Skip to content

Commit 77235de

Browse files
graphemeclustergraphemecluster
committed
Apply Suggested Changes
1 parent 11ad5f5 commit 77235de

File tree

5 files changed

+73
-74
lines changed

5 files changed

+73
-74
lines changed

‎src/compiler/checker.ts‎

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -31885,7 +31885,7 @@ export function createTypeChecker(host: TypeCheckerHost): TypeChecker {
3188531885
scanner.setScriptTarget(sourceFile.languageVersion);
3188631886
scanner.setLanguageVariant(sourceFile.languageVariant);
3188731887
scanner.setOnError((message, length, arg0) => {
31888-
// emulate `parseErrorAtPosition` from parser.ts
31888+
// For providing spelling suggestions
3188931889
const start = scanner!.getTokenEnd();
3189031890
if (message.category === DiagnosticCategory.Message && lastError && start === lastError.start && length === lastError.length) {
3189131891
const error = createDetachedDiagnostic(sourceFile.fileName, sourceFile.text, start, length, message, arg0);

‎src/compiler/scanner.ts‎

Lines changed: 39 additions & 40 deletions
Original file line numberDiff line numberDiff line change
@@ -21,7 +21,6 @@ import {
2121
KeywordSyntaxKind,
2222
LanguageFeatureMinimumTarget,
2323
LanguageVariant,
24-
lastOrUndefined,
2524
LineAndCharacter,
2625
MapLike,
2726
parsePseudoBigInt,
@@ -1614,7 +1613,7 @@ export function createScanner(languageVersion: ScriptTarget, skipTrivia: boolean
16141613
isRegularExpression && shouldEmitInvalidEscapeError && escapedValue >= 0xD800 && escapedValue <= 0xDBFF &&
16151614
pos + 6 < end && text.substring(pos, pos + 2) === "\\u" && charCodeUnchecked(pos + 2) !== CharacterCodes.openBrace
16161615
) {
1617-
// For regular expressions in Unicode mode, \u HexLeadSurrogate \u HexTrailSurrogate is treated as a single character
1616+
// For regular expressions in any Unicode mode, \u HexLeadSurrogate \u HexTrailSurrogate is treated as a single character
16181617
// for the purpose of determining whether a character class range is out of order
16191618
// https://tc39.es/ecma262/#prod-RegExpUnicodeEscapeSequence
16201619
const nextStart = pos;
@@ -2429,7 +2428,7 @@ export function createScanner(languageVersion: ScriptTarget, skipTrivia: boolean
24292428
pos = startOfRegExpBody;
24302429
let inEscape = false;
24312430
// Although nested character classes are allowed in Unicode Sets mode,
2432-
// an unescaped slash is nevertheless invalid even in a character class in Unicode mode.
2431+
// an unescaped slash is nevertheless invalid even in a character class in any Unicode mode.
24332432
// Additionally, parsing nested character classes will misinterpret regexes like `/[[]/`
24342433
// as unterminated, consuming characters beyond the slash. (This even applies to `/[[]/v`,
24352434
// which should be parsed as a well-terminated regex with an incomplete character class.)
@@ -2438,13 +2437,8 @@ export function createScanner(languageVersion: ScriptTarget, skipTrivia: boolean
24382437
while (true) {
24392438
// If we reach the end of a file, or hit a newline, then this is an unterminated
24402439
// regex. Report error and return what we have so far.
2441-
if (pos >= end) {
2442-
tokenFlags |= TokenFlags.Unterminated;
2443-
break;
2444-
}
2445-
2446-
const ch = charCodeUnchecked(pos);
2447-
if (isLineBreak(ch)) {
2440+
const ch = charCodeChecked(pos);
2441+
if (ch === CharacterCodes.EOF || isLineBreak(ch)) {
24482442
tokenFlags |= TokenFlags.Unterminated;
24492443
break;
24502444
}
@@ -2477,7 +2471,8 @@ export function createScanner(languageVersion: ScriptTarget, skipTrivia: boolean
24772471
pos = startOfRegExpBody;
24782472
inEscape = false;
24792473
let characterClassDepth = 0;
2480-
const bracketStack: CharacterCodes[] = [];
2474+
let inDecimalQuantifier = false;
2475+
let groupDepth = 0;
24812476
while (pos < endOfRegExpBody) {
24822477
const ch = charCodeUnchecked(pos);
24832478
if (inEscape) {
@@ -2493,18 +2488,23 @@ export function createScanner(languageVersion: ScriptTarget, skipTrivia: boolean
24932488
characterClassDepth--;
24942489
}
24952490
else if (!characterClassDepth) {
2496-
if (ch === CharacterCodes.openParen) {
2497-
bracketStack.push(CharacterCodes.closeParen);
2491+
if (ch === CharacterCodes.openBrace) {
2492+
inDecimalQuantifier = true;
24982493
}
2499-
else if (ch === CharacterCodes.openBrace) {
2500-
bracketStack.push(CharacterCodes.closeBrace);
2494+
else if (ch === CharacterCodes.closeBrace && inDecimalQuantifier) {
2495+
inDecimalQuantifier = false;
25012496
}
2502-
else if (ch === lastOrUndefined(bracketStack)) {
2503-
bracketStack.pop();
2504-
}
2505-
else if (ch === CharacterCodes.closeParen || ch === CharacterCodes.closeBracket || ch === CharacterCodes.closeBrace) {
2506-
// We encountered an unbalanced bracket outside a character class. Treat this position as the end of regex.
2507-
break;
2497+
else if (!inDecimalQuantifier) {
2498+
if (ch === CharacterCodes.openParen) {
2499+
groupDepth++;
2500+
}
2501+
else if (ch === CharacterCodes.closeParen && groupDepth) {
2502+
groupDepth--;
2503+
}
2504+
else if (ch === CharacterCodes.closeParen || ch === CharacterCodes.closeBracket || ch === CharacterCodes.closeBrace) {
2505+
// We encountered an unbalanced bracket outside a character class. Treat this position as the end of regex.
2506+
break;
2507+
}
25082508
}
25092509
}
25102510
pos++;
@@ -2517,9 +2517,9 @@ export function createScanner(languageVersion: ScriptTarget, skipTrivia: boolean
25172517
// Consume the slash character
25182518
pos++;
25192519
let regExpFlags = RegularExpressionFlags.None;
2520-
while (pos < end) {
2521-
const ch = codePointUnchecked(pos);
2522-
if (!isIdentifierPart(ch, languageVersion)) {
2520+
while (true) {
2521+
const ch = codePointChecked(pos);
2522+
if (ch === CharacterCodes.EOF || !isIdentifierPart(ch, languageVersion)) {
25232523
break;
25242524
}
25252525
if (reportErrors) {
@@ -2530,7 +2530,7 @@ export function createScanner(languageVersion: ScriptTarget, skipTrivia: boolean
25302530
else if (regExpFlags & flag) {
25312531
error(Diagnostics.Duplicate_regular_expression_flag, pos, 1);
25322532
}
2533-
else if (((regExpFlags | flag) & RegularExpressionFlags.UnicodeMode) === RegularExpressionFlags.UnicodeMode) {
2533+
else if (((regExpFlags | flag) & RegularExpressionFlags.AnyUnicodeMode) === RegularExpressionFlags.AnyUnicodeMode) {
25342534
error(Diagnostics.The_Unicode_u_flag_and_the_Unicode_Sets_v_flag_cannot_be_set_simultaneously, pos, 1);
25352535
}
25362536
else {
@@ -2560,9 +2560,9 @@ export function createScanner(languageVersion: ScriptTarget, skipTrivia: boolean
25602560
/** Grammar parameter */
25612561
var unicodeSetsMode = !!(regExpFlags & RegularExpressionFlags.UnicodeSets);
25622562
/** Grammar parameter */
2563-
var unicodeMode = !!(regExpFlags & RegularExpressionFlags.UnicodeMode);
2563+
var anyUnicodeMode = !!(regExpFlags & RegularExpressionFlags.AnyUnicodeMode);
25642564

2565-
if (unicodeMode) {
2565+
if (anyUnicodeMode) {
25662566
// Annex B treats any unicode mode as the strict syntax.
25672567
annexB = false;
25682568
}
@@ -2719,7 +2719,7 @@ export function createScanner(languageVersion: ScriptTarget, skipTrivia: boolean
27192719
error(Diagnostics.Incomplete_quantifier_Digit_expected, digitsStart, 0);
27202720
}
27212721
else {
2722-
if (unicodeMode) {
2722+
if (anyUnicodeMode) {
27232723
error(Diagnostics.Unexpected_0_Did_you_mean_to_escape_it_with_backslash, start, 1, String.fromCharCode(ch));
27242724
}
27252725
isPreviousTermQuantifiable = true;
@@ -2731,7 +2731,7 @@ export function createScanner(languageVersion: ScriptTarget, skipTrivia: boolean
27312731
}
27322732
}
27332733
else if (!min) {
2734-
if (unicodeMode) {
2734+
if (anyUnicodeMode) {
27352735
error(Diagnostics.Unexpected_0_Did_you_mean_to_escape_it_with_backslash, start, 1, String.fromCharCode(ch));
27362736
}
27372737
isPreviousTermQuantifiable = true;
@@ -2775,7 +2775,7 @@ export function createScanner(languageVersion: ScriptTarget, skipTrivia: boolean
27752775
// falls through
27762776
case CharacterCodes.closeBracket:
27772777
case CharacterCodes.closeBrace:
2778-
if (unicodeMode || ch === CharacterCodes.closeParen) {
2778+
if (anyUnicodeMode || ch === CharacterCodes.closeParen) {
27792779
error(Diagnostics.Unexpected_0_Did_you_mean_to_escape_it_with_backslash, pos, 1, String.fromCharCode(ch));
27802780
}
27812781
pos++;
@@ -2832,7 +2832,7 @@ export function createScanner(languageVersion: ScriptTarget, skipTrivia: boolean
28322832
scanGroupName(/*isReference*/ true);
28332833
scanExpectedChar(CharacterCodes.greaterThan);
28342834
}
2835-
else if (unicodeMode) {
2835+
else if (anyUnicodeMode) {
28362836
error(Diagnostics.k_must_be_followed_by_a_capturing_group_name_enclosed_in_angle_brackets, pos - 2, 2);
28372837
}
28382838
break;
@@ -2875,14 +2875,17 @@ export function createScanner(languageVersion: ScriptTarget, skipTrivia: boolean
28752875
Debug.assertEqual(charCodeUnchecked(pos - 1), CharacterCodes.backslash);
28762876
let ch = charCodeChecked(pos);
28772877
switch (ch) {
2878+
case CharacterCodes.EOF:
2879+
error(Diagnostics.Undetermined_character_escape, pos - 1, 1);
2880+
return "\\";
28782881
case CharacterCodes.c:
28792882
pos++;
28802883
ch = charCodeChecked(pos);
28812884
if (isASCIILetter(ch)) {
28822885
pos++;
28832886
return String.fromCharCode(ch & 0x1f);
28842887
}
2885-
if (unicodeMode) {
2888+
if (anyUnicodeMode) {
28862889
error(Diagnostics.c_must_be_followed_by_an_ASCII_letter, pos - 2, 2);
28872890
}
28882891
else if (atomEscape && annexB) {
@@ -2913,12 +2916,8 @@ export function createScanner(languageVersion: ScriptTarget, skipTrivia: boolean
29132916
pos++;
29142917
return String.fromCharCode(ch);
29152918
default:
2916-
if (pos >= end) {
2917-
error(Diagnostics.Undetermined_character_escape, pos - 1, 1);
2918-
return "\\";
2919-
}
29202919
pos--;
2921-
return scanEscapeSequence(/*shouldEmitInvalidEscapeError*/ unicodeMode, /*isRegularExpression*/ annexB ? "annex-b" : true);
2920+
return scanEscapeSequence(/*shouldEmitInvalidEscapeError*/ anyUnicodeMode, /*isRegularExpression*/ annexB ? "annex-b" : true);
29222921
}
29232922
}
29242923

@@ -3464,11 +3463,11 @@ export function createScanner(languageVersion: ScriptTarget, skipTrivia: boolean
34643463
}
34653464
}
34663465
scanExpectedChar(CharacterCodes.closeBrace);
3467-
if (!unicodeMode) {
3466+
if (!anyUnicodeMode) {
34683467
error(Diagnostics.Unicode_property_value_expressions_are_only_available_when_the_Unicode_u_flag_or_the_Unicode_Sets_v_flag_is_set, start, pos - start);
34693468
}
34703469
}
3471-
else if (unicodeMode) {
3470+
else if (anyUnicodeMode) {
34723471
error(Diagnostics._0_must_be_followed_by_a_Unicode_property_value_expression_enclosed_in_braces, pos - 2, 2, String.fromCharCode(ch));
34733472
}
34743473
return true;
@@ -3490,7 +3489,7 @@ export function createScanner(languageVersion: ScriptTarget, skipTrivia: boolean
34903489
}
34913490

34923491
function scanSourceCharacter(): string {
3493-
const size = unicodeMode ? charSize(charCodeChecked(pos)) : 1;
3492+
const size = anyUnicodeMode ? charSize(charCodeChecked(pos)) : 1;
34943493
pos += size;
34953494
return size > 0 ? text.substring(pos - size, pos) : "";
34963495
}

‎src/compiler/types.ts‎

Lines changed: 11 additions & 11 deletions
Original file line numberDiff line numberDiff line change
@@ -2765,17 +2765,17 @@ export interface RegularExpressionLiteral extends LiteralExpression {
27652765
// dprint-ignore
27662766
/** @internal */
27672767
export const enum RegularExpressionFlags {
2768-
None = 0,
2769-
HasIndices = 1 << 0, // d
2770-
Global = 1 << 1, // g
2771-
IgnoreCase = 1 << 2, // i
2772-
Multiline = 1 << 3, // m
2773-
DotAll = 1 << 4, // s
2774-
Unicode = 1 << 5, // u
2775-
UnicodeSets = 1 << 6, // v
2776-
Sticky = 1 << 7, // y
2777-
UnicodeMode = Unicode | UnicodeSets,
2778-
Modifiers = IgnoreCase | Multiline | DotAll,
2768+
None = 0,
2769+
HasIndices = 1 << 0, // d
2770+
Global = 1 << 1, // g
2771+
IgnoreCase = 1 << 2, // i
2772+
Multiline = 1 << 3, // m
2773+
DotAll = 1 << 4, // s
2774+
Unicode = 1 << 5, // u
2775+
UnicodeSets = 1 << 6, // v
2776+
Sticky = 1 << 7, // y
2777+
AnyUnicodeMode = Unicode | UnicodeSets,
2778+
Modifiers = IgnoreCase | Multiline | DotAll,
27792779
}
27802780

27812781
export interface NoSubstitutionTemplateLiteral extends LiteralExpression, TemplateLiteralLikeNode, Declaration {

‎src/testRunner/tests.ts‎

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -47,7 +47,7 @@ export * from "./unittests/paths.js";
4747
export * from "./unittests/printer.js";
4848
export * from "./unittests/programApi.js";
4949
export * from "./unittests/publicApi.js";
50-
export * from "./unittests/regExpParserRecovery.js";
50+
export * from "./unittests/regExpScannerRecovery.js";
5151
export * from "./unittests/reuseProgramStructure.js";
5252
export * from "./unittests/semver.js";
5353
export * from "./unittests/services/cancellableLanguageServiceOperations.js";

src/testRunner/unittests/regExpParserRecovery.ts renamed to src/testRunner/unittests/regExpScannerRecovery.ts

Lines changed: 21 additions & 21 deletions
Original file line numberDiff line numberDiff line change
@@ -1,6 +1,6 @@
11
import * as ts from "../_namespaces/ts.js";
22

3-
describe("unittests:: regExpParserRecovery", () => {
3+
describe("unittests:: regExpScannerRecovery", () => {
44
const testCases = [
55
"/",
66
"/[]",
@@ -16,7 +16,7 @@ describe("unittests:: regExpParserRecovery", () => {
1616
"/{[]}",
1717
"/([])",
1818
"/[)}({]",
19-
"/({[)}]})",
19+
"/({[]})",
2020
"/\\[",
2121
"/\\{",
2222
"/\\(",
@@ -42,27 +42,27 @@ describe("unittests:: regExpParserRecovery", () => {
4242
"/[\\}]",
4343
"/(\\})",
4444
"/{\\}}",
45-
"/({[\\]})]})",
45+
"/({[\\])]})",
4646
];
4747
const whiteSpaceSequences = [
4848
"",
4949
" ",
50-
"\t\v\r\n",
51-
"\u3000\u2028",
50+
"\t\f",
51+
"\u3000\u2003",
5252
];
53-
it("stops parsing unterminated regexes at correct position", () => {
54-
ts.forEach(testCases, testCase => {
55-
ts.forEach(whiteSpaceSequences, whiteSpaces => {
56-
const testCaseWithWhiteSpaces = testCase + whiteSpaces;
57-
const sources = [
58-
`const regex = ${testCaseWithWhiteSpaces};`,
59-
`(${testCaseWithWhiteSpaces});`,
60-
`([${testCaseWithWhiteSpaces}]);`,
61-
`({prop: ${testCaseWithWhiteSpaces}});`,
62-
`({prop: ([(${testCaseWithWhiteSpaces})])});`,
63-
`({[(${testCaseWithWhiteSpaces}).source]: 42});`,
64-
];
65-
ts.forEach(sources, source => {
53+
for (const testCase of testCases) {
54+
for (const whiteSpaces of whiteSpaceSequences) {
55+
const testCaseWithWhiteSpaces = testCase + whiteSpaces;
56+
const sources = [
57+
`const regex = ${testCaseWithWhiteSpaces};`,
58+
`(${testCaseWithWhiteSpaces});`,
59+
`([${testCaseWithWhiteSpaces}]);`,
60+
`({prop: ${testCaseWithWhiteSpaces}});`,
61+
`({prop: ([(${testCaseWithWhiteSpaces})])});`,
62+
`({[(${testCaseWithWhiteSpaces}).source]: 42});`,
63+
];
64+
for (const source of sources) {
65+
it("stops parsing unterminated regexes at correct position: " + JSON.stringify(source), () => {
6666
const { parseDiagnostics } = ts.createLanguageServiceSourceFile(
6767
/*fileName*/ "",
6868
ts.ScriptSnapshot.fromString(source),
@@ -75,7 +75,7 @@ describe("unittests:: regExpParserRecovery", () => {
7575
assert.equal(diagnostic.start, source.indexOf("/"), "Diagnostic should start at where the regex starts");
7676
assert.equal(diagnostic.length, testCase.length, "Diagnostic should end at where the regex ends");
7777
});
78-
});
79-
});
80-
});
78+
}
79+
}
80+
}
8181
});

0 commit comments

Comments
 (0)