agentic-qe 3.7.8 → 3.7.10

package/dist/shared/parsers/multi-language-parser.js

@@ -0,0 +1,1271 @@
+/**
+ * Multi-Language Parser Registry (ADR-076)
+ * Regex-based parsers for non-TypeScript languages.
+ * Extracts function/class/import declarations for test generation.
+ *
+ * Uses two key utilities to handle complex syntax that naive line-by-line regex misses:
+ * 1. joinMultiLineSignatures — collapses multi-line function/class declarations into single logical lines
+ * 2. matchBalancedBrackets — handles nested generics like Map<String, List<Foo>> correctly
+ */
+// ============================================================================
+// Shared Parsing Utilities
+// ============================================================================
+// Maximum line length for regex matching. Joined multi-line signatures can be long;
+// lines beyond this limit are skipped to prevent polynomial-time regex backtracking (CWE-1333).
+const MAX_REGEX_LINE_LEN = 1000;
+/**
+ * Join multi-line signatures into single logical lines.
+ * Handles: function params spanning lines, class declarations with long implements lists,
+ * multi-line annotations/attributes, and continuation lines.
+ *
+ * Returns { logicalLines, lineMap } where lineMap[logicalIndex] = originalStartLine (1-based).
+ */
+export function joinMultiLineSignatures(content) {
+    const rawLines = content.split('\n');
+    const logicalLines = [];
+    const lineMap = [];
+    let buffer = '';
+    let bufferStartLine = 0;
+    let openParens = 0;
+    let openAngles = 0;
+    let openSquare = 0;
+    for (let i = 0; i < rawLines.length; i++) {
+        const line = rawLines[i];
+        const trimmed = line.trim();
+        // Skip pure comment/blank lines — emit them as-is
+        if (trimmed === '' || trimmed.startsWith('//') || trimmed.startsWith('#') || trimmed.startsWith('*')) {
+            if (buffer) {
+                // Inside a continuation — keep accumulating
+                buffer += ' ' + trimmed;
+                continue;
+            }
+            logicalLines.push(line);
+            lineMap.push(i + 1);
+            continue;
+        }
+        if (!buffer) {
+            bufferStartLine = i + 1;
+        }
+        buffer = buffer ? buffer + ' ' + trimmed : line;
+        // Count bracket balance (ignoring strings/comments for simplicity — good enough for declarations)
+        for (const ch of trimmed) {
+            if (ch === '(')
+                openParens++;
+            else if (ch === ')')
+                openParens = Math.max(0, openParens - 1);
+            else if (ch === '<')
+                openAngles++;
+            else if (ch === '>')
+                openAngles = Math.max(0, openAngles - 1);
+            else if (ch === '[')
+                openSquare++;
+            else if (ch === ']')
+                openSquare = Math.max(0, openSquare - 1);
+        }
+        // Line is complete when all brackets are balanced
+        if (openParens === 0 && openAngles === 0 && openSquare === 0) {
+            logicalLines.push(buffer);
+            lineMap.push(bufferStartLine);
+            buffer = '';
+        }
+    }
+    // Flush remaining buffer
+    if (buffer) {
+        logicalLines.push(buffer);
+        lineMap.push(bufferStartLine);
+    }
+    return { logicalLines, lineMap };
+}
+/**
+ * Extract balanced bracket content starting at a given position.
+ * Handles nested brackets: Map<String, List<Foo>> returns "String, List<Foo>"
+ *
+ * @param str - The string to scan
+ * @param startIdx - Index of the opening bracket character
+ * @param open - Opening bracket char (default '<')
+ * @param close - Closing bracket char (default '>')
+ * @returns The content between brackets (excluding the brackets), or undefined if unbalanced
+ */
+export function matchBalancedBrackets(str, startIdx, open = '<', close = '>') {
+    if (str[startIdx] !== open)
+        return undefined;
+    let depth = 1;
+    let i = startIdx + 1;
+    while (i < str.length && depth > 0) {
+        if (str[i] === open)
+            depth++;
+        else if (str[i] === close)
+            depth--;
+        i++;
+    }
+    if (depth !== 0)
+        return undefined;
+    return str.substring(startIdx + 1, i - 1);
+}
+/**
+ * Extract the full type expression starting at an index, handling nested generics.
+ * E.g., from "Map<String, List<Foo>> remaining" extracts "Map<String, List<Foo>>"
+ */
+export function extractFullType(str, startIdx = 0) {
+    let i = startIdx;
+    // Skip whitespace
+    while (i < str.length && str[i] === ' ')
+        i++;
+    const typeStart = i;
+    // Walk through identifier chars
+    while (i < str.length && /[\w.]/.test(str[i]))
+        i++;
+    // If next char is '<', consume balanced generics
+    if (i < str.length && str[i] === '<') {
+        const inner = matchBalancedBrackets(str, i);
+        if (inner !== undefined) {
+            i += inner.length + 2; // +2 for < and >
+        }
+    }
+    // Handle array suffixes [], ?, !, etc.
+    while (i < str.length && /[[\]?!*&]/.test(str[i]))
+        i++;
+    return str.substring(typeStart, i).trim();
+}
+/**
+ * Collect decorators/attributes by scanning backwards from a line index.
+ * Handles multi-line attribute chains like:
+ *   #[derive(Debug)]
+ *   #[serde(rename_all = "camelCase")]
+ *   pub struct Foo { ... }
+ *
+ * @param lines - Array of source lines
+ * @param lineIdx - Index of the declaration line (0-based)
+ * @param prefix - Decorator prefix to look for ('@', '#[', '[')
+ */
+export function collectDecorators(lines, lineIdx, prefix) {
+    const decorators = [];
+    for (let j = lineIdx - 1; j >= 0; j--) {
+        const trimmed = lines[j].trim();
+        if (trimmed.startsWith(prefix)) {
+            decorators.unshift(trimmed);
+        }
+        else if (trimmed === '' || trimmed.startsWith('//') || trimmed.startsWith('*') || trimmed.startsWith('#')) {
+            // Allow blank lines and comments between decorators
+            continue;
+        }
+        else {
+            break;
+        }
+    }
+    return decorators;
+}
+/**
+ * Parse a balanced parenthesized parameter string.
+ * Unlike regex [^)]*, this handles nested parens/generics inside parameter types.
+ * Input: content after opening '(' — e.g. "id: Long, filter: Map<String, List<Int>>"
+ */
+export function extractParenContent(line, openIdx) {
+    return matchBalancedBrackets(line, openIdx, '(', ')');
+}
+/**
+ * Split a parameter string by commas, respecting nested generics depth.
+ * "String id, Map<String, List<Foo>> map" -> ["String id", "Map<String, List<Foo>> map"]
+ */
+function splitParamsGenericAware(paramsStr) {
+    const params = [];
+    let depth = 0;
+    let current = '';
+    for (const ch of paramsStr) {
+        if (ch === '<' || ch === '(')
+            depth++;
+        else if (ch === '>' || ch === ')')
+            depth--;
+        if (ch === ',' && depth === 0) {
+            params.push(current.trim());
+            current = '';
+        }
+        else {
+            current += ch;
+        }
+    }
+    if (current.trim())
+        params.push(current.trim());
+    return params;
+}
+/**
+ * Count braces in a line to track nesting depth.
+ * Returns the delta (opens - closes).
+ */
+function countBraceDelta(line) {
+    let delta = 0;
+    let inString = false;
+    let stringChar = '';
+    for (let i = 0; i < line.length; i++) {
+        const ch = line[i];
+        if (inString) {
+            if (ch === stringChar && line[i - 1] !== '\\')
+                inString = false;
+            continue;
+        }
+        if (ch === '"' || ch === '\'' || ch === '`') {
+            inString = true;
+            stringChar = ch;
+        }
+        else if (ch === '{') {
+            delta++;
+        }
+        else if (ch === '}') {
+            delta--;
+        }
+    }
+    return delta;
+}
+// --- Python Parser ---
+class PythonParser {
+    language = 'python';
+    supportedExtensions = ['.py'];
+    async parseFile(content, filePath) {
+        return {
+            functions: this.extractFunctions(content),
+            classes: this.extractClasses(content),
+            imports: this.extractImports(content),
+            language: 'python',
+            filePath,
+        };
+    }
+    extractFunctions(content) {
+        const functions = [];
+        const rawLines = content.split('\n');
+        const { logicalLines, lineMap } = joinMultiLineSignatures(content);
+        // Match: def function_name(params) -> return_type:
+        //    or: async def function_name(params) -> return_type:
+        const funcRegex = /^(\s*)(async\s+)?def\s+(\w+)\s*\(([^)]*)\)(?:\s*->\s*([^:]*\S))?\s*:/;
+        for (let i = 0; i < logicalLines.length; i++) {
+            if (logicalLines[i].length > MAX_REGEX_LINE_LEN)
+                continue;
+            const match = logicalLines[i].match(funcRegex);
+            if (match) {
+                const indent = match[1].length;
+                const isAsync = !!match[2];
+                const name = match[3];
+                const paramsStr = match[4];
+                const returnType = match[5]?.trim();
+                const origLine = lineMap[i] - 1; // 0-based index into rawLines
+                // Find end of function (next line at same or lower indent, or end of file)
+                let endLine = origLine + 1;
+                for (let j = origLine + 1; j < rawLines.length; j++) {
+                    const line = rawLines[j];
+                    if (line.trim() === '' || line.trim().startsWith('#'))
+                        continue;
+                    const lineIndent = line.match(/^(\s*)/)?.[1].length ?? 0;
+                    if (lineIndent <= indent && line.trim() !== '') {
+                        endLine = j;
+                        break;
+                    }
+                    endLine = j + 1;
+                }
+                const decorators = collectDecorators(rawLines, origLine, '@');
+                functions.push({
+                    name,
+                    parameters: this.parseParams(paramsStr),
+                    returnType,
+                    isAsync,
+                    isPublic: !name.startsWith('_'),
+                    complexity: 1,
+                    decorators,
+                    genericParams: [],
+                    startLine: lineMap[i],
+                    endLine,
+                });
+            }
+        }
+        return functions;
+    }
+    extractClasses(content) {
+        const classes = [];
+        const rawLines = content.split('\n');
+        const { logicalLines, lineMap } = joinMultiLineSignatures(content);
+        const classRegex = /^(\s*)class\s+(\w+)(?:\(([^)]*)\))?\s*:/;
+        for (let i = 0; i < logicalLines.length; i++) {
+            const match = logicalLines[i].match(classRegex);
+            if (match) {
+                const indent = match[1].length;
+                const name = match[2];
+                const bases = match[3]
+                    ? match[3].split(',').map((b) => b.trim())
+                    : [];
+                const origLine = lineMap[i] - 1;
+                // Find end of class
+                let endLine = origLine + 1;
+                for (let j = origLine + 1; j < rawLines.length; j++) {
+                    const line = rawLines[j];
+                    if (line.trim() === '' || line.trim().startsWith('#'))
+                        continue;
+                    const lineIndent = line.match(/^(\s*)/)?.[1].length ?? 0;
+                    if (lineIndent <= indent && line.trim() !== '') {
+                        endLine = j;
+                        break;
+                    }
+                    endLine = j + 1;
+                }
+                // Extract methods within class
+                const classContent = rawLines.slice(origLine + 1, endLine).join('\n');
+                const methods = this.extractFunctions(classContent).map((f) => ({
+                    ...f,
+                    startLine: f.startLine + origLine,
+                    endLine: f.endLine + origLine,
+                }));
+                const decorators = collectDecorators(rawLines, origLine, '@');
+                classes.push({
+                    name,
+                    methods,
+                    properties: [],
+                    isPublic: !name.startsWith('_'),
+                    implements: [],
+                    extends: bases[0] || undefined,
+                    decorators,
+                    startLine: lineMap[i],
+                    endLine,
+                });
+            }
+        }
+        return classes;
+    }
+    extractImports(content) {
+        const imports = [];
+        const lines = content.split('\n');
+        for (const line of lines) {
+            if (line.length > MAX_REGEX_LINE_LEN)
+                continue;
+            // from module import name1, name2
+            const fromMatch = line.match(/^\s*from\s+(\S+)\s+import\s+(.+)/);
+            if (fromMatch) {
+                const names = fromMatch[2]
+                    .split(',')
+                    .map((n) => n.trim().split(' as ')[0]);
+                imports.push({
+                    module: fromMatch[1],
+                    namedImports: names,
+                    isTypeOnly: false,
+                });
+                continue;
+            }
+            // import module
+            const importMatch = line.match(/^\s*import\s+(\S+)/);
+            if (importMatch) {
+                imports.push({
+                    module: importMatch[1],
+                    namedImports: [],
+                    isTypeOnly: false,
+                });
+            }
+        }
+        return imports;
+    }
+    parseParams(paramsStr) {
+        if (!paramsStr.trim())
+            return [];
+        return splitParamsGenericAware(paramsStr)
+            .map((p) => p.trim())
+            .filter((p) => p && p !== 'self' && p !== 'cls')
+            .map((p) => {
+            const [nameDefault, type] = p.split(':').map((s) => s.trim());
+            const [name, defaultValue] = nameDefault
+                .split('=')
+                .map((s) => s.trim());
+            return {
+                name,
+                type: type?.split('=')[0].trim() || undefined,
+                isOptional: !!defaultValue || name.startsWith('*'),
+                defaultValue: defaultValue || undefined,
+            };
+        });
+    }
+}
+// --- Java Parser ---
+class JavaParser {
+    language = 'java';
+    supportedExtensions = ['.java'];
+    async parseFile(content, filePath) {
+        return {
+            functions: this.extractFunctions(content),
+            classes: this.extractClasses(content),
+            imports: this.extractImports(content),
+            language: 'java',
+            filePath,
+        };
+    }
+    extractFunctions(content) {
+        const functions = [];
+        const rawLines = content.split('\n');
+        const { logicalLines, lineMap } = joinMultiLineSignatures(content);
+        // Match method declarations — use a simpler regex without <[^>]+> so generics don't break
+        const methodRegex = /^\s*(public|private|protected)?\s*(static)?\s*(async)?\s*(?:<\S+>\s+)?([\w<>,.[\]?]+(?:\s+[\w<>,.[\]?]+)*)\s+(\w+)\s*\(([^)]*)\)/;
+        for (let i = 0; i < logicalLines.length; i++) {
+            const line = logicalLines[i];
+            if (line.includes(' class ') || line.includes(' interface '))
+                continue;
+            if (line.length > MAX_REGEX_LINE_LEN)
+                continue;
+            const match = line.match(methodRegex);
+            if (match) {
+                const visibility = match[1] || 'default';
+                const name = match[5];
+                const paramsStr = match[6];
+                // Extract the full return type with balanced generics
+                const returnTypeRaw = match[4]?.trim();
+                let returnType = returnTypeRaw;
+                // If the raw return type was truncated by the regex, try to extract full type
+                if (returnTypeRaw && returnTypeRaw.includes('<')) {
+                    const angleIdx = returnTypeRaw.indexOf('<');
+                    const inner = matchBalancedBrackets(returnTypeRaw, angleIdx);
+                    if (inner !== undefined) {
+                        returnType = returnTypeRaw.substring(0, angleIdx) + '<' + inner + '>';
+                    }
+                }
+                const origLine = lineMap[i] - 1;
+                const decorators = collectDecorators(rawLines, origLine, '@');
+                functions.push({
+                    name,
+                    parameters: this.parseParams(paramsStr),
+                    returnType: returnType === 'void' ? undefined : returnType,
+                    isAsync: returnType?.includes('CompletableFuture') || false,
+                    isPublic: visibility === 'public',
+                    complexity: 1,
+                    decorators,
+                    genericParams: [],
+                    startLine: lineMap[i],
+                    endLine: lineMap[i],
+                });
+            }
+        }
+        return functions;
+    }
+    extractClasses(content) {
+        const classes = [];
+        const rawLines = content.split('\n');
+        const { logicalLines, lineMap } = joinMultiLineSignatures(content);
+        const classRegex = /^\s*(public|private|protected)?\s*(abstract)?\s*class\s+(\w+)(?:<[^{]*>)?(?:\s+extends\s+(\w+))?(?:\s+implements\s+([^{\s]+(?:\s+[^{\s]+)*))?\s*\{?/;
+        for (let i = 0; i < logicalLines.length; i++) {
+            if (logicalLines[i].length > MAX_REGEX_LINE_LEN)
+                continue;
+            const match = logicalLines[i].match(classRegex);
+            if (match) {
+                const name = match[3];
+                const extendsName = match[4];
+                const implementsList = match[5]
+                    ? match[5].split(',').map((s) => s.trim())
+                    : [];
+                const origLine = lineMap[i] - 1;
+                const decorators = collectDecorators(rawLines, origLine, '@');
+                classes.push({
+                    name,
+                    methods: [],
+                    properties: [],
+                    isPublic: (match[1] || 'default') === 'public',
+                    implements: implementsList,
+                    extends: extendsName,
+                    decorators,
+                    startLine: lineMap[i],
+                    endLine: lineMap[i],
+                });
+            }
+        }
+        return classes;
+    }
+    extractImports(content) {
+        const imports = [];
+        const lines = content.split('\n');
+        for (const line of lines) {
+            if (line.length > MAX_REGEX_LINE_LEN)
+                continue;
+            const match = line.match(/^\s*import +(static +)?(\S[^;]*);/);
+            if (match) {
+                const module = match[2].trim();
+                const parts = module.split('.');
+                const name = parts[parts.length - 1];
+                imports.push({
+                    module,
+                    namedImports: name === '*' ? [] : [name],
+                    isTypeOnly: false,
+                });
+            }
+        }
+        return imports;
+    }
+    parseParams(paramsStr) {
+        if (!paramsStr.trim())
+            return [];
+        return splitParamsGenericAware(paramsStr).map((p) => {
+            const parts = p.trim().split(/\s+/);
+            // Handle annotations: @NotNull String name
+            const filtered = parts.filter((part) => !part.startsWith('@'));
+            const name = filtered[filtered.length - 1] || '';
+            const type = filtered.slice(0, -1).join(' ') || undefined;
+            return { name, type, isOptional: false, defaultValue: undefined };
+        });
+    }
+}
+// --- C# Parser ---
+class CSharpParser {
+    language = 'csharp';
+    supportedExtensions = ['.cs'];
+    async parseFile(content, filePath) {
+        return {
+            functions: this.extractMethods(content),
+            classes: this.extractClasses(content),
+            imports: this.extractImports(content),
+            language: 'csharp',
+            filePath,
+        };
+    }
+    extractMethods(content) {
+        const functions = [];
+        const rawLines = content.split('\n');
+        const { logicalLines, lineMap } = joinMultiLineSignatures(content);
+        const methodRegex = /^\s*(public|private|protected|internal)?\s*(static)?\s*(async)?\s*([\w<>,.[\]?]+(?:\s+[\w<>,.[\]?]+)*)\s+(\w+)\s*\(([^)]*)\)/;
+        for (let i = 0; i < logicalLines.length; i++) {
+            const line = logicalLines[i];
+            if (line.includes(' class ') || line.includes(' interface ') || line.includes(' namespace '))
+                continue;
+            if (line.length > MAX_REGEX_LINE_LEN)
+                continue;
+            const match = line.match(methodRegex);
+            if (match) {
+                const visibility = match[1] || 'private';
+                const isAsync = !!match[3] || match[4]?.includes('Task') || false;
+                const returnTypeRaw = match[4]?.trim();
+                let returnType = returnTypeRaw;
+                // Extract full generic type
+                if (returnTypeRaw && returnTypeRaw.includes('<')) {
+                    const angleIdx = returnTypeRaw.indexOf('<');
+                    const inner = matchBalancedBrackets(returnTypeRaw, angleIdx);
+                    if (inner !== undefined) {
+                        returnType = returnTypeRaw.substring(0, angleIdx) + '<' + inner + '>';
+                    }
+                }
+                const name = match[5];
+                const paramsStr = match[6];
+                const origLine = lineMap[i] - 1;
+                const decorators = collectDecorators(rawLines, origLine, '[');
+                functions.push({
+                    name,
+                    parameters: this.parseParams(paramsStr),
+                    returnType: returnType === 'void' ? undefined : returnType,
+                    isAsync,
+                    isPublic: visibility === 'public',
+                    complexity: 1,
+                    decorators,
+                    genericParams: [],
+                    startLine: lineMap[i],
+                    endLine: lineMap[i],
+                });
+            }
+        }
+        return functions;
+    }
+    extractClasses(content) {
+        const classes = [];
+        const rawLines = content.split('\n');
+        const { logicalLines, lineMap } = joinMultiLineSignatures(content);
+        const classRegex = /^\s*(public|private|protected|internal)?\s*(abstract|sealed|static|partial)?\s*class\s+(\w+)(?:<[^{]*>)?(?:\s*:\s*([^{\s]+(?:\s+[^{\s]+)*))?\s*\{?/;
+        for (let i = 0; i < logicalLines.length; i++) {
+            if (logicalLines[i].length > MAX_REGEX_LINE_LEN)
+                continue;
+            const match = logicalLines[i].match(classRegex);
+            if (match) {
+                const name = match[3];
+                const bases = match[4]
+                    ? match[4].split(',').map((s) => s.trim())
+                    : [];
+                const origLine = lineMap[i] - 1;
+                const decorators = collectDecorators(rawLines, origLine, '[');
+                classes.push({
+                    name,
+                    methods: [],
+                    properties: [],
+                    isPublic: (match[1] || 'internal') === 'public',
+                    implements: bases.slice(1),
+                    extends: bases[0] || undefined,
+                    decorators,
+                    startLine: lineMap[i],
+                    endLine: lineMap[i],
+                });
+            }
+        }
+        return classes;
+    }
+    extractImports(content) {
+        const imports = [];
+        const lines = content.split('\n');
+        for (const line of lines) {
+            if (line.length > MAX_REGEX_LINE_LEN)
+                continue;
+            const match = line.match(/^\s*using +(static +)?(\S[^;]*);/);
+            if (match) {
+                imports.push({
+                    module: match[2].trim(),
+                    namedImports: [],
+                    isTypeOnly: false,
+                });
+            }
+        }
+        return imports;
+    }
+    parseParams(paramsStr) {
+        if (!paramsStr.trim())
+            return [];
+        return splitParamsGenericAware(paramsStr).map((p) => {
+            const parts = p.trim().split(/\s+/);
+            const name = parts[parts.length - 1] || '';
+            const type = parts.slice(0, -1).join(' ') || undefined;
+            return {
+                name,
+                type,
+                isOptional: type?.endsWith('?') || false,
+                defaultValue: undefined,
+            };
+        });
+    }
+}
+// --- Go Parser ---
+class GoParser {
+    language = 'go';
+    supportedExtensions = ['.go'];
+    async parseFile(content, filePath) {
+        return {
+            functions: this.extractFunctions(content),
+            classes: this.extractStructs(content),
+            imports: this.extractImports(content),
+            language: 'go',
+            filePath,
+        };
+    }
+    extractFunctions(content) {
+        const functions = [];
+        const rawLines = content.split('\n');
+        const { logicalLines, lineMap } = joinMultiLineSignatures(content);
+        // func Name(params) returnType {
+        // func (r *Receiver) Name(params) returnType {
+        const funcRegex = /^\s*func\s+(?:\((\w+)\s+\*?(\w+)\)\s+)?(\w+)\s*\(([^)]*)\)\s*([\w*[\](),\s]*)\s*\{?/;
+        // Track brace depth to filter nested function literals (Go closures).
+        // Go methods use receiver syntax at top level, so only depth 0 is valid.
+        let braceDepth = 0;
+        for (let i = 0; i < logicalLines.length; i++) {
+            const line = logicalLines[i];
+            if (line.length > MAX_REGEX_LINE_LEN) {
+                braceDepth += (line.match(/\{/g)?.length ?? 0) - (line.match(/\}/g)?.length ?? 0);
+                continue;
+            }
+            const match = line.match(funcRegex);
+            if (match && braceDepth === 0) {
+                const name = match[3];
+                const paramsStr = match[4];
+                const returnType = match[5]?.trim() || undefined;
+                functions.push({
+                    name,
+                    parameters: this.parseParams(paramsStr),
+                    returnType: returnType || undefined,
+                    isAsync: false,
+                    isPublic: name[0] === name[0].toUpperCase(),
+                    complexity: 1,
+                    decorators: [],
+                    genericParams: [],
+                    startLine: lineMap[i],
+                    endLine: lineMap[i],
+                });
+            }
+            // Update brace depth after checking for function match
+            braceDepth += countBraceDelta(line);
+            if (braceDepth < 0)
+                braceDepth = 0;
+        }
+        return functions;
+    }
+    extractStructs(content) {
+        const structs = [];
+        const { logicalLines, lineMap } = joinMultiLineSignatures(content);
+        const structRegex = /^\s*type\s+(\w+)\s+struct\s*\{/;
+        for (let i = 0; i < logicalLines.length; i++) {
+            const match = logicalLines[i].match(structRegex);
+            if (match) {
+                const name = match[1];
+                structs.push({
+                    name,
+                    methods: [],
+                    properties: [],
+                    isPublic: name[0] === name[0].toUpperCase(),
+                    implements: [],
+                    extends: undefined,
+                    decorators: [],
+                    startLine: lineMap[i],
+                    endLine: lineMap[i],
+                });
+            }
+        }
+        return structs;
+    }
+    extractImports(content) {
+        const imports = [];
+        const singleImport = /^\s*import\s+"([^"]+)"/;
+        const multiImportStart = /^\s*import\s*\(/;
+        const multiImportLine = /^\s*"([^"]+)"/;
+        const lines = content.split('\n');
+        let inMultiImport = false;
+        for (const line of lines) {
+            if (inMultiImport) {
+                if (line.includes(')')) {
+                    inMultiImport = false;
+                    continue;
+                }
+                const match = line.match(multiImportLine);
+                if (match)
+                    imports.push({
+                        module: match[1],
+                        namedImports: [],
+                        isTypeOnly: false,
+                    });
+            }
+            else {
+                const single = line.match(singleImport);
+                if (single) {
+                    imports.push({
+                        module: single[1],
+                        namedImports: [],
+                        isTypeOnly: false,
+                    });
+                }
+                else if (multiImportStart.test(line)) {
+                    inMultiImport = true;
+                }
+            }
+        }
+        return imports;
+    }
+    parseParams(paramsStr) {
+        if (!paramsStr.trim())
+            return [];
+        return splitParamsGenericAware(paramsStr).map((p) => {
+            const parts = p.trim().split(/\s+/);
+            const name = parts[0] || '';
+            const type = parts.slice(1).join(' ') || undefined;
+            return { name, type, isOptional: false, defaultValue: undefined };
+        });
+    }
+}
+// --- Rust Parser ---
+class RustParser {
+    language = 'rust';
+    supportedExtensions = ['.rs'];
+    async parseFile(content, filePath) {
+        return {
+            functions: this.extractFunctions(content),
+            classes: this.extractStructsAndImpls(content),
+            imports: this.extractImports(content),
+            language: 'rust',
+            filePath,
+        };
+    }
+    extractFunctions(content) {
+        const functions = [];
+        const rawLines = content.split('\n');
+        const { logicalLines, lineMap } = joinMultiLineSignatures(content);
+        // Updated regex: use [^{]* instead of [^>]+ for generics to handle lifetimes
+        const fnRegex = /^\s*(pub(?:\(crate\))? +)?(async +)?fn\s+(\w+)(?:<[^{]*>)?\s*\(([^)]*)\)(?:\s*->\s*([^{\s]+(?:\s+[^{\s]+)*))?\s*(?:where\s+[^{\s]+(?:\s+[^{\s]+)*)?\s*\{?/;
+        // Track brace depth to filter nested closures.
+        // Rust functions at depth 0 (top-level) and depth 1 (inside impl/mod blocks) are valid.
+        let braceDepth = 0;
+        for (let i = 0; i < logicalLines.length; i++) {
+            const line = logicalLines[i];
+            const match = line.match(fnRegex);
+            if (match && braceDepth <= 1) {
+                const isPub = !!match[1];
+                const isAsync = !!match[2];
+                const name = match[3];
+                const paramsStr = match[4];
+                const returnType = match[5]?.trim();
+                const origLine = lineMap[i] - 1;
+                const decorators = collectDecorators(rawLines, origLine, '#[');
+                functions.push({
+                    name,
+                    parameters: this.parseParams(paramsStr),
+                    returnType,
+                    isAsync,
+                    isPublic: isPub,
+                    complexity: 1,
+                    decorators,
+                    genericParams: [],
+                    startLine: lineMap[i],
+                    endLine: lineMap[i],
+                });
+            }
+            // Update brace depth after checking for function match
+            braceDepth += countBraceDelta(line);
+            if (braceDepth < 0)
+                braceDepth = 0;
+        }
+        return functions;
+    }
+    extractStructsAndImpls(content) {
+        const structs = [];
+        const rawLines = content.split('\n');
+        const { logicalLines, lineMap } = joinMultiLineSignatures(content);
+        // Updated: use [^{]* for generics with lifetimes
+        const structRegex = /^\s*(pub(?:\(crate\))?\s+)?struct\s+(\w+)(?:<[^{]*>)?\s*/;
+        for (let i = 0; i < logicalLines.length; i++) {
+            const match = logicalLines[i].match(structRegex);
+            if (match) {
+                const origLine = lineMap[i] - 1;
+                const decorators = collectDecorators(rawLines, origLine, '#[');
+                structs.push({
+                    name: match[2],
+                    methods: [],
+                    properties: [],
+                    isPublic: !!match[1],
+                    implements: [],
+                    extends: undefined,
+                    decorators,
+                    startLine: lineMap[i],
+                    endLine: lineMap[i],
+                });
+            }
+        }
+        return structs;
+    }
+    extractImports(content) {
+        const imports = [];
+        const lines = content.split('\n');
+        for (const line of lines) {
+            if (line.length > MAX_REGEX_LINE_LEN)
+                continue;
+            const match = line.match(/^\s*use +(\S[^;]*);/);
+            if (match) {
+                imports.push({
+                    module: match[1].trim(),
+                    namedImports: [],
+                    isTypeOnly: false,
+                });
+            }
+        }
+        return imports;
+    }
+    parseParams(paramsStr) {
+        if (!paramsStr.trim())
+            return [];
+        return splitParamsGenericAware(paramsStr)
+            .map((p) => p.trim())
+            .filter((p) => p && p !== '&self' && p !== '&mut self' && p !== 'self')
+            .map((p) => {
+            const colonIdx = p.indexOf(':');
+            if (colonIdx === -1)
+                return {
+                    name: p,
+                    type: undefined,
+                    isOptional: false,
+                    defaultValue: undefined,
+                };
+            const name = p.substring(0, colonIdx).trim();
+            const type = p.substring(colonIdx + 1).trim();
+            return { name, type, isOptional: false, defaultValue: undefined };
+        });
+    }
+}
+// --- Swift Parser ---
+class SwiftParser {
+    language = 'swift';
+    supportedExtensions = ['.swift'];
+    async parseFile(content, filePath) {
+        return {
+            functions: this.extractFunctions(content),
+            classes: this.extractClasses(content),
+            imports: this.extractImports(content),
+            language: 'swift',
+            filePath,
+        };
+    }
+    extractFunctions(content) {
+        const functions = [];
+        const { logicalLines, lineMap } = joinMultiLineSignatures(content);
+        const funcRegex = /^\s*(public|private|internal|open)?\s*(static)?\s*func\s+(\w+)(?:<[^{]*>)?\s*\(([^)]*)\)(?:\s*(async))?\s*(?:throws +)?(?:->\s*([^{\s]+(?:\s+[^{\s]+)*))?\s*\{?/;
+        for (let i = 0; i < logicalLines.length; i++) {
+            if (logicalLines[i].length > MAX_REGEX_LINE_LEN)
+                continue;
+            const match = logicalLines[i].match(funcRegex);
+            if (match) {
+                const visibility = match[1] || 'internal';
+                const isAsync = !!match[5];
+                const name = match[3];
+                const paramsStr = match[4];
+                const returnType = match[6]?.trim();
+                functions.push({
+                    name,
+                    parameters: this.parseParams(paramsStr),
+                    returnType,
+                    isAsync,
+                    isPublic: visibility === 'public' || visibility === 'open',
+                    complexity: 1,
+                    decorators: [],
+                    genericParams: [],
+                    startLine: lineMap[i],
+                    endLine: lineMap[i],
+                });
+            }
+        }
+        return functions;
+    }
+    extractClasses(content) {
+        const classes = [];
+        const { logicalLines, lineMap } = joinMultiLineSignatures(content);
+        const classRegex = /^\s*(public|private|internal|open)?\s*(class|struct|protocol)\s+(\w+)(?:<[^{]*>)?(?:\s*:\s*([^{\s]+(?:\s+[^{\s]+)*))?\s*\{?/;
+        for (let i = 0; i < logicalLines.length; i++) {
+            if (logicalLines[i].length > MAX_REGEX_LINE_LEN)
+                continue;
+            const match = logicalLines[i].match(classRegex);
+            if (match) {
+                const name = match[3];
+                const conformances = match[4]
+                    ? match[4].split(',').map((s) => s.trim())
+                    : [];
+                classes.push({
+                    name,
+                    methods: [],
+                    properties: [],
+                    isPublic: (match[1] || 'internal') === 'public' || match[1] === 'open',
+                    implements: conformances.slice(1),
+                    extends: conformances[0] || undefined,
+                    decorators: [],
+                    startLine: lineMap[i],
+                    endLine: lineMap[i],
+                });
+            }
+        }
+        return classes;
+    }
+    extractImports(content) {
+        const imports = [];
+        const lines = content.split('\n');
+        for (const line of lines) {
+            const match = line.match(/^\s*import\s+(\w+)/);
+            if (match)
+                imports.push({
+                    module: match[1],
+                    namedImports: [],
+                    isTypeOnly: false,
+                });
+        }
+        return imports;
+    }
+    parseParams(paramsStr) {
+        if (!paramsStr.trim())
+            return [];
+        return splitParamsGenericAware(paramsStr).map((p) => {
+            const parts = p.trim().split(':');
+            const labelAndName = parts[0]?.trim().split(/\s+/) || [];
+            const name = labelAndName[labelAndName.length - 1] || '';
+            const type = parts.slice(1).join(':').trim();
+            return {
+                name,
+                type: type || undefined,
+                isOptional: type?.endsWith('?') || false,
+                defaultValue: undefined,
+            };
+        });
+    }
+}
+// --- Kotlin Parser ---
+class KotlinParser {
+    language = 'kotlin';
+    supportedExtensions = ['.kt', '.kts'];
+    async parseFile(content, filePath) {
+        return {
+            functions: this.extractFunctions(content),
+            classes: this.extractClasses(content),
+            imports: this.extractImports(content),
+            language: 'kotlin',
+            filePath,
+        };
+    }
+    extractFunctions(content) {
+        const functions = [];
+        const rawLines = content.split('\n');
+        const { logicalLines, lineMap } = joinMultiLineSignatures(content);
+        const funcRegex = /^\s*(public|private|protected|internal)?\s*(suspend)?\s*fun\s+(?:<[^{]*> +)?(\w+)\s*\(([^)]*)\)(?:\s*:\s*([^{=\s]+(?:\s+[^{=\s]+)*))?\s*/;
+        // Track brace depth to filter nested lambdas that look like functions.
+        // Kotlin functions at depth 0 (top-level) and depth 1 (inside class bodies) are valid.
+        let braceDepth = 0;
+        for (let i = 0; i < logicalLines.length; i++) {
+            const line = logicalLines[i];
+            if (line.length > MAX_REGEX_LINE_LEN) {
+                braceDepth += (line.match(/\{/g)?.length ?? 0) - (line.match(/\}/g)?.length ?? 0);
+                continue;
+            }
+            const match = line.match(funcRegex);
+            if (match && braceDepth <= 1) {
+                const visibility = match[1] || 'public';
+                const isSuspend = !!match[2];
+                const name = match[3];
+                const paramsStr = match[4];
+                let returnType = match[5]?.trim();
+                // Extract full generic return type
+                if (returnType && returnType.includes('<')) {
+                    const angleIdx = returnType.indexOf('<');
+                    const inner = matchBalancedBrackets(returnType, angleIdx);
+                    if (inner !== undefined) {
+                        returnType = returnType.substring(0, angleIdx) + '<' + inner + '>';
+                    }
+                }
+                const origLine = lineMap[i] - 1;
+                const decorators = collectDecorators(rawLines, origLine, '@');
+                functions.push({
+                    name,
+                    parameters: this.parseParams(paramsStr),
+                    returnType,
+                    isAsync: isSuspend,
+                    isPublic: visibility === 'public',
+                    complexity: 1,
+                    decorators,
+                    genericParams: [],
+                    startLine: lineMap[i],
+                    endLine: lineMap[i],
+                });
+            }
+            // Update brace depth after checking for function match
+            braceDepth += countBraceDelta(line);
+            if (braceDepth < 0)
+                braceDepth = 0;
+        }
+        return functions;
+    }
+    extractClasses(content) {
+        const classes = [];
+        const rawLines = content.split('\n');
+        const { logicalLines, lineMap } = joinMultiLineSignatures(content);
+        const classRegex = /^\s*(public|private|protected|internal)?\s*(data|sealed|abstract|open)?\s*class\s+(\w+)(?:<[^{]*>)?(?:\s*(?:\([^)]*\))?\s*:\s*([^{\s]+(?:\s+[^{\s]+)*))?\s*\{?/;
+        for (let i = 0; i < logicalLines.length; i++) {
+            if (logicalLines[i].length > MAX_REGEX_LINE_LEN)
+                continue;
+            const match = logicalLines[i].match(classRegex);
+            if (match) {
+                const name = match[3];
+                const bases = match[4]
+                    ? match[4]
+                        .split(',')
+                        .map((s) => s.trim().split('(')[0].trim())
+                    : [];
+                const origLine = lineMap[i] - 1;
+                const decorators = collectDecorators(rawLines, origLine, '@');
+                classes.push({
+                    name,
+                    methods: [],
+                    properties: [],
+                    isPublic: (match[1] || 'public') === 'public',
+                    implements: bases.slice(1),
+                    extends: bases[0] || undefined,
+                    decorators,
+                    startLine: lineMap[i],
+                    endLine: lineMap[i],
+                });
+            }
+        }
+        return classes;
+    }
+    extractImports(content) {
+        const imports = [];
+        const lines = content.split('\n');
+        for (const line of lines) {
+            const match = line.match(/^\s*import\s+([^\s]+)/);
+            if (match)
+                imports.push({
+                    module: match[1],
+                    namedImports: [],
+                    isTypeOnly: false,
+                });
+        }
+        return imports;
+    }
+    parseParams(paramsStr) {
+        if (!paramsStr.trim())
+            return [];
+        return splitParamsGenericAware(paramsStr).map((p) => {
+            // Split on first colon only, since types can contain colons in rare cases
+            const colonIdx = p.indexOf(':');
+            if (colonIdx === -1) {
+                const cleaned = p.replace('val ', '').replace('var ', '').trim();
+                return { name: cleaned, type: undefined, isOptional: false, defaultValue: undefined };
+            }
+            const nameWithVal = p.substring(0, colonIdx)
+                .replace('val ', '')
+                .replace('var ', '')
+                .trim();
+            const typeAndDefault = p.substring(colonIdx + 1).trim();
+            const eqIdx = typeAndDefault.indexOf('=');
+            const type = eqIdx >= 0 ? typeAndDefault.substring(0, eqIdx).trim() : typeAndDefault;
+            const hasDefault = eqIdx >= 0;
+            return {
+                name: nameWithVal,
+                type,
+                isOptional: hasDefault || type?.endsWith('?') || false,
+                defaultValue: undefined,
+            };
+        });
+    }
+}
+// --- Dart Parser ---
+class DartParser {
+    language = 'dart';
+    supportedExtensions = ['.dart'];
+    async parseFile(content, filePath) {
+        return {
+            functions: this.extractFunctions(content),
+            classes: this.extractClasses(content),
+            imports: this.extractImports(content),
+            language: 'dart',
+            filePath,
+        };
+    }
+    extractFunctions(content) {
+        const functions = [];
+        const { logicalLines, lineMap } = joinMultiLineSignatures(content);
+        const funcRegex = /^\s*([\w<>,.?]+(?:\s+[\w<>,.?]+)*)\s+(\w+)\s*\(([^)]*)\)(?:\s*(async))?\s*\{/;
+        for (let i = 0; i < logicalLines.length; i++) {
+            const line = logicalLines[i];
+            if (line.includes(' class '))
+                continue;
+            if (line.length > MAX_REGEX_LINE_LEN)
+                continue;
+            const match = line.match(funcRegex);
+            if (match) {
+                const returnTypeRaw = match[1].trim();
+                let returnType = returnTypeRaw;
+                // Extract full generic return type
+                if (returnTypeRaw && returnTypeRaw.includes('<')) {
+                    const angleIdx = returnTypeRaw.indexOf('<');
+                    const inner = matchBalancedBrackets(returnTypeRaw, angleIdx);
+                    if (inner !== undefined) {
+                        returnType = returnTypeRaw.substring(0, angleIdx) + '<' + inner + '>';
+                    }
+                }
+                const name = match[2];
+                const paramsStr = match[3];
+                const isAsync = !!match[4] || returnType.includes('Future');
+                functions.push({
+                    name,
+                    parameters: this.parseParams(paramsStr),
+                    returnType: returnType === 'void' ? undefined : returnType,
+                    isAsync,
+                    isPublic: !name.startsWith('_'),
+                    complexity: 1,
+                    decorators: [],
+                    genericParams: [],
+                    startLine: lineMap[i],
+                    endLine: lineMap[i],
+                });
+            }
+        }
+        return functions;
+    }
+    extractClasses(content) {
+        const classes = [];
+        const { logicalLines, lineMap } = joinMultiLineSignatures(content);
+        const classRegex = /^\s*(abstract +)?class\s+(\w+)(?:<[^{]*>)?(?:\s+extends\s+(\w+))?(?:\s+(?:with|implements)\s+([^{\s]+(?:\s+[^{\s]+)*))?\s*\{/;
+        for (let i = 0; i < logicalLines.length; i++) {
+            if (logicalLines[i].length > MAX_REGEX_LINE_LEN)
+                continue;
+            const match = logicalLines[i].match(classRegex);
+            if (match) {
+                const name = match[2];
+                const extendsName = match[3];
+                const mixinsAndInterfaces = match[4]
+                    ? match[4].split(',').map((s) => s.trim())
+                    : [];
+                classes.push({
+                    name,
+                    methods: [],
+                    properties: [],
+                    isPublic: !name.startsWith('_'),
+                    implements: mixinsAndInterfaces,
+                    extends: extendsName,
+                    decorators: [],
+                    startLine: lineMap[i],
+                    endLine: lineMap[i],
+                });
+            }
+        }
+        return classes;
+    }
+    extractImports(content) {
+        const imports = [];
+        const lines = content.split('\n');
+        for (const line of lines) {
+            const match = line.match(/^\s*import\s+'([^']+)'/);
+            if (match)
+                imports.push({
+                    module: match[1],
+                    namedImports: [],
+                    isTypeOnly: false,
+                });
+        }
+        return imports;
+    }
+    parseParams(paramsStr) {
+        if (!paramsStr.trim())
+            return [];
+        // Remove curly/square brackets for named/optional params
+        const clean = paramsStr.replace(/[{}[\]]/g, '');
+        return splitParamsGenericAware(clean)
+            .map((p) => p.trim())
+            .filter((p) => p)
+            .map((p) => {
+            const parts = p
+                .replace('required ', '')
+                .trim()
+                .split(/\s+/);
+            const name = parts[parts.length - 1] || '';
+            const type = parts.slice(0, -1).join(' ') || undefined;
+            return {
+                name,
+                type,
+                isOptional: p.includes('?') || p.includes('{'),
+                defaultValue: undefined,
+            };
+        });
+    }
+}
+// ============================================================================
+// Parser Registry
+// ============================================================================
+/**
+ * Registry of all language parsers.
+ * Provides regex-based parsing for Python, Java, C#, Go, Rust, Swift, Kotlin, Dart.
+ */
+export class TreeSitterParserRegistry {
+    parsers = new Map();
+    constructor() {
+        this.register(new PythonParser());
+        this.register(new JavaParser());
+        this.register(new CSharpParser());
+        this.register(new GoParser());
+        this.register(new RustParser());
+        this.register(new SwiftParser());
+        this.register(new KotlinParser());
+        this.register(new DartParser());
+    }
+    register(parser) {
+        this.parsers.set(parser.language, parser);
+    }
+    getParser(language) {
+        return this.parsers.get(language);
+    }
+    async parseFile(content, filePath, language) {
+        const parser = this.parsers.get(language);
+        if (!parser)
+            return undefined;
+        return parser.parseFile(content, filePath);
+    }
+    getSupportedLanguages() {
+        return Array.from(this.parsers.keys());
+    }
+    supportsLanguage(language) {
+        return this.parsers.has(language);
+    }
+}
+export const treeSitterRegistry = new TreeSitterParserRegistry();
+// Export individual parsers for direct use
+export { PythonParser, JavaParser, CSharpParser, GoParser, RustParser, SwiftParser, KotlinParser, DartParser, };
+//# sourceMappingURL=multi-language-parser.js.map