Provide Syntax Checking for Regular Expressions (#55600)

Co-authored-by: Daniel Rosenwasser <DanielRosenwasser@users.noreply.github.com>
Co-authored-by: Jake Bailey <5341706+jakebailey@users.noreply.github.com>

src/compiler/core.ts

@@ -2289,7 +2289,7 @@ export function compareBooleans(a: boolean, b: boolean): Comparison {
  *
  * @internal
  */
-export function getSpellingSuggestion<T>(name: string, candidates: T[], getName: (candidate: T) => string | undefined): T | undefined {
+export function getSpellingSuggestion<T>(name: string, candidates: Iterable<T>, getName: (candidate: T) => string | undefined): T | undefined {
     const maximumLengthDifference = Math.max(2, Math.floor(name.length * 0.34));
     let bestDistance = Math.floor(name.length * 0.4) + 1; // If the best result is worse than this, don't bother.
     let bestCandidate: T | undefined;

src/compiler/diagnosticMessages.json

@@ -1649,6 +1649,154 @@
         "category": "Error",
         "code": 1498
     },
+    "Unknown regular expression flag.": {
+        "category": "Error",
+        "code": 1499
+    },
+    "Duplicate regular expression flag.": {
+        "category": "Error",
+        "code": 1500
+    },
+    "This regular expression flag is only available when targeting '{0}' or later.": {
+        "category": "Error",
+        "code": 1501
+    },
+    "The Unicode (u) flag and the Unicode Sets (v) flag cannot be set simultaneously.": {
+        "category": "Error",
+        "code": 1502
+    },
+    "Named capturing groups are only available when targeting 'ES2018' or later.": {
+        "category": "Error",
+        "code": 1503
+    },
+    "Subpattern flags must be present when there is a minus sign.": {
+        "category": "Error",
+        "code": 1504
+    },
+    "Incomplete quantifier. Digit expected.": {
+        "category": "Error",
+        "code": 1505
+    },
+    "Numbers out of order in quantifier.": {
+        "category": "Error",
+        "code": 1506
+    },
+    "There is nothing available for repetition.": {
+        "category": "Error",
+        "code": 1507
+    },
+    "Unexpected '{0}'. Did you mean to escape it with backslash?": {
+        "category": "Error",
+        "code": 1508
+    },
+    "This regular expression flag cannot be toggled within a subpattern.": {
+        "category": "Error",
+        "code": 1509
+    },
+    "'\\k' must be followed by a capturing group name enclosed in angle brackets.": {
+        "category": "Error",
+        "code": 1510
+    },
+    "'\\q' is only available inside character class.": {
+        "category": "Error",
+        "code": 1511
+    },
+    "'\\c' must be followed by an ASCII letter.": {
+        "category": "Error",
+        "code": 1512
+    },
+    "Undetermined character escape.": {
+        "category": "Error",
+        "code": 1513
+    },
+    "Expected a capturing group name.": {
+        "category": "Error",
+        "code": 1514
+    },
+    "Named capturing groups with the same name must be mutually exclusive to each other.": {
+        "category": "Error",
+        "code": 1515
+    },
+    "A character class range must not be bounded by another character class.": {
+        "category": "Error",
+        "code": 1516
+    },
+    "Range out of order in character class.": {
+        "category": "Error",
+        "code": 1517
+    },
+    "Anything that would possibly match more than a single character is invalid inside a negated character class.": {
+        "category": "Error",
+        "code": 1518
+    },
+    "Operators must not be mixed within a character class. Wrap it in a nested class instead.": {
+        "category": "Error",
+        "code": 1519
+    },
+    "Expected a class set oprand.": {
+        "category": "Error",
+        "code": 1520
+    },
+    "'\\q' must be followed by string alternatives enclosed in braces.": {
+        "category": "Error",
+        "code": 1521
+    },
+    "A character class must not contain a reserved double punctuator. Did you mean to escape it with backslash?": {
+        "category": "Error",
+        "code": 1522
+    },
+    "Expected a Unicode property name.": {
+        "category": "Error",
+        "code": 1523
+    },
+    "Unknown Unicode property name.": {
+        "category": "Error",
+        "code": 1524
+    },
+    "Expected a Unicode property value.": {
+        "category": "Error",
+        "code": 1525
+    },
+    "Unknown Unicode property value.": {
+        "category": "Error",
+        "code": 1526
+    },
+    "Expected a Unicode property name or value.": {
+        "category": "Error",
+        "code": 1527
+    },
+    "Any Unicode property that would possibly match more than a single character is only available when the Unicode Sets (v) flag is set.": {
+        "category": "Error",
+        "code": 1528
+    },
+    "Unknown Unicode property name or value.": {
+        "category": "Error",
+        "code": 1529
+    },
+    "Unicode property value expressions are only available when the Unicode (u) flag or the Unicode Sets (v) flag is set.": {
+        "category": "Error",
+        "code": 1530
+    },
+    "'\\{0}' must be followed by a Unicode property value expression enclosed in braces.": {
+        "category": "Error",
+        "code": 1531
+    },
+    "There is no capturing group named '{0}' in this regular expression.": {
+        "category": "Error",
+        "code": 1532
+    },
+    "A decimal escape must refer to an existent capturing group. There are only {0} capturing groups in this regular expression.": {
+        "category": "Error",
+        "code": 1533
+    },
+    "Decimal escapes are invalid when there are no capturing groups in a regular expression.": {
+        "category": "Error",
+        "code": 1534
+    },
+    "This character cannot be escaped in a regular expression.": {
+        "category": "Error",
+        "code": 1535
+    },
 
     "The types of '{0}' are incompatible between these types.": {
         "category": "Error",

src/compiler/parser.ts

@@ -62,6 +62,7 @@ import {
     DeleteExpression,
     Diagnostic,
     DiagnosticArguments,
+    DiagnosticCategory,
     DiagnosticMessage,
     Diagnostics,
     DiagnosticWithDetachedLocation,
@@ -113,6 +114,7 @@ import {
     HasModifiers,
     HeritageClause,
     Identifier,
+    identity,
     idText,
     IfStatement,
     ImportAttribute,
@@ -2142,7 +2144,11 @@ namespace Parser {
         // Don't report another error if it would just be at the same position as the last error.
         const lastError = lastOrUndefined(parseDiagnostics);
         let result: DiagnosticWithDetachedLocation | undefined;
-        if (!lastError || start !== lastError.start) {
+        if (message.category === DiagnosticCategory.Message && lastError && start === lastError.start && length === lastError.length) {
+            result = createDetachedDiagnostic(fileName, sourceText, start, length, message, ...args);
+            addRelatedInfo(lastError, result);
+        }
+        else if (!lastError || start !== lastError.start) {
             result = createDetachedDiagnostic(fileName, sourceText, start, length, message, ...args);
             parseDiagnostics.push(result);
         }
@@ -2398,7 +2404,7 @@ namespace Parser {
         }
 
         // The user alternatively might have misspelled or forgotten to add a space after a common keyword.
-        const suggestion = getSpellingSuggestion(expressionText, viableKeywordSuggestions, n => n) ?? getSpaceSuggestion(expressionText);
+        const suggestion = getSpellingSuggestion(expressionText, viableKeywordSuggestions, identity) ?? getSpaceSuggestion(expressionText);
         if (suggestion) {
             parseErrorAt(pos, node.end, Diagnostics.Unknown_keyword_or_identifier_Did_you_mean_0, suggestion);
             return;

src/compiler/program.ts

@@ -126,6 +126,7 @@ import {
     getLineStarts,
     getMatchedFileSpec,
     getMatchedIncludeSpec,
+    getNameOfScriptTarget,
     getNewLineCharacter,
     getNormalizedAbsolutePath,
     getNormalizedAbsolutePathWithoutRoot,
@@ -307,7 +308,6 @@ import {
     SyntaxKind,
     sys,
     System,
-    targetOptionDeclaration,
     toFileNameLowerCase,
     tokenToString,
     toPath as ts_toPath,
@@ -4780,7 +4780,7 @@ export function createProgram(rootNamesOrOptions: readonly string[] | CreateProg
                     message = Diagnostics.File_is_library_specified_here;
                     break;
                 }
-                const target = forEachEntry(targetOptionDeclaration.type, (value, key) => value === getEmitScriptTarget(options) ? key : undefined);
+                const target = getNameOfScriptTarget(getEmitScriptTarget(options));
                 configFileNode = target ? getOptionsSyntaxByValue("target", target) : undefined;
                 message = Diagnostics.File_is_default_library_for_target_specified_here;
                 break;

src/compiler/types.ts

@@ -2762,6 +2762,22 @@ export interface RegularExpressionLiteral extends LiteralExpression {
     readonly kind: SyntaxKind.RegularExpressionLiteral;
 }
 
+// dprint-ignore
+/** @internal */
+export const enum RegularExpressionFlags {
+    None        = 0,
+    HasIndices  = 1 << 0, // d
+    Global      = 1 << 1, // g
+    IgnoreCase  = 1 << 2, // i
+    Multiline   = 1 << 3, // m
+    DotAll      = 1 << 4, // s
+    Unicode     = 1 << 5, // u
+    UnicodeSets = 1 << 6, // v
+    Sticky      = 1 << 7, // y
+    UnicodeMode = Unicode | UnicodeSets,
+    Modifiers   = IgnoreCase | Multiline | DotAll,
+}
+
 export interface NoSubstitutionTemplateLiteral extends LiteralExpression, TemplateLiteralLikeNode, Declaration {
     readonly kind: SyntaxKind.NoSubstitutionTemplateLiteral;
     /** @internal */

src/compiler/utilities.ts

@@ -530,6 +530,7 @@ import {
     SymbolTable,
     SyntaxKind,
     TaggedTemplateExpression,
+    targetOptionDeclaration,
     TemplateExpression,
     TemplateLiteral,
     TemplateLiteralLikeNode,
@@ -8986,6 +8987,11 @@ export function getStrictOptionValue(compilerOptions: CompilerOptions, flag: Str
     return compilerOptions[flag] === undefined ? !!compilerOptions.strict : !!compilerOptions[flag];
 }
 
+/** @internal */
+export function getNameOfScriptTarget(scriptTarget: ScriptTarget): string | undefined {
+    return forEachEntry(targetOptionDeclaration.type, (value, key) => value === scriptTarget ? key : undefined);
+}
+
 /** @internal */
 export function getEmitStandardClassFields(compilerOptions: CompilerOptions) {
     return compilerOptions.useDefineForClassFields !== false && getEmitScriptTarget(compilerOptions) >= ScriptTarget.ES2022;

src/compiler/watch.ts

@@ -47,7 +47,6 @@ import {
     findIndex,
     flattenDiagnosticMessageText,
     forEach,
-    forEachEntry,
     ForegroundColorEscapeSequences,
     formatColorAndReset,
     formatDiagnostic,
@@ -59,6 +58,7 @@ import {
     getEmitScriptTarget,
     getImpliedNodeFormatForEmitWorker,
     getLineAndCharacterOfPosition,
+    getNameOfScriptTarget,
     getNewLineCharacter,
     getNormalizedAbsolutePath,
     getParsedCommandLineOfConfigFile,
@@ -95,7 +95,6 @@ import {
     sourceMapCommentRegExpDontCareLineStart,
     sys,
     System,
-    targetOptionDeclaration,
     WatchCompilerHost,
     WatchCompilerHostOfConfigFile,
     WatchCompilerHostOfFilesAndCompilerOptions,
@@ -542,7 +541,7 @@ export function fileIncludeReasonToDiagnostics(program: Program, reason: FileInc
         }
         case FileIncludeKind.LibFile: {
             if (reason.index !== undefined) return chainDiagnosticMessages(/*details*/ undefined, Diagnostics.Library_0_specified_in_compilerOptions, options.lib![reason.index]);
-            const target = forEachEntry(targetOptionDeclaration.type, (value, key) => value === getEmitScriptTarget(options) ? key : undefined);
+            const target = getNameOfScriptTarget(getEmitScriptTarget(options));
             const messageAndArgs: DiagnosticAndArguments = target ? [Diagnostics.Default_library_for_target_0, target] : [Diagnostics.Default_library];
             return chainDiagnosticMessages(/*details*/ undefined, ...messageAndArgs);
         }