Fix sqlite-vec dimension changes for local profiles

scripts/dev/bootstrap-local.mjs

@@ -551,12 +551,19 @@ function ensureCoreSchema(db) {
   `);
 }
 
-function tryInitVectorTables(db, dbPath) {
+function getEmbeddingDimensions(env) {
+  const defaultDimensions = env.EMBEDDING_PROFILE === 'openai-compatible' || env.EMBEDDING_PROFILE === 'custom'
+    ? '1024'
+    : '1536';
+  return Number(env.EMBEDDING_DIMENSIONS || defaultDimensions);
+}
+
+function tryInitVectorTables(db, dbPath, env) {
   const extension = process.platform === 'darwin' ? 'dylib' : process.platform === 'win32' ? 'dll' : 'so';
   const extensionPath = process.env.SQLITE_VEC_EXTENSION_PATH || path.join(repoDir, 'vendor', 'sqlite-extensions', `vec0.${extension}`);
-  const dimensions = Number(process.env.EMBEDDING_DIMENSIONS || '1536');
+  const dimensions = getEmbeddingDimensions(env);
   if (!Number.isInteger(dimensions) || dimensions <= 0) {
-    throw new Error(`Invalid EMBEDDING_DIMENSIONS="${process.env.EMBEDDING_DIMENSIONS}"`);
+    throw new Error(`Invalid EMBEDDING_DIMENSIONS="${env.EMBEDDING_DIMENSIONS}"`);
   }
 
   try {
@@ -585,11 +592,12 @@ function main() {
 
   ensureEnvFile();
 
-  const env = parseEnvFile(targetEnv);
+  const env = { ...parseEnvFile(targetEnv), ...process.env };
   if (process.env.SQLITE_DB_PATH) {
     ensureEnvValue('SQLITE_DB_PATH', process.env.SQLITE_DB_PATH);
+    env.SQLITE_DB_PATH = process.env.SQLITE_DB_PATH;
   }
-  const dbPath = expandPath(process.env.SQLITE_DB_PATH || env.SQLITE_DB_PATH || getDefaultDbPath());
+  const dbPath = expandPath(env.SQLITE_DB_PATH || getDefaultDbPath());
   fs.mkdirSync(path.dirname(dbPath), { recursive: true });
   if (!fs.existsSync(dbPath)) {
     fs.closeSync(fs.openSync(dbPath, 'w'));
@@ -598,7 +606,7 @@ function main() {
   const db = new Database(dbPath);
   try {
     ensureCoreSchema(db);
-    tryInitVectorTables(db, dbPath);
+    tryInitVectorTables(db, dbPath, env);
   } finally {
     db.close();
   }

scripts/rebuild-embeddings.ts

@@ -1,6 +1,7 @@
 import { getSQLiteClient } from '@/services/database/sqlite-client';
 import { NodeEmbedder } from '@/services/typescript/embed-nodes';
 import { UniversalEmbedder } from '@/services/typescript/embed-universal';
+import { getVectorBackendType } from '@/services/vectorBackend';
 
 async function maybeRecreateQdrantCollections() {
   if (process.env.VECTOR_BACKEND !== 'qdrant' || process.env.QDRANT_RECREATE_COLLECTIONS !== 'true') {
@@ -30,6 +31,10 @@ async function maybeRecreateQdrantCollections() {
 async function main() {
   const sqlite = getSQLiteClient();
   await maybeRecreateQdrantCollections();
+  if (getVectorBackendType() === 'sqlite-vec') {
+    sqlite.recreateVectorTables();
+    console.log('[rebuild-embeddings] Recreated sqlite-vec tables for the active embedding dimensions.');
+  }
   const nodeRows = sqlite.query<{ id: number; source?: string | null }>(`
     SELECT id, source
     FROM nodes

src/services/database/sqlite-client.ts

@@ -18,6 +18,7 @@ export interface SQLiteQueryResult<T = any> {
 }
 
 type FtsSurfaceName = 'nodes' | 'chunks';
+type VectorTableName = 'vec_nodes' | 'vec_chunks';
 
 interface IntegrityProbeResult {
   ok: boolean;
@@ -283,33 +284,82 @@ class SQLiteClient {
   public ensureVectorExtensions(): void {
     try {
       const dimensions = getEmbeddingProviderInfo().dimensions;
-      // Test for vec_nodes and vec_chunks; create them if missing
-      const hasVecNodes = this.db.prepare("SELECT name FROM sqlite_master WHERE type='table' AND name=?").get('vec_nodes');
-      if (!hasVecNodes) {
-        this.db.exec(`
-          CREATE VIRTUAL TABLE vec_nodes USING vec0(
-            node_id INTEGER PRIMARY KEY,
-            embedding FLOAT[${dimensions}]
-          );
-        `);
-        console.log('Created vec_nodes virtual table');
-      }
-
-      const hasVecChunks = this.db.prepare("SELECT name FROM sqlite_master WHERE type='table' AND name=?").get('vec_chunks');
-      if (!hasVecChunks) {
-        this.db.exec(`
-          CREATE VIRTUAL TABLE vec_chunks USING vec0(
-            chunk_id INTEGER PRIMARY KEY,
-            embedding FLOAT[${dimensions}]
-          );
-        `);
-        console.log('Created vec_chunks virtual table');
-      }
+      this.ensureVectorTable('vec_nodes', dimensions);
+      this.ensureVectorTable('vec_chunks', dimensions);
     } catch (error) {
       console.warn('Vector extension not available:', error);
     }
   }
 
+  private getVectorTableSql(tableName: VectorTableName): string | null {
+    const row = this.db.prepare("SELECT sql FROM sqlite_master WHERE type='table' AND name=?")
+      .get(tableName) as { sql?: string } | undefined;
+    return row?.sql || null;
+  }
+
+  public getVectorTableDimensions(): Record<'nodes' | 'chunks', number | null> {
+    return {
+      nodes: this.getVectorTableDimension('vec_nodes'),
+      chunks: this.getVectorTableDimension('vec_chunks'),
+    };
+  }
+
+  private getVectorTableDimension(tableName: VectorTableName): number | null {
+    const sql = this.getVectorTableSql(tableName);
+    const match = sql?.match(/embedding\s+FLOAT\[(\d+)\]/i);
+    return match ? Number(match[1]) : null;
+  }
+
+  private createVectorTable(tableName: VectorTableName, dimensions: number): void {
+    const idColumn = tableName === 'vec_nodes' ? 'node_id' : 'chunk_id';
+    this.db.exec(`
+      CREATE VIRTUAL TABLE ${tableName} USING vec0(
+        ${idColumn} INTEGER PRIMARY KEY,
+        embedding FLOAT[${dimensions}]
+      );
+    `);
+  }
+
+  private ensureVectorTable(tableName: VectorTableName, dimensions: number): void {
+    const existingSql = this.getVectorTableSql(tableName);
+    if (!existingSql) {
+      this.createVectorTable(tableName, dimensions);
+      console.log(`Created ${tableName} virtual table`);
+      return;
+    }
+
+    const existingDimensions = this.getVectorTableDimension(tableName);
+    if (existingDimensions === dimensions) {
+      return;
+    }
+
+    const rowCount = this.countVectorRows(tableName);
+    if (rowCount === 0 || this.maintenanceMode) {
+      this.db.exec(`DROP TABLE IF EXISTS ${tableName};`);
+      this.createVectorTable(tableName, dimensions);
+      console.warn(`Recreated ${tableName} virtual table for ${dimensions} dimensions (was ${existingDimensions ?? 'unknown'}).`);
+      return;
+    }
+
+    console.warn(
+      `${tableName} uses ${existingDimensions ?? 'unknown'} dimensions, but the active embedding profile requires ${dimensions}. ` +
+      'Run npm run rebuild:embeddings to recreate derived vectors.'
+    );
+  }
+
+  public recreateVectorTables(): void {
+    if (this.readOnly) {
+      throw new Error('Cannot recreate vector tables while SQLITE_READONLY=true.');
+    }
+    const dimensions = getEmbeddingProviderInfo().dimensions;
+    this.db.exec(`
+      DROP TABLE IF EXISTS vec_nodes;
+      DROP TABLE IF EXISTS vec_chunks;
+    `);
+    this.createVectorTable('vec_nodes', dimensions);
+    this.createVectorTable('vec_chunks', dimensions);
+  }
+
   private ensureVectorTables(): void {
     if (this.readOnly) {
       return;
@@ -1333,7 +1383,7 @@ class SQLiteClient {
     );
   }
 
-  private countVectorRows(tableName: 'vec_nodes' | 'vec_chunks'): number {
+  private countVectorRows(tableName: VectorTableName): number {
     try {
       const exists = this.db.prepare("SELECT 1 FROM sqlite_master WHERE type='table' AND name=?").get(tableName);
       if (!exists) return 0;

src/services/vectorBackend/sqlite-vec-backend.ts

@@ -99,10 +99,25 @@ export class SqliteVecBackend implements VectorBackend {
   async healthCheck(): Promise<VectorBackendHealth> {
     const sqlite = getSQLiteClient();
     const ok = await sqlite.checkVectorExtension();
+    const expectedDimensions = getEmbeddingProviderInfo().dimensions;
+    const tableDimensions = sqlite.getVectorTableDimensions();
+    const mismatchedTables = Object.entries(tableDimensions)
+      .filter(([, dimensions]) => dimensions !== null && dimensions !== expectedDimensions)
+      .map(([table, dimensions]) => `${table}=${dimensions}`);
+
+    if (mismatchedTables.length > 0) {
+      return {
+        ok: false,
+        backend: 'sqlite-vec',
+        dimensions: expectedDimensions,
+        detail: `sqlite-vec table dimensions mismatch (${mismatchedTables.join(', ')}), expected ${expectedDimensions}. Run npm run rebuild:embeddings.`,
+      };
+    }
+
     return {
       ok,
       backend: 'sqlite-vec',
-      dimensions: getEmbeddingProviderInfo().dimensions,
+      dimensions: expectedDimensions,
       detail: ok ? 'sqlite-vec extension loaded' : 'sqlite-vec extension unavailable',
     };
   }