Refactored and moved all keys and configuration into .env files and provided samples

retriever/.env.sample

@@ -0,0 +1,16 @@
+# Azure AI Search connection settings
+AZURE_AISEARCH_ENDPOINT=https://iva-demo-vector-service.search.windows.net
+AZURE_AISEARCH_KEY=your-azure-search-key
+AZURE_AISEARCH_INDEX_NAME=iva-vector-demo
+
+# If you're using Azure OpenAI API, you'll need to set these variables
+AZURE_OPENAI_API_KEY=your-azure-openai-key
+AZURE_OPENAI_API_INSTANCE_NAME=iva-open-ai
+AZURE_OPENAI_API_DEPLOYMENT_NAME=text-embedding-3-small
+AZURE_OPENAI_API_EMBEDDINGS_DEPLOYMENT_NAME=
+AZURE_OPENAI_API_VERSION=2024-08-01-preview
+
+# Or you can use the OpenAI API directly
+OPENAI_API_KEY=
+DEBUG=true
+

retriever/index.js

@@ -5,6 +5,7 @@ import {
 import { OpenAIEmbeddings, AzureOpenAIEmbeddings } from "@langchain/openai";
 
 const query = process.argv[2] || "What is CX Automation?";
+const company = process.argv[3] || "default";
 
 // the RAG widget uses the OpenAIEmbeddings class but the config will not work because you cannot pass in the api-version param. DO NOT USE
 // const embedding = new OpenAIEmbeddings({
@@ -34,11 +35,25 @@ function getSourceId(document) {
   if (document.metadata) {
     const mergedMetadata = Object.values(document.metadata).join("");
     const metatDataObj = JSON.parse(mergedMetadata);
-    return metatDataObj.source;
-  } else return undefined;
-}
 
-const resultDocuments = await store.similaritySearch(query);
+    if ("sourceURL" in metatDataObj) {
+      return metatDataObj.sourceURL;
+    }
+    if ("source" in metatDataObj) {
+      return metatDataObj.source;
+    }
+    if ("source_id" in metatDataObj) {
+      return metatDataObj.source_id;
+    }
+    if ("sourceName" in metatDataObj) {
+      return metatDataObj.sourceName;
+    }
+  } else return "no source found";
+}
+const filter = {
+  filterExpression: `search.in(company, '${company}')`,
+};
+const resultDocuments = await store.similaritySearch(query, 20, filter);
 const sources = resultDocuments.map((doc) => ({
   source_id: getSourceId(doc),
   text: doc.pageContent,

retriever/package.json

@@ -3,7 +3,7 @@
   "version": "1.0.0",
   "main": "index.js",
   "scripts": {
-    "test": "node --env-file=.env index.js \"What is Verint?\""
+    "test": "node --env-file=.env index.js \"What is Verint?\" \"Verint\""
   },
   "author": "",
   "license": "MIT",