doc2vec 1.0.1 → 1.0.2

package/Dockerfile

@@ -1,19 +1,10 @@
-FROM node:20-slim AS base
+FROM node:20-slim
 
-# Install AWS CLI
-RUN apt-get update && \
-    apt-get install -y curl unzip git && \
-    curl "https://awscli.amazonaws.com/awscli-exe-linux-x86_64.zip" -o "awscliv2.zip" && \
-    unzip awscliv2.zip && \
-    ./aws/install && \
-    rm -rf awscliv2.zip aws && \
-    apt-get clean && rm -rf /var/lib/apt/lists/*
+WORKDIR /app
 
-# Copy project files
-COPY package.json /usr/src/app/
-COPY package-lock.json /usr/src/app/
-COPY tsconfig.json /usr/src/app/
-COPY *.ts /usr/src/app/
-WORKDIR /usr/src/app
-
-RUN npm install
+# Install Python and build tools
+RUN apt-get update && apt-get install -y \
+  python3 \
+  make \
+  g++ \
+  && apt-get clean

package/Dockerfile.old

@@ -0,0 +1,19 @@
+FROM node:20-slim AS base
+
+# Install AWS CLI
+RUN apt-get update && \
+    apt-get install -y curl unzip git && \
+    curl "https://awscli.amazonaws.com/awscli-exe-linux-x86_64.zip" -o "awscliv2.zip" && \
+    unzip awscliv2.zip && \
+    ./aws/install && \
+    rm -rf awscliv2.zip aws && \
+    apt-get clean && rm -rf /var/lib/apt/lists/*
+
+# Copy project files
+COPY package.json /usr/src/app/
+COPY package-lock.json /usr/src/app/
+COPY tsconfig.json /usr/src/app/
+COPY *.ts /usr/src/app/
+WORKDIR /usr/src/app
+
+RUN npm install

package/dist/doc2vec.js

@@ -690,16 +690,13 @@ class Doc2Vec {
                 logger.error(`Error during chunking or embedding for ${filePath}:`, error);
             }
         }, logger);
-        // 3. Update cleanup logic for URL rewriting
         logger.section('CLEANUP');
         if (dbConnection.type === 'sqlite') {
             logger.info(`Running SQLite cleanup for local directory ${config.path}`);
-            // Pass the config object directly to our consolidated function
             this.removeObsoleteFilesSQLite(dbConnection.db, processedFiles, config, logger);
         }
         else if (dbConnection.type === 'qdrant') {
             logger.info(`Running Qdrant cleanup for local directory ${config.path} in collection ${dbConnection.collectionName}`);
-            // Pass the config object directly to our consolidated function
             await this.removeObsoleteFilesQdrant(dbConnection, processedFiles, config, logger);
         }
         logger.info(`Finished processing local directory: ${config.path}`);
@@ -790,12 +787,11 @@ class Doc2Vec {
             logger.error(`Error reading directory ${dirPath}:`, error);
         }
     }
-    // SQLite consolidated function
     removeObsoleteFilesSQLite(db, processedFiles, pathConfig, logger) {
         const getChunksForPathStmt = db.prepare(`
-        SELECT chunk_id, url FROM vec_items
-        WHERE url LIKE ? || '%'
-    `);
+            SELECT chunk_id, url FROM vec_items
+            WHERE url LIKE ? || '%'
+        `);
         const deleteChunkStmt = db.prepare(`DELETE FROM vec_items WHERE chunk_id = ?`);
         // Determine if we're using URL rewriting or direct file paths
         const isRewriteMode = typeof pathConfig === 'object' && pathConfig.url_rewrite_prefix;

package/package.json

@@ -1,6 +1,6 @@
 {
     "name": "doc2vec",
-    "version": "1.0.1",
+    "version": "1.0.2",
     "type": "commonjs",
     "description": "",
     "main": "doc2vec.ts",
@@ -13,7 +13,9 @@
     },
     "scripts": {
         "build": "tsc && chmod 755 dist/doc2vec.js",
-        "start": "node dist/doc2vec.js"
+        "start": "node dist/doc2vec.js",
+        "prepublishOnly": "npm run build",
+        "prepare": "npm run build"
     },
     "keywords": [],
     "author": "",