doc2vec 1.1.1 → 1.3.0

package/mcp/package.json

@@ -1,41 +0,0 @@
-{
-  "name": "sqlite-vec-mcp-server",
-  "version": "1.0.0",
-  "description": "MCP Server for querying documentation with sqlite-vec",
-  "main": "build/index.js",
-  "type": "module",
-  "bin": {
-    "doc-query": "./build/index.js"
-  },
-  "scripts": {
-    "build": "tsc && chmod 755 build/index.js",
-    "start": "node build/index.js",
-    "dev": "npm run build && npm start"
-  },
-  "files": [
-    "build"
-  ],
-  "keywords": [
-    "mcp",
-    "sqlite-vec",
-    "vector",
-    "search"
-  ],
-  "author": "",
-  "license": "ISC",
-  "dependencies": {
-    "@modelcontextprotocol/sdk": "^1.9.0",
-    "better-sqlite3": "^11.8.1",
-    "dotenv": "^16.4.7",
-    "express": "^5.1.0",
-    "openai": "^4.89.0",
-    "sqlite-vec": "0.1.7-alpha.2",
-    "zod": "^3.24.2"
-  },
-  "devDependencies": {
-    "@types/better-sqlite3": "^7.6.13",
-    "@types/express": "^5.0.1",
-    "@types/node": "^22.14.0",
-    "typescript": "^5.8.3"
-  }
-}

package/mcp/src/index.ts

@@ -1,214 +0,0 @@
-// src/index.ts
-import 'dotenv/config'; // Load .env file
-import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js";
-import { SSEServerTransport } from "@modelcontextprotocol/sdk/server/sse.js";
-import express, { Request, Response } from "express";
-import { z } from "zod";
-
-import * as sqliteVec from "sqlite-vec";
-import Database, { Database as DatabaseType } from "better-sqlite3";
-import { OpenAI } from 'openai';
-import path from 'path';
-import fs from 'fs'; // Import fs for checking file existence
-
-// --- Configuration & Environment Check ---
-
-const openAIApiKey = process.env.OPENAI_API_KEY;
-const dbDir = process.env.SQLITE_DB_DIR || __dirname; // Default to current dir if not set
-
-if (!openAIApiKey) {
-    console.error("Error: OPENAI_API_KEY environment variable is not set.");
-    process.exit(1);
-}
-if (!fs.existsSync(dbDir)) {
-    console.warn(`Warning: SQLITE_DB_DIR (${dbDir}) does not exist. Databases may not be found.`);
-    process.exit(1);
-}
-
-const openai = new OpenAI({
-    apiKey: openAIApiKey,
-});
-
-export interface QueryResult {
-    chunk_id: string;
-    distance: number;
-    content: string;
-    url?: string;
-    embedding?: Float32Array | number[];
-    [key: string]: unknown;
-}
-
-async function createEmbeddings(text: string): Promise<number[]> {
-    try {
-        console.log("Calling OpenAI embeddings API...");
-        const startTime = Date.now();
-        const response = await openai.embeddings.create({
-            model: 'text-embedding-3-large', // Or your preferred model
-            input: text,
-        });
-        const duration = Date.now() - startTime;
-        console.log(`OpenAI embeddings received in ${duration}ms.`);
-        if (!response.data?.[0]?.embedding) {
-            throw new Error("Failed to get embedding from OpenAI response.");
-        }
-        return response.data[0].embedding;
-    } catch (error) {
-        console.error("Error creating OpenAI embeddings:", error);
-        throw new Error(`Failed to create embeddings: ${error instanceof Error ? error.message : String(error)}`);
-    }
-}
-
-function queryCollection(queryEmbedding: number[], filter: { product_name: string; version?: string }, topK: number = 10): QueryResult[] {
-    const dbPath = path.join(dbDir, `${filter.product_name}.db`);
-
-    if (!fs.existsSync(dbPath)) {
-        throw new Error(`Database file not found at ${dbPath}`);
-    }
-
-    let db: DatabaseType | null = null;
-    try {
-        db = new Database(dbPath);
-        console.log(`[DB ${dbPath}] Opened connection.`);
-        sqliteVec.load(db);
-        console.log(`[DB ${dbPath}] sqliteVec loaded.`);
-        let query = `
-              SELECT
-                  *,
-                  distance
-              FROM vec_items
-              WHERE embedding MATCH @query_embedding`;
-      
-        if (filter.product_name) query += ` AND product_name = @product_name`;
-        if (filter.version) query += ` AND version = @version`;
-      
-        query += `
-              ORDER BY distance
-              LIMIT @top_k;`;
-      
-        const stmt = db.prepare(query);
-        console.log(`[DB ${dbPath}] Query prepared. Executing...`);
-        const startTime = Date.now();
-        const rows = stmt.all({
-          query_embedding: new Float32Array(queryEmbedding),
-          product_name: filter.product_name,
-          version: filter.version,
-          top_k: topK,
-        });
-        const duration = Date.now() - startTime;
-        console.log(`[DB ${dbPath}] Query executed in ${duration}ms. Found ${rows.length} rows.`);
-      
-        rows.forEach((row: any) => {
-          delete row.embedding;
-        })
-      
-        return rows as QueryResult[];
-    } catch (error) {
-        console.error(`Error querying collection in ${dbPath}:`, error);
-        throw new Error(`Database query failed: ${error instanceof Error ? error.message : String(error)}`);
-    } finally {
-        if (db) {
-            db.close();
-        }
-    }
-}
-
-async function queryDocumentation(queryText: string, productName: string, version?: string, limit: number = 4): Promise<{ distance: number, content: string, url?: string }[]> {
-    const queryEmbedding = await createEmbeddings(queryText);
-    const results = queryCollection(queryEmbedding, { product_name: productName, version: version }, limit);
-    return results.map((qr: QueryResult) => ({
-        distance: qr.distance,
-        content: qr.content,
-        ...(qr.url && { url: qr.url }),
-    }));
-}
-
-
-// --- MCP Server Setup ---
-const serverName = "sqlite-vec-doc-query"; // Store name for logging
-const serverVersion = "1.0.0"; // Store version for logging
-
-const server = new McpServer({
-    name: serverName,
-    version: serverVersion,
-    capabilities: {},
-});
-
-// --- Define the MCP Tool ---
-server.tool(
-    "query-documentation",
-    "Query documentation stored in a sqlite-vec database using vector search.",
-    {
-        queryText: z.string().min(1).describe("The natural language query to search for."),
-        productName: z.string().min(1).describe("The name of the product documentation database to search within (e.g., 'my-product'). Corresponds to the DB filename without .db."),
-        version: z.string().optional().describe("The specific version of the product documentation (e.g., '1.2.0'). Optional."),
-        limit: z.number().int().positive().optional().default(4).describe("Maximum number of results to return. Defaults to 4."),
-    },
-    async ({ queryText, productName, version, limit }) => {
-        console.log(`Received query: text="${queryText}", product="${productName}", version="${version || 'any'}", limit=${limit}`);
-
-        try {
-            const results = await queryDocumentation(queryText, productName, version, limit);
-
-            if (results.length === 0) {
-                return {
-                    content: [{ type: "text", text: `No relevant documentation found for "${queryText}" in product "${productName}" ${version ? `(version ${version})` : ''}.` }],
-                };
-            }
-
-            const formattedResults = results.map((r, index) =>
-                [
-                    `Result ${index + 1}:`,
-                    `  Content: ${r.content}`,
-                    `  Distance: ${r.distance.toFixed(4)}`,
-                    r.url ? `  URL: ${r.url}` : null,
-                    "---"
-                ].filter(line => line !== null).join("\n")
-            ).join("\n");
-
-            const responseText = `Found ${results.length} relevant documentation snippets for "${queryText}" in product "${productName}" ${version ? `(version ${version})` : ''}:\n\n${formattedResults}`;
-            console.log(`Handler finished processing. Payload size (approx): ${responseText.length} chars. Returning response object...`);
-
-            return {
-                content: [{ type: "text", text: responseText }],
-            };
-        } catch (error: any) {
-            console.error("Error processing 'query-documentation' tool:", error);
-            return {
-                content: [{ type: "text", text: `Error querying documentation: ${error.message}` }],
-            };
-        }
-    }
-);
-
-const app = express();
-
-// to support multiple simultaneous connections we have a lookup object from
-// sessionId to transport
-const transports: {[sessionId: string]: SSEServerTransport} = {};
-
-app.get("/sse", async (_: Request, res: Response) => {
-  const transport = new SSEServerTransport('/messages', res);
-  transports[transport.sessionId] = transport;
-  res.on("close", () => {
-    delete transports[transport.sessionId];
-  });
-  await server.connect(transport);
-});
-
-app.post("/messages", async (req: Request, res: Response) => {
-  const sessionId = req.query.sessionId as string;
-  const transport = transports[sessionId];
-  if (transport) {
-    await transport.handlePostMessage(req, res);
-  } else {
-    res.status(400).send('No transport found for sessionId');
-  }
-});
-
-const PORT = process.env.PORT || 3001;
-
-const webserver = app.listen(PORT, () => {
-    console.log(`Server is running on port ${PORT}`);
-});
-
-webserver.keepAliveTimeout = 3000;

package/mcp/tsconfig.json

@@ -1,16 +0,0 @@
-{
-    "compilerOptions": {
-      "target": "ES2022",
-      "module": "Node16",
-      "moduleResolution": "Node16",
-      "outDir": "./build",
-      "rootDir": "./src",
-      "strict": true,
-      "esModuleInterop": true,
-      "skipLibCheck": true,
-      "forceConsistentCasingInFileNames": true,
-      "resolveJsonModule": true
-    },
-    "include": ["src/**/*"],
-    "exclude": ["node_modules", "build"]
-}

package/tsconfig.json

@@ -1,18 +0,0 @@
-{
-    "compilerOptions": {
-      "target": "ES2020",
-      "module": "CommonJS",
-      "lib": ["es2022", "dom"],
-      "declaration": false,
-      "sourceMap": false,
-      "outDir": "./dist",
-      "rootDir": "./",
-      "strict": true,
-      "esModuleInterop": true,
-      "skipLibCheck": true,
-      "forceConsistentCasingInFileNames": true,
-      "moduleResolution": "node"
-    },
-    "include": ["**/*.ts"],
-    "exclude": ["node_modules", "mcp"]
-  }