lance-context 0.1.0 → 1.0.1

package/dist/search/ast-chunker.js

@@ -0,0 +1,261 @@
+import * as ts from 'typescript';
+import * as fs from 'fs/promises';
+import * as path from 'path';
+const MAX_CHUNK_LINES = 150;
+const MIN_CHUNK_LINES = 10;
+/**
+ * AST-aware code chunker for TypeScript and JavaScript files.
+ * Chunks code by logical units (functions, classes, etc.) rather than arbitrary line counts.
+ */
+export class ASTChunker {
+    /**
+     * Check if a file can be parsed with AST chunking
+     */
+    static canParse(filepath) {
+        const ext = path.extname(filepath).toLowerCase();
+        return ['.ts', '.tsx', '.js', '.jsx', '.mts', '.cts', '.mjs', '.cjs'].includes(ext);
+    }
+    /**
+     * Parse a file and return AST-aware chunks
+     */
+    async chunkFile(filepath) {
+        const content = await fs.readFile(filepath, 'utf-8');
+        const ext = path.extname(filepath).toLowerCase();
+        // Determine script kind
+        let scriptKind = ts.ScriptKind.TS;
+        if (ext === '.tsx' || ext === '.jsx') {
+            scriptKind = ts.ScriptKind.TSX;
+        }
+        else if (ext === '.js' || ext === '.mjs' || ext === '.cjs') {
+            scriptKind = ts.ScriptKind.JS;
+        }
+        // Parse the file
+        const sourceFile = ts.createSourceFile(filepath, content, ts.ScriptTarget.Latest, true, scriptKind);
+        const chunks = [];
+        const lines = content.split('\n');
+        // Collect imports as a single chunk
+        const imports = [];
+        // Process top-level statements
+        ts.forEachChild(sourceFile, (node) => {
+            if (ts.isImportDeclaration(node) || ts.isExportDeclaration(node)) {
+                imports.push(node);
+                return;
+            }
+            const nodeChunks = this.extractChunks(node, sourceFile, lines);
+            chunks.push(...nodeChunks);
+        });
+        // Add imports as the first chunk if present
+        if (imports.length > 0) {
+            const firstImport = imports[0];
+            const lastImport = imports[imports.length - 1];
+            const startLine = sourceFile.getLineAndCharacterOfPosition(firstImport.getStart(sourceFile)).line;
+            const endLine = sourceFile.getLineAndCharacterOfPosition(lastImport.getEnd()).line;
+            chunks.unshift({
+                content: lines.slice(startLine, endLine + 1).join('\n'),
+                startLine: startLine + 1,
+                endLine: endLine + 1,
+                type: 'import',
+                name: 'imports',
+            });
+        }
+        // Sort chunks by start line
+        chunks.sort((a, b) => a.startLine - b.startLine);
+        // Split any chunks that are too large
+        const finalChunks = [];
+        for (const chunk of chunks) {
+            if (chunk.endLine - chunk.startLine + 1 > MAX_CHUNK_LINES) {
+                const splitChunks = this.splitLargeChunk(chunk, lines);
+                finalChunks.push(...splitChunks);
+            }
+            else {
+                finalChunks.push(chunk);
+            }
+        }
+        return finalChunks;
+    }
+    /**
+     * Extract chunks from a node
+     */
+    extractChunks(node, sourceFile, lines) {
+        const chunks = [];
+        const endLine = sourceFile.getLineAndCharacterOfPosition(node.getEnd()).line;
+        // Get leading comments/decorators
+        const fullStart = node.getFullStart();
+        const actualStartLine = sourceFile.getLineAndCharacterOfPosition(fullStart).line;
+        if (ts.isFunctionDeclaration(node)) {
+            chunks.push({
+                content: lines.slice(actualStartLine, endLine + 1).join('\n'),
+                startLine: actualStartLine + 1,
+                endLine: endLine + 1,
+                type: 'function',
+                name: node.name?.getText(sourceFile),
+            });
+        }
+        else if (ts.isClassDeclaration(node)) {
+            // For classes, we have options:
+            // 1. Keep the whole class together if it's small enough
+            // 2. Split into class header + individual methods if too large
+            const classLines = endLine - actualStartLine + 1;
+            if (classLines <= MAX_CHUNK_LINES) {
+                // Keep class together
+                chunks.push({
+                    content: lines.slice(actualStartLine, endLine + 1).join('\n'),
+                    startLine: actualStartLine + 1,
+                    endLine: endLine + 1,
+                    type: 'class',
+                    name: node.name?.getText(sourceFile),
+                });
+            }
+            else {
+                // Split class into methods
+                const className = node.name?.getText(sourceFile) || 'AnonymousClass';
+                // Get class header (decorators + class declaration up to first member)
+                const members = node.members;
+                if (members.length > 0) {
+                    const firstMemberStart = sourceFile.getLineAndCharacterOfPosition(members[0].getFullStart()).line;
+                    // Class header chunk
+                    if (firstMemberStart > actualStartLine) {
+                        chunks.push({
+                            content: lines.slice(actualStartLine, firstMemberStart).join('\n'),
+                            startLine: actualStartLine + 1,
+                            endLine: firstMemberStart,
+                            type: 'class',
+                            name: `${className} (header)`,
+                        });
+                    }
+                    // Individual method chunks
+                    for (const member of members) {
+                        const memberChunks = this.extractMemberChunk(member, sourceFile, lines, className);
+                        chunks.push(...memberChunks);
+                    }
+                }
+                else {
+                    // Empty class
+                    chunks.push({
+                        content: lines.slice(actualStartLine, endLine + 1).join('\n'),
+                        startLine: actualStartLine + 1,
+                        endLine: endLine + 1,
+                        type: 'class',
+                        name: className,
+                    });
+                }
+            }
+        }
+        else if (ts.isInterfaceDeclaration(node)) {
+            chunks.push({
+                content: lines.slice(actualStartLine, endLine + 1).join('\n'),
+                startLine: actualStartLine + 1,
+                endLine: endLine + 1,
+                type: 'interface',
+                name: node.name.getText(sourceFile),
+            });
+        }
+        else if (ts.isTypeAliasDeclaration(node)) {
+            chunks.push({
+                content: lines.slice(actualStartLine, endLine + 1).join('\n'),
+                startLine: actualStartLine + 1,
+                endLine: endLine + 1,
+                type: 'type',
+                name: node.name.getText(sourceFile),
+            });
+        }
+        else if (ts.isVariableStatement(node)) {
+            const declarations = node.declarationList.declarations;
+            const names = declarations
+                .map((d) => (ts.isIdentifier(d.name) ? d.name.getText(sourceFile) : undefined))
+                .filter(Boolean)
+                .join(', ');
+            chunks.push({
+                content: lines.slice(actualStartLine, endLine + 1).join('\n'),
+                startLine: actualStartLine + 1,
+                endLine: endLine + 1,
+                type: 'variable',
+                name: names || undefined,
+            });
+        }
+        else if (ts.isExpressionStatement(node) || ts.isExportAssignment(node)) {
+            chunks.push({
+                content: lines.slice(actualStartLine, endLine + 1).join('\n'),
+                startLine: actualStartLine + 1,
+                endLine: endLine + 1,
+                type: 'other',
+            });
+        }
+        else if (ts.isEnumDeclaration(node)) {
+            chunks.push({
+                content: lines.slice(actualStartLine, endLine + 1).join('\n'),
+                startLine: actualStartLine + 1,
+                endLine: endLine + 1,
+                type: 'other',
+                name: node.name.getText(sourceFile),
+            });
+        }
+        return chunks;
+    }
+    /**
+     * Extract a chunk for a class member
+     */
+    extractMemberChunk(member, sourceFile, lines, className) {
+        const fullStart = member.getFullStart();
+        const startLine = sourceFile.getLineAndCharacterOfPosition(fullStart).line;
+        const endLine = sourceFile.getLineAndCharacterOfPosition(member.getEnd()).line;
+        let name;
+        let type = 'other';
+        if (ts.isMethodDeclaration(member) ||
+            ts.isGetAccessorDeclaration(member) ||
+            ts.isSetAccessorDeclaration(member)) {
+            type = 'method';
+            name = `${className}.${member.name?.getText(sourceFile) || 'anonymous'}`;
+        }
+        else if (ts.isConstructorDeclaration(member)) {
+            type = 'method';
+            name = `${className}.constructor`;
+        }
+        else if (ts.isPropertyDeclaration(member)) {
+            type = 'variable';
+            name = `${className}.${member.name?.getText(sourceFile) || 'property'}`;
+        }
+        return [
+            {
+                content: lines.slice(startLine, endLine + 1).join('\n'),
+                startLine: startLine + 1,
+                endLine: endLine + 1,
+                type,
+                name,
+            },
+        ];
+    }
+    /**
+     * Split a large chunk into smaller pieces
+     */
+    splitLargeChunk(chunk, _allLines) {
+        const chunkLines = chunk.content.split('\n');
+        const totalLines = chunkLines.length;
+        const chunks = [];
+        // Split into roughly equal parts, each under MAX_CHUNK_LINES
+        const numParts = Math.ceil(totalLines / MAX_CHUNK_LINES);
+        const linesPerPart = Math.ceil(totalLines / numParts);
+        for (let i = 0; i < numParts; i++) {
+            const startIdx = i * linesPerPart;
+            const endIdx = Math.min((i + 1) * linesPerPart, totalLines);
+            const partLines = chunkLines.slice(startIdx, endIdx);
+            if (partLines.length < MIN_CHUNK_LINES && chunks.length > 0) {
+                // Merge with previous chunk if too small
+                const lastChunk = chunks[chunks.length - 1];
+                lastChunk.content += '\n' + partLines.join('\n');
+                lastChunk.endLine = chunk.startLine + endIdx - 1;
+            }
+            else {
+                chunks.push({
+                    content: partLines.join('\n'),
+                    startLine: chunk.startLine + startIdx,
+                    endLine: chunk.startLine + endIdx - 1,
+                    type: chunk.type,
+                    name: chunk.name ? `${chunk.name} (part ${i + 1})` : undefined,
+                });
+            }
+        }
+        return chunks;
+    }
+}
+//# sourceMappingURL=ast-chunker.js.map

package/dist/search/ast-chunker.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"ast-chunker.js","sourceRoot":"","sources":["../../src/search/ast-chunker.ts"],"names":[],"mappings":"AAAA,OAAO,KAAK,EAAE,MAAM,YAAY,CAAC;AACjC,OAAO,KAAK,EAAE,MAAM,aAAa,CAAC;AAClC,OAAO,KAAK,IAAI,MAAM,MAAM,CAAC;AAU7B,MAAM,eAAe,GAAG,GAAG,CAAC;AAC5B,MAAM,eAAe,GAAG,EAAE,CAAC;AAE3B;;;GAGG;AACH,MAAM,OAAO,UAAU;IACrB;;OAEG;IACH,MAAM,CAAC,QAAQ,CAAC,QAAgB;QAC9B,MAAM,GAAG,GAAG,IAAI,CAAC,OAAO,CAAC,QAAQ,CAAC,CAAC,WAAW,EAAE,CAAC;QACjD,OAAO,CAAC,KAAK,EAAE,MAAM,EAAE,KAAK,EAAE,MAAM,EAAE,MAAM,EAAE,MAAM,EAAE,MAAM,EAAE,MAAM,CAAC,CAAC,QAAQ,CAAC,GAAG,CAAC,CAAC;IACtF,CAAC;IAED;;OAEG;IACH,KAAK,CAAC,SAAS,CAAC,QAAgB;QAC9B,MAAM,OAAO,GAAG,MAAM,EAAE,CAAC,QAAQ,CAAC,QAAQ,EAAE,OAAO,CAAC,CAAC;QACrD,MAAM,GAAG,GAAG,IAAI,CAAC,OAAO,CAAC,QAAQ,CAAC,CAAC,WAAW,EAAE,CAAC;QAEjD,wBAAwB;QACxB,IAAI,UAAU,GAAG,EAAE,CAAC,UAAU,CAAC,EAAE,CAAC;QAClC,IAAI,GAAG,KAAK,MAAM,IAAI,GAAG,KAAK,MAAM,EAAE,CAAC;YACrC,UAAU,GAAG,EAAE,CAAC,UAAU,CAAC,GAAG,CAAC;QACjC,CAAC;aAAM,IAAI,GAAG,KAAK,KAAK,IAAI,GAAG,KAAK,MAAM,IAAI,GAAG,KAAK,MAAM,EAAE,CAAC;YAC7D,UAAU,GAAG,EAAE,CAAC,UAAU,CAAC,EAAE,CAAC;QAChC,CAAC;QAED,iBAAiB;QACjB,MAAM,UAAU,GAAG,EAAE,CAAC,gBAAgB,CACpC,QAAQ,EACR,OAAO,EACP,EAAE,CAAC,YAAY,CAAC,MAAM,EACtB,IAAI,EACJ,UAAU,CACX,CAAC;QAEF,MAAM,MAAM,GAAe,EAAE,CAAC;QAC9B,MAAM,KAAK,GAAG,OAAO,CAAC,KAAK,CAAC,IAAI,CAAC,CAAC;QAElC,oCAAoC;QACpC,MAAM,OAAO,GAAc,EAAE,CAAC;QAE9B,+BAA+B;QAC/B,EAAE,CAAC,YAAY,CAAC,UAAU,EAAE,CAAC,IAAI,EAAE,EAAE;YACnC,IAAI,EAAE,CAAC,mBAAmB,CAAC,IAAI,CAAC,IAAI,EAAE,CAAC,mBAAmB,CAAC,IAAI,CAAC,EAAE,CAAC;gBACjE,OAAO,CAAC,IAAI,CAAC,IAAI,CAAC,CAAC;gBACnB,OAAO;YACT,CAAC;YAED,MAAM,UAAU,GAAG,IAAI,CAAC,aAAa,CAAC,IAAI,EAAE,UAAU,EAAE,KAAK,CAAC,CAAC;YAC/D,MAAM,CAAC,IAAI,CAAC,GAAG,UAAU,CAAC,CAAC;QAC7B,CAAC,CAAC,CAAC;QAEH,4CAA4C;QAC5C,IAAI,OAAO,CAAC,MAAM,GAAG,CAAC,EAAE,CAAC;YACvB,MAAM,WAAW,GAAG,OAAO,CAAC,CAAC,CAAC,CAAC;YAC/B,MAAM,UAAU,GAAG,OAAO,CAAC,OAAO,CAAC,MAAM,GAAG,CAAC,CAAC,CAAC;YAC/C,MAAM,SAAS,GAAG,UAAU,CAAC,6BAA6B,CACxD,WAAW,CAAC,QAAQ,CAAC,UAAU,CAAC,CACjC,CAAC,IAAI,CAAC;YACP,MAAM,OAAO,GAAG,UAAU,CAAC,6BAA6B,CAAC,UAAU,CAAC,MAAM,EAAE,CAAC,CAAC,IAAI,CAAC;YAEnF,MAAM,CAAC,OAAO,CAAC;gBACb,OAAO,EAAE,KAAK,CAAC,KAAK,CAAC,SAAS,EAAE,OAAO,GAAG,CAAC,CAAC,CAAC,IAAI,CAAC,IAAI,CAAC;gBACvD,SAAS,EAAE,SAAS,GAAG,CAAC;gBACxB,OAAO,EAAE,OAAO,GAAG,CAAC;gBACpB,IAAI,EAAE,QAAQ;gBACd,IAAI,EAAE,SAAS;aAChB,CAAC,CAAC;QACL,CAAC;QAED,4BAA4B;QAC5B,MAAM,CAAC,IAAI,CAAC,CAAC,CAAC,EAAE,CAAC,EAAE,EAAE,CAAC,CAAC,CAAC,SAAS,GAAG,CAAC,CAAC,SAAS,CAAC,CAAC;QAEjD,sCAAsC;QACtC,MAAM,WAAW,GAAe,EAAE,CAAC;QACnC,KAAK,MAAM,KAAK,IAAI,MAAM,EAAE,CAAC;YAC3B,IAAI,KAAK,CAAC,OAAO,GAAG,KAAK,CAAC,SAAS,GAAG,CAAC,GAAG,eAAe,EAAE,CAAC;gBAC1D,MAAM,WAAW,GAAG,IAAI,CAAC,eAAe,CAAC,KAAK,EAAE,KAAK,CAAC,CAAC;gBACvD,WAAW,CAAC,IAAI,CAAC,GAAG,WAAW,CAAC,CAAC;YACnC,CAAC;iBAAM,CAAC;gBACN,WAAW,CAAC,IAAI,CAAC,KAAK,CAAC,CAAC;YAC1B,CAAC;QACH,CAAC;QAED,OAAO,WAAW,CAAC;IACrB,CAAC;IAED;;OAEG;IACK,aAAa,CAAC,IAAa,EAAE,UAAyB,EAAE,KAAe;QAC7E,MAAM,MAAM,GAAe,EAAE,CAAC;QAC9B,MAAM,OAAO,GAAG,UAAU,CAAC,6BAA6B,CAAC,IAAI,CAAC,MAAM,EAAE,CAAC,CAAC,IAAI,CAAC;QAE7E,kCAAkC;QAClC,MAAM,SAAS,GAAG,IAAI,CAAC,YAAY,EAAE,CAAC;QACtC,MAAM,eAAe,GAAG,UAAU,CAAC,6BAA6B,CAAC,SAAS,CAAC,CAAC,IAAI,CAAC;QAEjF,IAAI,EAAE,CAAC,qBAAqB,CAAC,IAAI,CAAC,EAAE,CAAC;YACnC,MAAM,CAAC,IAAI,CAAC;gBACV,OAAO,EAAE,KAAK,CAAC,KAAK,CAAC,eAAe,EAAE,OAAO,GAAG,CAAC,CAAC,CAAC,IAAI,CAAC,IAAI,CAAC;gBAC7D,SAAS,EAAE,eAAe,GAAG,CAAC;gBAC9B,OAAO,EAAE,OAAO,GAAG,CAAC;gBACpB,IAAI,EAAE,UAAU;gBAChB,IAAI,EAAE,IAAI,CAAC,IAAI,EAAE,OAAO,CAAC,UAAU,CAAC;aACrC,CAAC,CAAC;QACL,CAAC;aAAM,IAAI,EAAE,CAAC,kBAAkB,CAAC,IAAI,CAAC,EAAE,CAAC;YACvC,gCAAgC;YAChC,wDAAwD;YACxD,+DAA+D;YAC/D,MAAM,UAAU,GAAG,OAAO,GAAG,eAAe,GAAG,CAAC,CAAC;YAEjD,IAAI,UAAU,IAAI,eAAe,EAAE,CAAC;gBAClC,sBAAsB;gBACtB,MAAM,CAAC,IAAI,CAAC;oBACV,OAAO,EAAE,KAAK,CAAC,KAAK,CAAC,eAAe,EAAE,OAAO,GAAG,CAAC,CAAC,CAAC,IAAI,CAAC,IAAI,CAAC;oBAC7D,SAAS,EAAE,eAAe,GAAG,CAAC;oBAC9B,OAAO,EAAE,OAAO,GAAG,CAAC;oBACpB,IAAI,EAAE,OAAO;oBACb,IAAI,EAAE,IAAI,CAAC,IAAI,EAAE,OAAO,CAAC,UAAU,CAAC;iBACrC,CAAC,CAAC;YACL,CAAC;iBAAM,CAAC;gBACN,2BAA2B;gBAC3B,MAAM,SAAS,GAAG,IAAI,CAAC,IAAI,EAAE,OAAO,CAAC,UAAU,CAAC,IAAI,gBAAgB,CAAC;gBAErE,uEAAuE;gBACvE,MAAM,OAAO,GAAG,IAAI,CAAC,OAAO,CAAC;gBAC7B,IAAI,OAAO,CAAC,MAAM,GAAG,CAAC,EAAE,CAAC;oBACvB,MAAM,gBAAgB,GAAG,UAAU,CAAC,6BAA6B,CAC/D,OAAO,CAAC,CAAC,CAAC,CAAC,YAAY,EAAE,CAC1B,CAAC,IAAI,CAAC;oBAEP,qBAAqB;oBACrB,IAAI,gBAAgB,GAAG,eAAe,EAAE,CAAC;wBACvC,MAAM,CAAC,IAAI,CAAC;4BACV,OAAO,EAAE,KAAK,CAAC,KAAK,CAAC,eAAe,EAAE,gBAAgB,CAAC,CAAC,IAAI,CAAC,IAAI,CAAC;4BAClE,SAAS,EAAE,eAAe,GAAG,CAAC;4BAC9B,OAAO,EAAE,gBAAgB;4BACzB,IAAI,EAAE,OAAO;4BACb,IAAI,EAAE,GAAG,SAAS,WAAW;yBAC9B,CAAC,CAAC;oBACL,CAAC;oBAED,2BAA2B;oBAC3B,KAAK,MAAM,MAAM,IAAI,OAAO,EAAE,CAAC;wBAC7B,MAAM,YAAY,GAAG,IAAI,CAAC,kBAAkB,CAAC,MAAM,EAAE,UAAU,EAAE,KAAK,EAAE,SAAS,CAAC,CAAC;wBACnF,MAAM,CAAC,IAAI,CAAC,GAAG,YAAY,CAAC,CAAC;oBAC/B,CAAC;gBACH,CAAC;qBAAM,CAAC;oBACN,cAAc;oBACd,MAAM,CAAC,IAAI,CAAC;wBACV,OAAO,EAAE,KAAK,CAAC,KAAK,CAAC,eAAe,EAAE,OAAO,GAAG,CAAC,CAAC,CAAC,IAAI,CAAC,IAAI,CAAC;wBAC7D,SAAS,EAAE,eAAe,GAAG,CAAC;wBAC9B,OAAO,EAAE,OAAO,GAAG,CAAC;wBACpB,IAAI,EAAE,OAAO;wBACb,IAAI,EAAE,SAAS;qBAChB,CAAC,CAAC;gBACL,CAAC;YACH,CAAC;QACH,CAAC;aAAM,IAAI,EAAE,CAAC,sBAAsB,CAAC,IAAI,CAAC,EAAE,CAAC;YAC3C,MAAM,CAAC,IAAI,CAAC;gBACV,OAAO,EAAE,KAAK,CAAC,KAAK,CAAC,eAAe,EAAE,OAAO,GAAG,CAAC,CAAC,CAAC,IAAI,CAAC,IAAI,CAAC;gBAC7D,SAAS,EAAE,eAAe,GAAG,CAAC;gBAC9B,OAAO,EAAE,OAAO,GAAG,CAAC;gBACpB,IAAI,EAAE,WAAW;gBACjB,IAAI,EAAE,IAAI,CAAC,IAAI,CAAC,OAAO,CAAC,UAAU,CAAC;aACpC,CAAC,CAAC;QACL,CAAC;aAAM,IAAI,EAAE,CAAC,sBAAsB,CAAC,IAAI,CAAC,EAAE,CAAC;YAC3C,MAAM,CAAC,IAAI,CAAC;gBACV,OAAO,EAAE,KAAK,CAAC,KAAK,CAAC,eAAe,EAAE,OAAO,GAAG,CAAC,CAAC,CAAC,IAAI,CAAC,IAAI,CAAC;gBAC7D,SAAS,EAAE,eAAe,GAAG,CAAC;gBAC9B,OAAO,EAAE,OAAO,GAAG,CAAC;gBACpB,IAAI,EAAE,MAAM;gBACZ,IAAI,EAAE,IAAI,CAAC,IAAI,CAAC,OAAO,CAAC,UAAU,CAAC;aACpC,CAAC,CAAC;QACL,CAAC;aAAM,IAAI,EAAE,CAAC,mBAAmB,CAAC,IAAI,CAAC,EAAE,CAAC;YACxC,MAAM,YAAY,GAAG,IAAI,CAAC,eAAe,CAAC,YAAY,CAAC;YACvD,MAAM,KAAK,GAAG,YAAY;iBACvB,GAAG,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,CAAC,EAAE,CAAC,YAAY,CAAC,CAAC,CAAC,IAAI,CAAC,CAAC,CAAC,CAAC,CAAC,CAAC,IAAI,CAAC,OAAO,CAAC,UAAU,CAAC,CAAC,CAAC,CAAC,SAAS,CAAC,CAAC;iBAC9E,MAAM,CAAC,OAAO,CAAC;iBACf,IAAI,CAAC,IAAI,CAAC,CAAC;YAEd,MAAM,CAAC,IAAI,CAAC;gBACV,OAAO,EAAE,KAAK,CAAC,KAAK,CAAC,eAAe,EAAE,OAAO,GAAG,CAAC,CAAC,CAAC,IAAI,CAAC,IAAI,CAAC;gBAC7D,SAAS,EAAE,eAAe,GAAG,CAAC;gBAC9B,OAAO,EAAE,OAAO,GAAG,CAAC;gBACpB,IAAI,EAAE,UAAU;gBAChB,IAAI,EAAE,KAAK,IAAI,SAAS;aACzB,CAAC,CAAC;QACL,CAAC;aAAM,IAAI,EAAE,CAAC,qBAAqB,CAAC,IAAI,CAAC,IAAI,EAAE,CAAC,kBAAkB,CAAC,IAAI,CAAC,EAAE,CAAC;YACzE,MAAM,CAAC,IAAI,CAAC;gBACV,OAAO,EAAE,KAAK,CAAC,KAAK,CAAC,eAAe,EAAE,OAAO,GAAG,CAAC,CAAC,CAAC,IAAI,CAAC,IAAI,CAAC;gBAC7D,SAAS,EAAE,eAAe,GAAG,CAAC;gBAC9B,OAAO,EAAE,OAAO,GAAG,CAAC;gBACpB,IAAI,EAAE,OAAO;aACd,CAAC,CAAC;QACL,CAAC;aAAM,IAAI,EAAE,CAAC,iBAAiB,CAAC,IAAI,CAAC,EAAE,CAAC;YACtC,MAAM,CAAC,IAAI,CAAC;gBACV,OAAO,EAAE,KAAK,CAAC,KAAK,CAAC,eAAe,EAAE,OAAO,GAAG,CAAC,CAAC,CAAC,IAAI,CAAC,IAAI,CAAC;gBAC7D,SAAS,EAAE,eAAe,GAAG,CAAC;gBAC9B,OAAO,EAAE,OAAO,GAAG,CAAC;gBACpB,IAAI,EAAE,OAAO;gBACb,IAAI,EAAE,IAAI,CAAC,IAAI,CAAC,OAAO,CAAC,UAAU,CAAC;aACpC,CAAC,CAAC;QACL,CAAC;QAED,OAAO,MAAM,CAAC;IAChB,CAAC;IAED;;OAEG;IACK,kBAAkB,CACxB,MAAuB,EACvB,UAAyB,EACzB,KAAe,EACf,SAAiB;QAEjB,MAAM,SAAS,GAAG,MAAM,CAAC,YAAY,EAAE,CAAC;QACxC,MAAM,SAAS,GAAG,UAAU,CAAC,6BAA6B,CAAC,SAAS,CAAC,CAAC,IAAI,CAAC;QAC3E,MAAM,OAAO,GAAG,UAAU,CAAC,6BAA6B,CAAC,MAAM,CAAC,MAAM,EAAE,CAAC,CAAC,IAAI,CAAC;QAE/E,IAAI,IAAwB,CAAC;QAC7B,IAAI,IAAI,GAAqB,OAAO,CAAC;QAErC,IACE,EAAE,CAAC,mBAAmB,CAAC,MAAM,CAAC;YAC9B,EAAE,CAAC,wBAAwB,CAAC,MAAM,CAAC;YACnC,EAAE,CAAC,wBAAwB,CAAC,MAAM,CAAC,EACnC,CAAC;YACD,IAAI,GAAG,QAAQ,CAAC;YAChB,IAAI,GAAG,GAAG,SAAS,IAAI,MAAM,CAAC,IAAI,EAAE,OAAO,CAAC,UAAU,CAAC,IAAI,WAAW,EAAE,CAAC;QAC3E,CAAC;aAAM,IAAI,EAAE,CAAC,wBAAwB,CAAC,MAAM,CAAC,EAAE,CAAC;YAC/C,IAAI,GAAG,QAAQ,CAAC;YAChB,IAAI,GAAG,GAAG,SAAS,cAAc,CAAC;QACpC,CAAC;aAAM,IAAI,EAAE,CAAC,qBAAqB,CAAC,MAAM,CAAC,EAAE,CAAC;YAC5C,IAAI,GAAG,UAAU,CAAC;YAClB,IAAI,GAAG,GAAG,SAAS,IAAI,MAAM,CAAC,IAAI,EAAE,OAAO,CAAC,UAAU,CAAC,IAAI,UAAU,EAAE,CAAC;QAC1E,CAAC;QAED,OAAO;YACL;gBACE,OAAO,EAAE,KAAK,CAAC,KAAK,CAAC,SAAS,EAAE,OAAO,GAAG,CAAC,CAAC,CAAC,IAAI,CAAC,IAAI,CAAC;gBACvD,SAAS,EAAE,SAAS,GAAG,CAAC;gBACxB,OAAO,EAAE,OAAO,GAAG,CAAC;gBACpB,IAAI;gBACJ,IAAI;aACL;SACF,CAAC;IACJ,CAAC;IAED;;OAEG;IACK,eAAe,CAAC,KAAe,EAAE,SAAmB;QAC1D,MAAM,UAAU,GAAG,KAAK,CAAC,OAAO,CAAC,KAAK,CAAC,IAAI,CAAC,CAAC;QAC7C,MAAM,UAAU,GAAG,UAAU,CAAC,MAAM,CAAC;QACrC,MAAM,MAAM,GAAe,EAAE,CAAC;QAE9B,6DAA6D;QAC7D,MAAM,QAAQ,GAAG,IAAI,CAAC,IAAI,CAAC,UAAU,GAAG,eAAe,CAAC,CAAC;QACzD,MAAM,YAAY,GAAG,IAAI,CAAC,IAAI,CAAC,UAAU,GAAG,QAAQ,CAAC,CAAC;QAEtD,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,QAAQ,EAAE,CAAC,EAAE,EAAE,CAAC;YAClC,MAAM,QAAQ,GAAG,CAAC,GAAG,YAAY,CAAC;YAClC,MAAM,MAAM,GAAG,IAAI,CAAC,GAAG,CAAC,CAAC,CAAC,GAAG,CAAC,CAAC,GAAG,YAAY,EAAE,UAAU,CAAC,CAAC;YAC5D,MAAM,SAAS,GAAG,UAAU,CAAC,KAAK,CAAC,QAAQ,EAAE,MAAM,CAAC,CAAC;YAErD,IAAI,SAAS,CAAC,MAAM,GAAG,eAAe,IAAI,MAAM,CAAC,MAAM,GAAG,CAAC,EAAE,CAAC;gBAC5D,yCAAyC;gBACzC,MAAM,SAAS,GAAG,MAAM,CAAC,MAAM,CAAC,MAAM,GAAG,CAAC,CAAC,CAAC;gBAC5C,SAAS,CAAC,OAAO,IAAI,IAAI,GAAG,SAAS,CAAC,IAAI,CAAC,IAAI,CAAC,CAAC;gBACjD,SAAS,CAAC,OAAO,GAAG,KAAK,CAAC,SAAS,GAAG,MAAM,GAAG,CAAC,CAAC;YACnD,CAAC;iBAAM,CAAC;gBACN,MAAM,CAAC,IAAI,CAAC;oBACV,OAAO,EAAE,SAAS,CAAC,IAAI,CAAC,IAAI,CAAC;oBAC7B,SAAS,EAAE,KAAK,CAAC,SAAS,GAAG,QAAQ;oBACrC,OAAO,EAAE,KAAK,CAAC,SAAS,GAAG,MAAM,GAAG,CAAC;oBACrC,IAAI,EAAE,KAAK,CAAC,IAAI;oBAChB,IAAI,EAAE,KAAK,CAAC,IAAI,CAAC,CAAC,CAAC,GAAG,KAAK,CAAC,IAAI,UAAU,CAAC,GAAG,CAAC,GAAG,CAAC,CAAC,CAAC,SAAS;iBAC/D,CAAC,CAAC;YACL,CAAC;QACH,CAAC;QAED,OAAO,MAAM,CAAC;IAChB,CAAC;CACF"}

package/dist/search/clustering.d.ts

@@ -0,0 +1,77 @@
+/**
+ * K-means clustering for semantic code grouping.
+ * Groups code chunks by similarity to discover concept areas in the codebase.
+ */
+/**
+ * Represents a discovered concept cluster
+ */
+export interface ConceptCluster {
+    /** Unique cluster identifier (0-indexed) */
+    id: number;
+    /** Human-readable label derived from representative chunks */
+    label: string;
+    /** Number of code chunks in this cluster */
+    size: number;
+    /** Representative chunk IDs for this cluster */
+    representativeChunks: string[];
+    /** Centroid vector for this cluster */
+    centroid: number[];
+    /** Keywords extracted from representative chunks */
+    keywords: string[];
+}
+/**
+ * Result of clustering operation
+ */
+export interface ClusteringResult {
+    /** Total number of clusters created */
+    clusterCount: number;
+    /** All discovered concept clusters */
+    clusters: ConceptCluster[];
+    /** Mapping from chunk ID to cluster ID */
+    assignments: Map<string, number>;
+}
+/**
+ * Options for clustering operation
+ */
+export interface ClusteringOptions {
+    /** Target number of clusters (default: auto-determined based on chunk count) */
+    numClusters?: number;
+    /** Maximum iterations for k-means (default: 100) */
+    maxIterations?: number;
+    /** Convergence threshold (default: 0.001) */
+    convergenceThreshold?: number;
+    /** Number of representative chunks to store per cluster (default: 3) */
+    numRepresentatives?: number;
+}
+/**
+ * Chunk data needed for clustering
+ */
+export interface ChunkForClustering {
+    id: string;
+    content: string;
+    filepath: string;
+    embedding: number[];
+    symbolName?: string;
+    symbolType?: string;
+}
+/**
+ * Calculate cosine similarity between two vectors
+ */
+export declare function cosineSimilarity(a: number[], b: number[]): number;
+/**
+ * Calculate Euclidean distance between two vectors
+ */
+export declare function euclideanDistance(a: number[], b: number[]): number;
+/**
+ * Perform k-means clustering on embeddings
+ */
+export declare function kMeansClustering(chunks: ChunkForClustering[], options?: ClusteringOptions): ClusteringResult;
+/**
+ * Find the cluster ID for a given embedding (nearest centroid)
+ */
+export declare function assignToCluster(embedding: number[], clusters: ConceptCluster[]): number;
+/**
+ * Calculate silhouette score for clustering quality assessment
+ */
+export declare function calculateSilhouetteScore(chunks: ChunkForClustering[], assignments: Map<string, number>, clusters: ConceptCluster[]): number;
+//# sourceMappingURL=clustering.d.ts.map

package/dist/search/clustering.d.ts.map

@@ -0,0 +1 @@
+{"version":3,"file":"clustering.d.ts","sourceRoot":"","sources":["../../src/search/clustering.ts"],"names":[],"mappings":"AAAA;;;GAGG;AAEH;;GAEG;AACH,MAAM,WAAW,cAAc;IAC7B,4CAA4C;IAC5C,EAAE,EAAE,MAAM,CAAC;IACX,8DAA8D;IAC9D,KAAK,EAAE,MAAM,CAAC;IACd,4CAA4C;IAC5C,IAAI,EAAE,MAAM,CAAC;IACb,gDAAgD;IAChD,oBAAoB,EAAE,MAAM,EAAE,CAAC;IAC/B,uCAAuC;IACvC,QAAQ,EAAE,MAAM,EAAE,CAAC;IACnB,oDAAoD;IACpD,QAAQ,EAAE,MAAM,EAAE,CAAC;CACpB;AAED;;GAEG;AACH,MAAM,WAAW,gBAAgB;IAC/B,uCAAuC;IACvC,YAAY,EAAE,MAAM,CAAC;IACrB,sCAAsC;IACtC,QAAQ,EAAE,cAAc,EAAE,CAAC;IAC3B,0CAA0C;IAC1C,WAAW,EAAE,GAAG,CAAC,MAAM,EAAE,MAAM,CAAC,CAAC;CAClC;AAED;;GAEG;AACH,MAAM,WAAW,iBAAiB;IAChC,gFAAgF;IAChF,WAAW,CAAC,EAAE,MAAM,CAAC;IACrB,oDAAoD;IACpD,aAAa,CAAC,EAAE,MAAM,CAAC;IACvB,6CAA6C;IAC7C,oBAAoB,CAAC,EAAE,MAAM,CAAC;IAC9B,wEAAwE;IACxE,kBAAkB,CAAC,EAAE,MAAM,CAAC;CAC7B;AAED;;GAEG;AACH,MAAM,WAAW,kBAAkB;IACjC,EAAE,EAAE,MAAM,CAAC;IACX,OAAO,EAAE,MAAM,CAAC;IAChB,QAAQ,EAAE,MAAM,CAAC;IACjB,SAAS,EAAE,MAAM,EAAE,CAAC;IACpB,UAAU,CAAC,EAAE,MAAM,CAAC;IACpB,UAAU,CAAC,EAAE,MAAM,CAAC;CACrB;AAED;;GAEG;AACH,wBAAgB,gBAAgB,CAAC,CAAC,EAAE,MAAM,EAAE,EAAE,CAAC,EAAE,MAAM,EAAE,GAAG,MAAM,CAuBjE;AAED;;GAEG;AACH,wBAAgB,iBAAiB,CAAC,CAAC,EAAE,MAAM,EAAE,EAAE,CAAC,EAAE,MAAM,EAAE,GAAG,MAAM,CAYlE;AAwED;;GAEG;AACH,wBAAgB,gBAAgB,CAC9B,MAAM,EAAE,kBAAkB,EAAE,EAC5B,OAAO,GAAE,iBAAsB,GAC9B,gBAAgB,CAqLlB;AA4JD;;GAEG;AACH,wBAAgB,eAAe,CAAC,SAAS,EAAE,MAAM,EAAE,EAAE,QAAQ,EAAE,cAAc,EAAE,GAAG,MAAM,CAavF;AAED;;GAEG;AACH,wBAAgB,wBAAwB,CACtC,MAAM,EAAE,kBAAkB,EAAE,EAC5B,WAAW,EAAE,GAAG,CAAC,MAAM,EAAE,MAAM,CAAC,EAChC,QAAQ,EAAE,cAAc,EAAE,GACzB,MAAM,CA6DR"}