : never;
>Pair : Pair<S>
> : ^^^^^^^

type C = Pair<"😀x">;
>C : ["😀", "x"]
> : ^^^^^^^^^^^

declare let c: C;
>c : ["😀", "x"]
> : ^^^^^^^^^^^

const chk3: ["😀", "x"] = c;
>chk3 : ["😀", "x"]
> : ^^^^^^^^^^^
>c : ["😀", "x"]
> : ^^^^^^^^^^^

// A leading BMP character then a surrogate pair
type D = Pair<"x😀">;
>D : ["x", "😀"]
> : ^^^^^^^^^^^

declare let d: D;
>d : ["x", "😀"]
> : ^^^^^^^^^^^

const chk4: ["x", "😀"] = d;
>chk4 : ["x", "😀"]
> : ^^^^^^^^^^^
>d : ["x", "😀"]
> : ^^^^^^^^^^^

28 changes: 28 additions & 0 deletions tests/cases/compiler/templateLiteralTypesInferenceRunes.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,28 @@
// @strict: true
// @noEmit: true

// Template literal type inference should pull off whole code points (runes)
// rather than splitting surrogate pairs (e.g. emoji) into lone surrogates.
// https://github.com/microsoft/TypeScript/issues/63533

type Heads<S> = S extends `${infer C}${infer R}` ? [C, R] : never;

type A = Heads<"😀abc">;
declare let a: A;
const chk: ["😀", "abc"] = a;

// Multiple surrogate-pair characters in a row
type B = Heads<"😀😁">;
declare let b: B;
const chk2: ["😀", "😁"] = b;

// Surrogate pair followed by a BMP character
type Pair<S> = S extends `${infer C1}${infer C2}` ? [C1, C2] : never;
type C = Pair<"😀x">;
declare let c: C;
const chk3: ["😀", "x"] = c;

// A leading BMP character then a surrogate pair
type D = Pair<"x😀">;
declare let d: D;
const chk4: ["x", "😀"] = d;
Loading