From dbc3d0f670994de8c2a248a582795f49d4b37a3a Mon Sep 17 00:00:00 2001
From: "Dina Berry (She/her)" <diberry@microsoft.com>
Date: Fri, 22 May 2026 07:26:16 -0700
Subject: [PATCH 01/10] fix: address code review issues in select-algorithm
 samples

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
---
 ai/select-algorithm-dotnet/Models/Configuration.cs       | 2 +-
 ai/select-algorithm-go/src/utils.go                      | 2 +-
 .../com/azure/documentdb/selectalgorithm/CompareAll.java | 9 ++++-----
 .../java/com/azure/documentdb/selectalgorithm/Utils.java | 9 ++++-----
 ai/select-algorithm-python/src/utils.py                  | 2 +-
 ai/select-algorithm-typescript/src/utils.ts              | 9 +++++++--
 6 files changed, 18 insertions(+), 15 deletions(-)

diff --git a/ai/select-algorithm-dotnet/Models/Configuration.cs b/ai/select-algorithm-dotnet/Models/Configuration.cs
index cbca25b..9f241a8 100644
--- a/ai/select-algorithm-dotnet/Models/Configuration.cs
+++ b/ai/select-algorithm-dotnet/Models/Configuration.cs
@@ -37,5 +37,5 @@ public class VectorSearchConfiguration
 
 public class DataFilesConfiguration
 {
-    public string WithVectors { get; set; } = "data/Hotels_Vector.json";
+    public string WithVectors { get; set; } = "../data/Hotels_Vector.json";
 }
diff --git a/ai/select-algorithm-go/src/utils.go b/ai/select-algorithm-go/src/utils.go
index c358892..c6042b6 100644
--- a/ai/select-algorithm-go/src/utils.go
+++ b/ai/select-algorithm-go/src/utils.go
@@ -60,7 +60,7 @@ func LoadConfig() (*Config, error) {
 	return &Config{
 		ClusterName:  getEnvOrDefault("DOCUMENTDB_CLUSTER_NAME", ""),
 		DatabaseName: getEnvOrDefault("AZURE_DOCUMENTDB_DATABASENAME", "Hotels"),
-		DataFile:     getEnvOrDefault("DATA_FILE_WITH_VECTORS", "data/Hotels_Vector.json"),
+		DataFile:     getEnvOrDefault("DATA_FILE_WITH_VECTORS", "../data/Hotels_Vector.json"),
 		VectorField:  getEnvOrDefault("EMBEDDED_FIELD", "DescriptionVector"),
 		ModelName:    getEnvOrDefault("AZURE_OPENAI_EMBEDDING_MODEL", "text-embedding-3-small"),
 		Dimensions:   dimensions,
diff --git a/ai/select-algorithm-java/src/main/java/com/azure/documentdb/selectalgorithm/CompareAll.java b/ai/select-algorithm-java/src/main/java/com/azure/documentdb/selectalgorithm/CompareAll.java
index 66281ed..e41fd46 100644
--- a/ai/select-algorithm-java/src/main/java/com/azure/documentdb/selectalgorithm/CompareAll.java
+++ b/ai/select-algorithm-java/src/main/java/com/azure/documentdb/selectalgorithm/CompareAll.java
@@ -28,7 +28,7 @@ public static void run() {
         int topK = Integer.parseInt(Utils.getEnv("TOP_K", "5"));
 
         String databaseName = Utils.getEnv("AZURE_DOCUMENTDB_DATABASENAME", "Hotels");
-        String dataFile = Utils.getEnv("DATA_FILE_WITH_VECTORS", "data/Hotels_Vector.json");
+        String dataFile = Utils.getEnv("DATA_FILE_WITH_VECTORS", "../data/Hotels_Vector.json");
         String vectorField = Utils.getEnv("EMBEDDED_FIELD", "DescriptionVector");
         int dimensions = Integer.parseInt(Utils.getEnv("EMBEDDING_DIMENSIONS", "1536"));
         String model = Utils.getEnv("AZURE_OPENAI_EMBEDDING_MODEL", "text-embedding-3-small");
@@ -113,12 +113,11 @@ public static void run() {
 
             int successCount = (int) results.stream().filter(r -> !r.top1Name().equals("(failed)")).count();
             if (successCount == 0) {
-                System.out.println("\n❌ All 9 comparisons failed — no algorithm returned results.");
-                System.exit(1);
-            } else {
-                System.out.printf("%nSummary: %d succeeded, %d failed%n", successCount, 9 - successCount);
+                throw new IllegalStateException("All 9 comparisons failed — no algorithm returned results.");
             }
 
+            System.out.printf("%nSummary: %d succeeded, %d failed%n", successCount, 9 - successCount);
+
             // Cleanup: drop the comparison collection
             System.out.println("\n  Cleanup: dropping comparison collection...");
             collection.drop();
diff --git a/ai/select-algorithm-java/src/main/java/com/azure/documentdb/selectalgorithm/Utils.java b/ai/select-algorithm-java/src/main/java/com/azure/documentdb/selectalgorithm/Utils.java
index b8b761e..30dca39 100644
--- a/ai/select-algorithm-java/src/main/java/com/azure/documentdb/selectalgorithm/Utils.java
+++ b/ai/select-algorithm-java/src/main/java/com/azure/documentdb/selectalgorithm/Utils.java
@@ -7,7 +7,6 @@
 import com.azure.core.credential.AccessToken;
 import com.azure.identity.DefaultAzureCredential;
 import com.azure.identity.DefaultAzureCredentialBuilder;
-import com.mongodb.ConnectionString;
 import com.mongodb.MongoClientSettings;
 import com.mongodb.MongoCredential;
 import com.mongodb.client.MongoClient;
@@ -39,8 +38,7 @@ public static MongoClient getMongoClient() {
             throw new IllegalStateException("DOCUMENTDB_CLUSTER_NAME environment variable is required");
         }
 
-        String connectionUri = String.format(
-                "mongodb+srv://%s.global.mongocluster.cosmos.azure.com/", clusterName);
+        String clusterHost = String.format("%s.global.mongocluster.cosmos.azure.com", clusterName);
 
         // Use custom OIDC callback with DefaultAzureCredential
         // This chains through CLI, managed identity, etc.
@@ -56,9 +54,10 @@ public static MongoClient getMongoClient() {
                 });
 
         MongoClientSettings settings = MongoClientSettings.builder()
-                .applyConnectionString(new ConnectionString(connectionUri))
+                .applyToClusterSettings(builder -> builder.srvHost(clusterHost))
+                .applyToSslSettings(builder -> builder.enabled(true))
                 .credential(mongoCredential)
-                .retryWrites(false)
+                .retryWrites(true)
                 .build();
 
         return MongoClients.create(settings);
diff --git a/ai/select-algorithm-python/src/utils.py b/ai/select-algorithm-python/src/utils.py
index de0f8b4..e97ea9a 100644
--- a/ai/select-algorithm-python/src/utils.py
+++ b/ai/select-algorithm-python/src/utils.py
@@ -64,7 +64,7 @@ def get_config() -> Dict[str, Any]:
     """Load configuration from environment variables."""
     return {
         'database_name': os.getenv('AZURE_DOCUMENTDB_DATABASENAME', 'Hotels'),
-        'data_file': os.getenv('DATA_FILE_WITH_VECTORS', 'data/Hotels_Vector.json'),
+        'data_file': os.getenv('DATA_FILE_WITH_VECTORS', '../data/Hotels_Vector.json'),
         'vector_field': os.getenv('EMBEDDED_FIELD', 'DescriptionVector'),
         'model_name': os.getenv('AZURE_OPENAI_EMBEDDING_MODEL', 'text-embedding-3-small'),
         'dimensions': int(os.getenv('EMBEDDING_DIMENSIONS', '1536')),
diff --git a/ai/select-algorithm-typescript/src/utils.ts b/ai/select-algorithm-typescript/src/utils.ts
index 09ec3dd..5f99337 100644
--- a/ai/select-algorithm-typescript/src/utils.ts
+++ b/ai/select-algorithm-typescript/src/utils.ts
@@ -20,9 +20,14 @@ export function getConfig() {
 
 export const AzureIdentityTokenCallback = async (params: OIDCCallbackParams, credential: TokenCredential): Promise<OIDCResponse> => {
     const tokenResponse: AccessToken | null = await credential.getToken(['https://ossrdbms-aad.database.windows.net/.default']);
+
+    if (!tokenResponse || !tokenResponse.token) {
+        throw new Error('Failed to acquire token');
+    }
+
     return {
-        accessToken: tokenResponse?.token || '',
-        expiresInSeconds: Math.floor(((tokenResponse?.expiresOnTimestamp || 0) - Date.now()) / 1000)
+        accessToken: tokenResponse.token,
+        expiresInSeconds: Math.floor((tokenResponse.expiresOnTimestamp - Date.now()) / 1000)
     };
 };
 

From 8a2796cb5560e6979e595db4bd627d4826de6c06 Mon Sep 17 00:00:00 2001
From: "Dina Berry (She/her)" <diberry@microsoft.com>
Date: Fri, 22 May 2026 07:49:28 -0700
Subject: [PATCH 02/10] Fix data paths (local data/) and Java
 retryWrites(false)

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
---
 ai/select-algorithm-dotnet/Models/Configuration.cs        | 2 +-
 ai/select-algorithm-dotnet/README.md                      | 6 +++---
 ai/select-algorithm-dotnet/appsettings.json               | 2 +-
 ai/select-algorithm-go/README.md                          | 8 ++++----
 ai/select-algorithm-go/src/utils.go                       | 2 +-
 ai/select-algorithm-java/README.md                        | 4 ++--
 .../com/azure/documentdb/selectalgorithm/CompareAll.java  | 2 +-
 .../java/com/azure/documentdb/selectalgorithm/Utils.java  | 2 +-
 ai/select-algorithm-python/README.md                      | 4 ++--
 ai/select-algorithm-python/src/utils.py                   | 2 +-
 10 files changed, 17 insertions(+), 17 deletions(-)

diff --git a/ai/select-algorithm-dotnet/Models/Configuration.cs b/ai/select-algorithm-dotnet/Models/Configuration.cs
index 9f241a8..cbca25b 100644
--- a/ai/select-algorithm-dotnet/Models/Configuration.cs
+++ b/ai/select-algorithm-dotnet/Models/Configuration.cs
@@ -37,5 +37,5 @@ public class VectorSearchConfiguration
 
 public class DataFilesConfiguration
 {
-    public string WithVectors { get; set; } = "../data/Hotels_Vector.json";
+    public string WithVectors { get; set; } = "data/Hotels_Vector.json";
 }
diff --git a/ai/select-algorithm-dotnet/README.md b/ai/select-algorithm-dotnet/README.md
index 2621f77..fd30957 100644
--- a/ai/select-algorithm-dotnet/README.md
+++ b/ai/select-algorithm-dotnet/README.md
@@ -46,7 +46,7 @@ Demonstrates three vector index algorithms available in Azure DocumentDB:
        "EmbeddingSizeBatch": 16
      },
      "DataFiles": {
-       "WithVectors": "../data/Hotels_Vector.json"
+       "WithVectors": "data/Hotels_Vector.json"
      }
    }
    ```
@@ -54,7 +54,7 @@ Demonstrates three vector index algorithms available in Azure DocumentDB:
 3. Copy the shared data file:
 
    ```bash
-   cp ../data/Hotels_Vector.json .
+   cp data/Hotels_Vector.json .
    ```
 
 4. Restore packages:
@@ -78,7 +78,7 @@ dotnet run
 | `MongoDB:ClusterName` | (required) | DocumentDB cluster name |
 | `AzureOpenAI:Endpoint` | (required) | Azure OpenAI endpoint |
 | `AzureOpenAI:EmbeddingModel` | (required) | Embedding model deployment name |
-| `DataFiles:WithVectors` | `../data/Hotels_Vector.json` | Path to vectors JSON file |
+| `DataFiles:WithVectors` | `data/Hotels_Vector.json` | Path to vectors JSON file |
 | `Embedding:EmbeddedField` | `DescriptionVector` | Field name containing embeddings |
 | `Embedding:Dimensions` | `1536` | Vector dimensions |
 | `MongoDB:DatabaseName` | `Hotels` | Target database name |
diff --git a/ai/select-algorithm-dotnet/appsettings.json b/ai/select-algorithm-dotnet/appsettings.json
index 68ee696..3b38afb 100644
--- a/ai/select-algorithm-dotnet/appsettings.json
+++ b/ai/select-algorithm-dotnet/appsettings.json
@@ -14,7 +14,7 @@
     "EmbeddingModel": "text-embedding-3-small"
   },
   "DataFiles": {
-    "WithVectors": "../data/Hotels_Vector.json"
+    "WithVectors": "data/Hotels_Vector.json"
   },
   "Embedding": {
     "EmbeddedField": "DescriptionVector",
diff --git a/ai/select-algorithm-go/README.md b/ai/select-algorithm-go/README.md
index f03828e..de14f79 100644
--- a/ai/select-algorithm-go/README.md
+++ b/ai/select-algorithm-go/README.md
@@ -38,7 +38,7 @@ This sample demonstrates how to compare different vector search algorithms (IVF,
    AZURE_OPENAI_EMBEDDING_ENDPOINT=https://your-resource.openai.azure.com
    AZURE_OPENAI_EMBEDDING_MODEL=text-embedding-3-small
    AZURE_DOCUMENTDB_DATABASENAME=Hotels
-   DATA_FILE_WITH_VECTORS=../data/Hotels_Vector.json
+   DATA_FILE_WITH_VECTORS=data/Hotels_Vector.json
    EMBEDDED_FIELD=DescriptionVector
    EMBEDDING_DIMENSIONS=1536
    ```
@@ -46,10 +46,10 @@ This sample demonstrates how to compare different vector search algorithms (IVF,
 3. **Copy the shared data file** into this directory:
 
    ```bash
-   cp ../data/Hotels_Vector.json .
+   cp data/Hotels_Vector.json .
    ```
 
-   The `DATA_FILE_WITH_VECTORS` env var defaults to `../data/Hotels_Vector.json`.
+   The `DATA_FILE_WITH_VECTORS` env var defaults to `data/Hotels_Vector.json`.
 
 4. **Install dependencies**:
 
@@ -116,7 +116,7 @@ go run ./src/...
 | `AZURE_OPENAI_EMBEDDING_ENDPOINT` | *(required)* | Azure OpenAI endpoint |
 | `AZURE_OPENAI_EMBEDDING_MODEL` | `text-embedding-3-small` | Embedding model name |
 | `AZURE_DOCUMENTDB_DATABASENAME` | `Hotels` | Database name |
-| `DATA_FILE_WITH_VECTORS` | `../data/Hotels_Vector.json` | Path to data file |
+| `DATA_FILE_WITH_VECTORS` | `data/Hotels_Vector.json` | Path to data file |
 | `EMBEDDED_FIELD` | `DescriptionVector` | Field containing embeddings |
 | `EMBEDDING_DIMENSIONS` | `1536` | Embedding vector dimensions |
 | `LOAD_SIZE_BATCH` | `100` | Batch size for data insertion |
diff --git a/ai/select-algorithm-go/src/utils.go b/ai/select-algorithm-go/src/utils.go
index c6042b6..c358892 100644
--- a/ai/select-algorithm-go/src/utils.go
+++ b/ai/select-algorithm-go/src/utils.go
@@ -60,7 +60,7 @@ func LoadConfig() (*Config, error) {
 	return &Config{
 		ClusterName:  getEnvOrDefault("DOCUMENTDB_CLUSTER_NAME", ""),
 		DatabaseName: getEnvOrDefault("AZURE_DOCUMENTDB_DATABASENAME", "Hotels"),
-		DataFile:     getEnvOrDefault("DATA_FILE_WITH_VECTORS", "../data/Hotels_Vector.json"),
+		DataFile:     getEnvOrDefault("DATA_FILE_WITH_VECTORS", "data/Hotels_Vector.json"),
 		VectorField:  getEnvOrDefault("EMBEDDED_FIELD", "DescriptionVector"),
 		ModelName:    getEnvOrDefault("AZURE_OPENAI_EMBEDDING_MODEL", "text-embedding-3-small"),
 		Dimensions:   dimensions,
diff --git a/ai/select-algorithm-java/README.md b/ai/select-algorithm-java/README.md
index 2449f40..119efb0 100644
--- a/ai/select-algorithm-java/README.md
+++ b/ai/select-algorithm-java/README.md
@@ -37,7 +37,7 @@ This sample demonstrates how to compare all three vector search index algorithms
 3. Copy the shared data file:
 
    ```bash
-   cp ../data/Hotels_Vector.json .
+   cp data/Hotels_Vector.json .
    ```
 
 ## Build
@@ -81,7 +81,7 @@ $env:ALGORITHM="compare"; mvn exec:java
 | `DOCUMENTDB_CLUSTER_NAME` | (required) | DocumentDB cluster name |
 | `AZURE_OPENAI_EMBEDDING_ENDPOINT` | (required) | Azure OpenAI endpoint |
 | `AZURE_OPENAI_EMBEDDING_MODEL` | (required) | Embedding model deployment name |
-| `DATA_FILE_WITH_VECTORS` | `../data/Hotels_Vector.json` | Path to vectors JSON file |
+| `DATA_FILE_WITH_VECTORS` | `data/Hotels_Vector.json` | Path to vectors JSON file |
 | `EMBEDDED_FIELD` | `DescriptionVector` | Field name containing embeddings |
 | `EMBEDDING_DIMENSIONS` | `1536` | Vector dimensions |
 | `AZURE_DOCUMENTDB_DATABASENAME` | `Hotels` | Target database name |
diff --git a/ai/select-algorithm-java/src/main/java/com/azure/documentdb/selectalgorithm/CompareAll.java b/ai/select-algorithm-java/src/main/java/com/azure/documentdb/selectalgorithm/CompareAll.java
index e41fd46..2d22e64 100644
--- a/ai/select-algorithm-java/src/main/java/com/azure/documentdb/selectalgorithm/CompareAll.java
+++ b/ai/select-algorithm-java/src/main/java/com/azure/documentdb/selectalgorithm/CompareAll.java
@@ -28,7 +28,7 @@ public static void run() {
         int topK = Integer.parseInt(Utils.getEnv("TOP_K", "5"));
 
         String databaseName = Utils.getEnv("AZURE_DOCUMENTDB_DATABASENAME", "Hotels");
-        String dataFile = Utils.getEnv("DATA_FILE_WITH_VECTORS", "../data/Hotels_Vector.json");
+        String dataFile = Utils.getEnv("DATA_FILE_WITH_VECTORS", "data/Hotels_Vector.json");
         String vectorField = Utils.getEnv("EMBEDDED_FIELD", "DescriptionVector");
         int dimensions = Integer.parseInt(Utils.getEnv("EMBEDDING_DIMENSIONS", "1536"));
         String model = Utils.getEnv("AZURE_OPENAI_EMBEDDING_MODEL", "text-embedding-3-small");
diff --git a/ai/select-algorithm-java/src/main/java/com/azure/documentdb/selectalgorithm/Utils.java b/ai/select-algorithm-java/src/main/java/com/azure/documentdb/selectalgorithm/Utils.java
index 30dca39..641dcb5 100644
--- a/ai/select-algorithm-java/src/main/java/com/azure/documentdb/selectalgorithm/Utils.java
+++ b/ai/select-algorithm-java/src/main/java/com/azure/documentdb/selectalgorithm/Utils.java
@@ -57,7 +57,7 @@ public static MongoClient getMongoClient() {
                 .applyToClusterSettings(builder -> builder.srvHost(clusterHost))
                 .applyToSslSettings(builder -> builder.enabled(true))
                 .credential(mongoCredential)
-                .retryWrites(true)
+                .retryWrites(false)
                 .build();
 
         return MongoClients.create(settings);
diff --git a/ai/select-algorithm-python/README.md b/ai/select-algorithm-python/README.md
index 1fe7746..becef58 100644
--- a/ai/select-algorithm-python/README.md
+++ b/ai/select-algorithm-python/README.md
@@ -57,7 +57,7 @@ Compare IVF, HNSW, and DiskANN vector index algorithms in Azure DocumentDB. Each
 3. Copy the shared data file:
 
    ```bash
-   cp ../data/Hotels_Vector.json .
+   cp data/Hotels_Vector.json .
    ```
 
 4. Ensure you're logged in to Azure:
@@ -83,7 +83,7 @@ The script creates a single `hotels` collection, loads data once, then for each
 | `DOCUMENTDB_CLUSTER_NAME` | (required) | DocumentDB cluster name |
 | `AZURE_OPENAI_EMBEDDING_ENDPOINT` | (required) | Azure OpenAI endpoint |
 | `AZURE_OPENAI_EMBEDDING_MODEL` | (required) | Embedding model deployment name |
-| `DATA_FILE_WITH_VECTORS` | `../data/Hotels_Vector.json` | Path to vectors JSON file |
+| `DATA_FILE_WITH_VECTORS` | `data/Hotels_Vector.json` | Path to vectors JSON file |
 | `EMBEDDED_FIELD` | `DescriptionVector` | Field name containing embeddings |
 | `EMBEDDING_DIMENSIONS` | `1536` | Vector dimensions |
 | `AZURE_DOCUMENTDB_DATABASENAME` | `Hotels` | Target database name |
diff --git a/ai/select-algorithm-python/src/utils.py b/ai/select-algorithm-python/src/utils.py
index e97ea9a..de0f8b4 100644
--- a/ai/select-algorithm-python/src/utils.py
+++ b/ai/select-algorithm-python/src/utils.py
@@ -64,7 +64,7 @@ def get_config() -> Dict[str, Any]:
     """Load configuration from environment variables."""
     return {
         'database_name': os.getenv('AZURE_DOCUMENTDB_DATABASENAME', 'Hotels'),
-        'data_file': os.getenv('DATA_FILE_WITH_VECTORS', '../data/Hotels_Vector.json'),
+        'data_file': os.getenv('DATA_FILE_WITH_VECTORS', 'data/Hotels_Vector.json'),
         'vector_field': os.getenv('EMBEDDED_FIELD', 'DescriptionVector'),
         'model_name': os.getenv('AZURE_OPENAI_EMBEDDING_MODEL', 'text-embedding-3-small'),
         'dimensions': int(os.getenv('EMBEDDING_DIMENSIONS', '1536')),

From 3167d632572e81294edea848fec4466c85cc7930 Mon Sep 17 00:00:00 2001
From: "Dina Berry (She/her)" <diberry@microsoft.com>
Date: Fri, 22 May 2026 08:03:47 -0700
Subject: [PATCH 03/10] Fix TS README: data path is local data/ not ../data/

---
 ai/select-algorithm-typescript/README.md | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/ai/select-algorithm-typescript/README.md b/ai/select-algorithm-typescript/README.md
index 73414a8..254eee3 100644
--- a/ai/select-algorithm-typescript/README.md
+++ b/ai/select-algorithm-typescript/README.md
@@ -52,13 +52,14 @@ Compare IVF, HNSW, and DiskANN vector index algorithms in Azure DocumentDB using
     | `LOAD_SIZE_BATCH` | Batch size for data insertion |
     | `SIMILARITY` | Similarity metric: `COS`, `L2`, or `IP` |
 
-5. **Copy the shared data file** into this directory:
+5. **Create a `data/` directory and copy the data file:**
 
     ```bash
-    cp ../data/Hotels_Vector.json .
+    mkdir -p data
+    cp ../data/Hotels_Vector.json data/
     ```
 
-    The `DATA_FILE_WITH_VECTORS` env var defaults to `../data/Hotels_Vector.json`.
+    The `DATA_FILE_WITH_VECTORS` env var defaults to `data/Hotels_Vector.json`.
 
 6. **Build the project:**
 

From 63da6f3f1cbb73867397f6f5ef508d0aa9ff31fb Mon Sep 17 00:00:00 2001
From: "Dina Berry (She/her)" <diberry@microsoft.com>
Date: Fri, 22 May 2026 08:05:10 -0700
Subject: [PATCH 04/10] Rename .NET config section MongoDB -> DocumentDB

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
---
 ai/select-algorithm-dotnet/CompareAll.cs           |  4 ++--
 ai/select-algorithm-dotnet/Models/Configuration.cs |  4 ++--
 ai/select-algorithm-dotnet/README.md               |  8 ++++----
 ai/select-algorithm-dotnet/Utils.cs                |  4 ++--
 ai/select-algorithm-dotnet/appsettings.json        |  2 +-
 ai/select-algorithm-dotnet/quickstart.md           | 10 +++++-----
 6 files changed, 16 insertions(+), 16 deletions(-)

diff --git a/ai/select-algorithm-dotnet/CompareAll.cs b/ai/select-algorithm-dotnet/CompareAll.cs
index 9eb9c75..2cf3d10 100644
--- a/ai/select-algorithm-dotnet/CompareAll.cs
+++ b/ai/select-algorithm-dotnet/CompareAll.cs
@@ -30,11 +30,11 @@ public static void Run(AppConfiguration appConfig)
         Console.WriteLine(new string('=', 60));
 
         // Use config values with env var overrides for compare-specific settings
-        var databaseName = appConfig.MongoDB.DatabaseName;
+        var databaseName = appConfig.DocumentDB.DatabaseName;
         var dataFile = appConfig.DataFiles.WithVectors;
         var vectorField = appConfig.Embedding.EmbeddedField;
         var dimensions = appConfig.Embedding.Dimensions;
-        var batchSize = appConfig.MongoDB.LoadBatchSize;
+        var batchSize = appConfig.DocumentDB.LoadBatchSize;
         var queryText = Environment.GetEnvironmentVariable("QUERY_TEXT") ?? "luxury hotel near the beach";
         var topK = int.Parse(Environment.GetEnvironmentVariable("TOP_K") ?? "5");
 
diff --git a/ai/select-algorithm-dotnet/Models/Configuration.cs b/ai/select-algorithm-dotnet/Models/Configuration.cs
index cbca25b..3b4cb09 100644
--- a/ai/select-algorithm-dotnet/Models/Configuration.cs
+++ b/ai/select-algorithm-dotnet/Models/Configuration.cs
@@ -3,7 +3,7 @@ namespace SelectAlgorithm.Models;
 public class AppConfiguration
 {
     public AzureOpenAIConfiguration AzureOpenAI { get; set; } = new();
-    public MongoDBConfiguration MongoDB { get; set; } = new();
+    public DocumentDBConfiguration DocumentDB { get; set; } = new();
     public EmbeddingConfiguration Embedding { get; set; } = new();
     public VectorSearchConfiguration VectorSearch { get; set; } = new();
     public DataFilesConfiguration DataFiles { get; set; } = new();
@@ -15,7 +15,7 @@ public class AzureOpenAIConfiguration
     public string EmbeddingModel { get; set; } = "text-embedding-3-small";
 }
 
-public class MongoDBConfiguration
+public class DocumentDBConfiguration
 {
     public string ClusterName { get; set; } = string.Empty;
     public string DatabaseName { get; set; } = "Hotels";
diff --git a/ai/select-algorithm-dotnet/README.md b/ai/select-algorithm-dotnet/README.md
index fd30957..32e25cf 100644
--- a/ai/select-algorithm-dotnet/README.md
+++ b/ai/select-algorithm-dotnet/README.md
@@ -35,7 +35,7 @@ Demonstrates three vector index algorithms available in Azure DocumentDB:
        "EmbeddingModel": "text-embedding-3-small",
        "Endpoint": "https://<your-resource>.openai.azure.com"
      },
-     "MongoDB": {
+     "DocumentDB": {
        "ClusterName": "<your-cluster-name>",
        "DatabaseName": "Hotels",
        "LoadBatchSize": 100
@@ -75,14 +75,14 @@ dotnet run
 
 | Setting (appsettings.json) | Default | Description |
 |---------------------------|---------|-------------|
-| `MongoDB:ClusterName` | (required) | DocumentDB cluster name |
+| `DocumentDB:ClusterName` | (required) | DocumentDB cluster name |
 | `AzureOpenAI:Endpoint` | (required) | Azure OpenAI endpoint |
 | `AzureOpenAI:EmbeddingModel` | (required) | Embedding model deployment name |
 | `DataFiles:WithVectors` | `data/Hotels_Vector.json` | Path to vectors JSON file |
 | `Embedding:EmbeddedField` | `DescriptionVector` | Field name containing embeddings |
 | `Embedding:Dimensions` | `1536` | Vector dimensions |
-| `MongoDB:DatabaseName` | `Hotels` | Target database name |
-| `MongoDB:LoadBatchSize` | `100` | Batch size for data loading |
+| `DocumentDB:DatabaseName` | `Hotels` | Target database name |
+| `DocumentDB:LoadBatchSize` | `100` | Batch size for data loading |
 | `Embedding:EmbeddingSizeBatch` | `16` | Batch size for embedding requests |
 
 **Additional environment variables for compare mode:**
diff --git a/ai/select-algorithm-dotnet/Utils.cs b/ai/select-algorithm-dotnet/Utils.cs
index 62590ad..01d97a2 100644
--- a/ai/select-algorithm-dotnet/Utils.cs
+++ b/ai/select-algorithm-dotnet/Utils.cs
@@ -34,9 +34,9 @@ public static class Utils
 {
     public static IMongoClient GetMongoClientPasswordless(AppConfiguration config)
     {
-        var clusterName = config.MongoDB.ClusterName;
+        var clusterName = config.DocumentDB.ClusterName;
         if (string.IsNullOrEmpty(clusterName))
-            throw new InvalidOperationException("MongoDB:ClusterName is required in appsettings.json");
+            throw new InvalidOperationException("DocumentDB:ClusterName is required in appsettings.json");
 
         var credential = new DefaultAzureCredential();
 
diff --git a/ai/select-algorithm-dotnet/appsettings.json b/ai/select-algorithm-dotnet/appsettings.json
index 3b38afb..4e62a99 100644
--- a/ai/select-algorithm-dotnet/appsettings.json
+++ b/ai/select-algorithm-dotnet/appsettings.json
@@ -1,5 +1,5 @@
 {
-  "MongoDB": {
+  "DocumentDB": {
     "DatabaseName": "Hotels",
     "ClusterName": "<your-cluster-name>",
     "LoadBatchSize": 50
diff --git a/ai/select-algorithm-dotnet/quickstart.md b/ai/select-algorithm-dotnet/quickstart.md
index e94c2da..140a83d 100644
--- a/ai/select-algorithm-dotnet/quickstart.md
+++ b/ai/select-algorithm-dotnet/quickstart.md
@@ -149,7 +149,7 @@ Find the [sample code](https://github.com/Azure-Samples/documentdb-samples/tree/
    ```bash
    export AzureOpenAI__Endpoint="https://<your-resource>.openai.azure.com"
    export AzureOpenAI__EmbeddingModel="text-embedding-3-small"
-   export MongoDB__ClusterName="<your-cluster-name>"
+   export DocumentDB__ClusterName="<your-cluster-name>"
    export DataFiles__WithVectors="data/Hotels_Vector.json"
    export AZURE_TENANT_ID="<your-tenant-id>"
    ```
@@ -159,7 +159,7 @@ Find the [sample code](https://github.com/Azure-Samples/documentdb-samples/tree/
    ```powershell
    $env:AzureOpenAI__Endpoint="https://<your-resource>.openai.azure.com"
    $env:AzureOpenAI__EmbeddingModel="text-embedding-3-small"
-   $env:MongoDB__ClusterName="<your-cluster-name>"
+   $env:DocumentDB__ClusterName="<your-cluster-name>"
    $env:DataFiles__WithVectors="data/Hotels_Vector.json"
    $env:AZURE_TENANT_ID="<your-tenant-id>"
    ```
@@ -171,7 +171,7 @@ Find the [sample code](https://github.com/Azure-Samples/documentdb-samples/tree/
    - `<your-cluster-name>`: Your Azure DocumentDB cluster name
    - `<your-tenant-id>`: Your Microsoft Entra tenant ID
 
-   These environment variables override the matching values in `appsettings.json`. For example, `MongoDB__ClusterName` overrides `MongoDB:ClusterName` and `AzureOpenAI__Endpoint` overrides `AzureOpenAI:Endpoint`.
+   These environment variables override the matching values in `appsettings.json`. For example, `DocumentDB__ClusterName` overrides `DocumentDB:ClusterName` and `AzureOpenAI__Endpoint` overrides `AzureOpenAI:Endpoint`.
 
    You should always prefer passwordless authentication. For more information on setting up managed identity and the full range of your authentication options, see [Authenticate .NET apps to Azure services by using the Azure SDK for .NET](/dotnet/azure/sdk/authentication).
 
@@ -205,7 +205,7 @@ Find the [sample code](https://github.com/Azure-Samples/documentdb-samples/tree/
        "Endpoint": "https://<your-resource>.openai.azure.com",
        "EmbeddingModel": "text-embedding-3-small"
      },
-     "MongoDB": {
+     "DocumentDB": {
        "ClusterName": "<your-cluster-name>",
        "DatabaseName": "Hotels",
        "LoadBatchSize": 100
@@ -226,7 +226,7 @@ Find the [sample code](https://github.com/Azure-Samples/documentdb-samples/tree/
    }
    ```
 
-   You can keep placeholder values in `appsettings.json` and override them at runtime with environment variables such as `AzureOpenAI__Endpoint` and `MongoDB__ClusterName`.
+   You can keep placeholder values in `appsettings.json` and override them at runtime with environment variables such as `AzureOpenAI__Endpoint` and `DocumentDB__ClusterName`.
 
 ## Create code files
 

From 87cbef42378367b19b77487b853fe9c3a375c18d Mon Sep 17 00:00:00 2001
From: "Dina Berry (She/her)" <diberry@microsoft.com>
Date: Fri, 22 May 2026 08:13:42 -0700
Subject: [PATCH 05/10] Fix READMEs: data paths, connection string language,
 env var instructions

- All 5 READMEs: fix data copy to 'mkdir -p data && cp ../data/Hotels_Vector.json data/'
- Python, Java: replace 'connection strings' with 'configuration values'
- TypeScript: replace .env/.env.example instructions with shell export pattern,
  annotate SIMILARITY as unused in compare-all mode, fix step numbering (1-5)
- .NET: add note that azd is optional, users can edit appsettings.json directly
- appsettings.json: LoadBatchSize 50 -> 100 to match all other languages

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
---
 ai/select-algorithm-dotnet/README.md        |  8 +++----
 ai/select-algorithm-dotnet/appsettings.json |  2 +-
 ai/select-algorithm-go/README.md            |  4 ++--
 ai/select-algorithm-java/README.md          |  6 ++---
 ai/select-algorithm-python/README.md        |  6 ++---
 ai/select-algorithm-typescript/README.md    | 26 ++++++++++-----------
 6 files changed, 26 insertions(+), 26 deletions(-)

diff --git a/ai/select-algorithm-dotnet/README.md b/ai/select-algorithm-dotnet/README.md
index 32e25cf..abde5f6 100644
--- a/ai/select-algorithm-dotnet/README.md
+++ b/ai/select-algorithm-dotnet/README.md
@@ -19,13 +19,13 @@ Demonstrates three vector index algorithms available in Azure DocumentDB:
 
 1. **Configure environment:**
 
-   The .NET sample uses `appsettings.json` for configuration. After deploying with `azd up`, you can export values:
+   The .NET sample uses `appsettings.json` for configuration. You can set values directly in `appsettings.json`, or if you deployed with `azd up`, export your provisioned values first:
 
    ```bash
    azd env get-values
    ```
 
-   Then update `appsettings.json` with your Azure resource values.
+   > **Note:** `azd` is optional. You can skip it and edit `appsettings.json` manually with your Azure resource values.
 
 2. Edit `appsettings.json` with your configuration:
 
@@ -51,10 +51,10 @@ Demonstrates three vector index algorithms available in Azure DocumentDB:
    }
    ```
 
-3. Copy the shared data file:
+3. Copy the shared data file into the local `data/` directory:
 
    ```bash
-   cp data/Hotels_Vector.json .
+   mkdir -p data && cp ../data/Hotels_Vector.json data/
    ```
 
 4. Restore packages:
diff --git a/ai/select-algorithm-dotnet/appsettings.json b/ai/select-algorithm-dotnet/appsettings.json
index 4e62a99..472d7ff 100644
--- a/ai/select-algorithm-dotnet/appsettings.json
+++ b/ai/select-algorithm-dotnet/appsettings.json
@@ -2,7 +2,7 @@
   "DocumentDB": {
     "DatabaseName": "Hotels",
     "ClusterName": "<your-cluster-name>",
-    "LoadBatchSize": 50
+    "LoadBatchSize": 100
   },
   "VectorSearch": {
     "Similarity": "",
diff --git a/ai/select-algorithm-go/README.md b/ai/select-algorithm-go/README.md
index de14f79..452cde8 100644
--- a/ai/select-algorithm-go/README.md
+++ b/ai/select-algorithm-go/README.md
@@ -43,10 +43,10 @@ This sample demonstrates how to compare different vector search algorithms (IVF,
    EMBEDDING_DIMENSIONS=1536
    ```
 
-3. **Copy the shared data file** into this directory:
+3. **Copy the shared data file** into the local `data/` directory:
 
    ```bash
-   cp data/Hotels_Vector.json .
+   mkdir -p data && cp ../data/Hotels_Vector.json data/
    ```
 
    The `DATA_FILE_WITH_VECTORS` env var defaults to `data/Hotels_Vector.json`.
diff --git a/ai/select-algorithm-java/README.md b/ai/select-algorithm-java/README.md
index 119efb0..b73f613 100644
--- a/ai/select-algorithm-java/README.md
+++ b/ai/select-algorithm-java/README.md
@@ -20,7 +20,7 @@ This sample demonstrates how to compare all three vector search index algorithms
    azd env get-values > .env
    ```
 
-   This creates a `.env` file at the repository root with the connection strings and endpoints needed to run the sample.
+   This creates a `.env` file at the repository root with the configuration values and endpoints needed to run the sample.
 
    Alternatively, copy the example and fill in values manually:
 
@@ -34,10 +34,10 @@ This sample demonstrates how to compare all three vector search index algorithms
    - `AZURE_OPENAI_EMBEDDING_MODEL` — deployment name (e.g., `text-embedding-3-small`)
    - `DATA_FILE_WITH_VECTORS` — path to the pre-computed vectors JSON file
 
-3. Copy the shared data file:
+3. Copy the shared data file into the local `data/` directory:
 
    ```bash
-   cp data/Hotels_Vector.json .
+   mkdir -p data && cp ../data/Hotels_Vector.json data/
    ```
 
 ## Build
diff --git a/ai/select-algorithm-python/README.md b/ai/select-algorithm-python/README.md
index becef58..7c9ba5c 100644
--- a/ai/select-algorithm-python/README.md
+++ b/ai/select-algorithm-python/README.md
@@ -40,7 +40,7 @@ Compare IVF, HNSW, and DiskANN vector index algorithms in Azure DocumentDB. Each
    azd env get-values > .env
    ```
 
-   This creates a `.env` file at the repository root with the connection strings and endpoints needed to run the sample.
+   This creates a `.env` file at the repository root with the configuration values and endpoints needed to run the sample.
 
    Alternatively, copy the example and fill in values manually:
 
@@ -54,10 +54,10 @@ Compare IVF, HNSW, and DiskANN vector index algorithms in Azure DocumentDB. Each
    pip install -r ../requirements.txt
    ```
 
-3. Copy the shared data file:
+3. Copy the shared data file into the local `data/` directory:
 
    ```bash
-   cp data/Hotels_Vector.json .
+   mkdir -p data && cp ../data/Hotels_Vector.json data/
    ```
 
 4. Ensure you're logged in to Azure:
diff --git a/ai/select-algorithm-typescript/README.md b/ai/select-algorithm-typescript/README.md
index 254eee3..094bd4d 100644
--- a/ai/select-algorithm-typescript/README.md
+++ b/ai/select-algorithm-typescript/README.md
@@ -25,18 +25,18 @@ Compare IVF, HNSW, and DiskANN vector index algorithms in Azure DocumentDB using
 
 3. **Configure environment variables:**
 
-    After deploying with `azd up`, create a `.env` file with your provisioned resource values:
+    This sample reads configuration from environment variables. Export them in your shell before running:
 
     ```bash
-    azd env get-values > .env
-    ```
-
-    This creates a `.env` file in the project folder with the connection strings and endpoints needed to run the sample.
-
-    Alternatively, copy the example and fill in values manually:
-
-    ```bash
-    cp .env.example .env
+    export DOCUMENTDB_CLUSTER_NAME=your-cluster
+    export AZURE_OPENAI_EMBEDDING_ENDPOINT=https://your-resource.openai.azure.com
+    export AZURE_OPENAI_EMBEDDING_MODEL=text-embedding-3-small
+    export AZURE_OPENAI_EMBEDDING_API_VERSION=2023-05-15
+    export AZURE_DOCUMENTDB_DATABASENAME=Hotels
+    export DATA_FILE_WITH_VECTORS=data/Hotels_Vector.json
+    export EMBEDDED_FIELD=DescriptionVector
+    export EMBEDDING_DIMENSIONS=1536
+    export LOAD_SIZE_BATCH=100
     ```
 
     | Variable | Description |
@@ -50,9 +50,9 @@ Compare IVF, HNSW, and DiskANN vector index algorithms in Azure DocumentDB using
     | `EMBEDDED_FIELD` | Field name containing the vector (default: `DescriptionVector`) |
     | `EMBEDDING_DIMENSIONS` | Vector dimensions (default: `1536`) |
     | `LOAD_SIZE_BATCH` | Batch size for data insertion |
-    | `SIMILARITY` | Similarity metric: `COS`, `L2`, or `IP` |
+    | `SIMILARITY` | Not used in compare-all mode — all 3 similarity metrics (COS, L2, IP) are tested automatically |
 
-5. **Create a `data/` directory and copy the data file:**
+4. **Create a `data/` directory and copy the data file:**
 
     ```bash
     mkdir -p data
@@ -61,7 +61,7 @@ Compare IVF, HNSW, and DiskANN vector index algorithms in Azure DocumentDB using
 
     The `DATA_FILE_WITH_VECTORS` env var defaults to `data/Hotels_Vector.json`.
 
-6. **Build the project:**
+5. **Build the project:**
 
     ```bash
     npm run build

From feba4f3b5b4df91fb287a760341c283fb1f4dae2 Mon Sep 17 00:00:00 2001
From: "Dina Berry (She/her)" <diberry@microsoft.com>
Date: Fri, 22 May 2026 08:31:48 -0700
Subject: [PATCH 06/10] Fix READMEs: replace .env with shell exports, remove
 env:init script

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
---
 ai/select-algorithm-go/README.md            | 28 ++++++---------------
 ai/select-algorithm-java/README.md          | 26 +++++++------------
 ai/select-algorithm-python/README.md        | 18 ++++++-------
 ai/select-algorithm-typescript/package.json |  1 -
 4 files changed, 25 insertions(+), 48 deletions(-)

diff --git a/ai/select-algorithm-go/README.md b/ai/select-algorithm-go/README.md
index 452cde8..18d6f86 100644
--- a/ai/select-algorithm-go/README.md
+++ b/ai/select-algorithm-go/README.md
@@ -20,27 +20,16 @@ This sample demonstrates how to compare different vector search algorithms (IVF,
 
 2. **Configure environment variables:**
 
-   After deploying with `azd up`, create a `.env` file with your provisioned resource values:
+   After deploying with `azd up`, use the provisioned output values to set these required environment variables in your shell:
 
    ```bash
-   azd env get-values > .env
-   ```
-
-   Alternatively, copy the example and fill in values manually:
-
-   ```bash
-   cp .env.example .env
-   ```
-
-   Required variables:
-   ```env
-   DOCUMENTDB_CLUSTER_NAME=your-cluster-name
-   AZURE_OPENAI_EMBEDDING_ENDPOINT=https://your-resource.openai.azure.com
-   AZURE_OPENAI_EMBEDDING_MODEL=text-embedding-3-small
-   AZURE_DOCUMENTDB_DATABASENAME=Hotels
-   DATA_FILE_WITH_VECTORS=data/Hotels_Vector.json
-   EMBEDDED_FIELD=DescriptionVector
-   EMBEDDING_DIMENSIONS=1536
+   export DOCUMENTDB_CLUSTER_NAME=your-cluster-name
+   export AZURE_OPENAI_EMBEDDING_ENDPOINT=https://your-resource.openai.azure.com
+   export AZURE_OPENAI_EMBEDDING_MODEL=text-embedding-3-small
+   export AZURE_DOCUMENTDB_DATABASENAME=Hotels
+   export DATA_FILE_WITH_VECTORS=data/Hotels_Vector.json
+   export EMBEDDED_FIELD=DescriptionVector
+   export EMBEDDING_DIMENSIONS=1536
    ```
 
 3. **Copy the shared data file** into the local `data/` directory:
@@ -147,7 +136,6 @@ go run ./src/...
 
 ```
 select-algorithm-go/
-├── .env.example          # Environment variable template
 ├── go.mod                # Go module dependencies
 ├── go.sum                # Go module checksums
 ├── output/               # Sample output files
diff --git a/ai/select-algorithm-java/README.md b/ai/select-algorithm-java/README.md
index b73f613..e9daf45 100644
--- a/ai/select-algorithm-java/README.md
+++ b/ai/select-algorithm-java/README.md
@@ -14,27 +14,19 @@ This sample demonstrates how to compare all three vector search index algorithms
 
 1. ### Configure environment variables
 
-   After deploying with `azd up`, create a `.env` file with your provisioned resource values:
+   After deploying with `azd up`, use the provisioned output values to set the required environment variables in your terminal:
 
    ```bash
-   azd env get-values > .env
+   export DOCUMENTDB_CLUSTER_NAME=your-cluster-name
+   export AZURE_OPENAI_EMBEDDING_ENDPOINT=https://your-resource.openai.azure.com
+   export AZURE_OPENAI_EMBEDDING_MODEL=text-embedding-3-small
+   export AZURE_DOCUMENTDB_DATABASENAME=Hotels
+   export DATA_FILE_WITH_VECTORS=data/Hotels_Vector.json
+   export EMBEDDED_FIELD=DescriptionVector
+   export EMBEDDING_DIMENSIONS=1536
    ```
 
-   This creates a `.env` file at the repository root with the configuration values and endpoints needed to run the sample.
-
-   Alternatively, copy the example and fill in values manually:
-
-   ```bash
-   cp .env.example .env
-   ```
-
-2. Update `.env` with your Azure resource details (if not using `azd`):
-   - `DOCUMENTDB_CLUSTER_NAME` — your DocumentDB cluster name
-   - `AZURE_OPENAI_EMBEDDING_ENDPOINT` — your Azure OpenAI endpoint
-   - `AZURE_OPENAI_EMBEDDING_MODEL` — deployment name (e.g., `text-embedding-3-small`)
-   - `DATA_FILE_WITH_VECTORS` — path to the pre-computed vectors JSON file
-
-3. Copy the shared data file into the local `data/` directory:
+2. Copy the shared data file into the local `data/` directory:
 
    ```bash
    mkdir -p data && cp ../data/Hotels_Vector.json data/
diff --git a/ai/select-algorithm-python/README.md b/ai/select-algorithm-python/README.md
index 7c9ba5c..fc01802 100644
--- a/ai/select-algorithm-python/README.md
+++ b/ai/select-algorithm-python/README.md
@@ -34,18 +34,16 @@ Compare IVF, HNSW, and DiskANN vector index algorithms in Azure DocumentDB. Each
 
 1. ### Configure environment variables
 
-   After deploying with `azd up`, create a `.env` file with your provisioned resource values:
+   After deploying with `azd up`, use the provisioned output values to set the required environment variables in your terminal:
 
    ```bash
-   azd env get-values > .env
-   ```
-
-   This creates a `.env` file at the repository root with the configuration values and endpoints needed to run the sample.
-
-   Alternatively, copy the example and fill in values manually:
-
-   ```bash
-   cp .env.example .env
+   export DOCUMENTDB_CLUSTER_NAME=your-cluster-name
+   export AZURE_OPENAI_EMBEDDING_ENDPOINT=https://your-resource.openai.azure.com
+   export AZURE_OPENAI_EMBEDDING_MODEL=text-embedding-3-small
+   export AZURE_DOCUMENTDB_DATABASENAME=Hotels
+   export DATA_FILE_WITH_VECTORS=data/Hotels_Vector.json
+   export EMBEDDED_FIELD=DescriptionVector
+   export EMBEDDING_DIMENSIONS=1536
    ```
 
 2. Install dependencies:
diff --git a/ai/select-algorithm-typescript/package.json b/ai/select-algorithm-typescript/package.json
index 5c1f24a..8a9d155 100644
--- a/ai/select-algorithm-typescript/package.json
+++ b/ai/select-algorithm-typescript/package.json
@@ -4,7 +4,6 @@
   "description": "Compare IVF, HNSW, and DiskANN vector index algorithms in Azure DocumentDB",
   "type": "module",
   "scripts": {
-    "env:init": "azd env get-values > .env",
     "build": "tsc",
     "start": "node dist/compare-all.js"
   },

From 3fbb310792f1ddc1ab83aaee5727d579e8a0b203 Mon Sep 17 00:00:00 2001
From: "Dina Berry (She/her)" <diberry@microsoft.com>
Date: Fri, 22 May 2026 08:42:10 -0700
Subject: [PATCH 07/10] Fix .NET README env var docs, remove misleading
 algorithm env knobs

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
---
 ai/select-algorithm-dotnet/README.md     | 17 ++++++++++++++++-
 ai/select-algorithm-java/README.md       | 12 +++---------
 ai/select-algorithm-python/README.md     |  4 ++--
 ai/select-algorithm-typescript/README.md |  5 +++--
 4 files changed, 24 insertions(+), 14 deletions(-)

diff --git a/ai/select-algorithm-dotnet/README.md b/ai/select-algorithm-dotnet/README.md
index abde5f6..01d3345 100644
--- a/ai/select-algorithm-dotnet/README.md
+++ b/ai/select-algorithm-dotnet/README.md
@@ -51,13 +51,28 @@ Demonstrates three vector index algorithms available in Azure DocumentDB:
    }
    ```
 
+   > **Note:** .NET configuration also supports environment variable overrides. Use double-underscore (`__`) as the hierarchy separator:
+   > ```bash
+   > export DocumentDB__ClusterName=your-cluster-name
+   > export DocumentDB__DatabaseName=Hotels
+   > export AzureOpenAI__Endpoint=https://your-resource.openai.azure.com
+   > export AzureOpenAI__EmbeddingModel=text-embedding-3-small
+   > ```
+   > Environment variables take precedence over `appsettings.json` values.
+
 3. Copy the shared data file into the local `data/` directory:
 
    ```bash
    mkdir -p data && cp ../data/Hotels_Vector.json data/
    ```
 
-4. Restore packages:
+4. Sign in to Azure for passwordless authentication:
+
+   ```bash
+   az login
+   ```
+
+5. Restore packages:
 
    ```bash
    dotnet restore
diff --git a/ai/select-algorithm-java/README.md b/ai/select-algorithm-java/README.md
index e9daf45..85361f7 100644
--- a/ai/select-algorithm-java/README.md
+++ b/ai/select-algorithm-java/README.md
@@ -46,16 +46,10 @@ Compare all 9 algorithm × similarity combinations:
 mvn exec:java -Pcompare
 ```
 
-Or via the `ALGORITHM` environment variable:
-
-```bash
-ALGORITHM=compare mvn exec:java
-```
-
 On Windows (PowerShell):
 
 ```powershell
-$env:ALGORITHM="compare"; mvn exec:java
+mvn exec:java -Pcompare
 ```
 
 ## Algorithms
@@ -79,12 +73,12 @@ $env:ALGORITHM="compare"; mvn exec:java
 | `AZURE_DOCUMENTDB_DATABASENAME` | `Hotels` | Target database name |
 | `LOAD_SIZE_BATCH` | `100` | Batch size for data loading |
 | `EMBEDDING_SIZE_BATCH` | `16` | Batch size for embedding requests |
-| `ALGORITHM` | (empty = all) | Which algorithm to run |
-| `SIMILARITY` | (empty = all) | Similarity metric: `COS`, `L2`, `IP` |
 | `QUERY_TEXT` | `luxury hotel near the beach` | Search query text |
 | `TOP_K` | `5` | Number of results per search |
 | `VERBOSE` | `false` | Print detailed per-index results |
 
+`CompareAll.java` always runs all 9 algorithm/metric combinations. It does not read `ALGORITHM` or `SIMILARITY` environment variables.
+
 ## Authentication
 
 This sample uses **passwordless authentication** via `DefaultAzureCredential`:
diff --git a/ai/select-algorithm-python/README.md b/ai/select-algorithm-python/README.md
index fc01802..e867fbc 100644
--- a/ai/select-algorithm-python/README.md
+++ b/ai/select-algorithm-python/README.md
@@ -87,8 +87,8 @@ The script creates a single `hotels` collection, loads data once, then for each
 | `AZURE_DOCUMENTDB_DATABASENAME` | `Hotels` | Target database name |
 | `LOAD_SIZE_BATCH` | `100` | Batch size for data loading |
 | `EMBEDDING_SIZE_BATCH` | `16` | Batch size for embedding requests |
-| `ALGORITHM` | (empty = all) | Which algorithm to run |
-| `SIMILARITY` | (empty = all) | Similarity metric: `COS`, `L2`, `IP` |
 | `QUERY_TEXT` | `luxury hotel near the beach` | Search query text |
 | `TOP_K` | `5` | Number of results per search |
 | `VERBOSE` | `false` | Show all k results per combo |
+
+`compare_all.py` always runs all 9 algorithm/metric combinations. It does not read `ALGORITHM` or `SIMILARITY` environment variables.
diff --git a/ai/select-algorithm-typescript/README.md b/ai/select-algorithm-typescript/README.md
index 094bd4d..302df29 100644
--- a/ai/select-algorithm-typescript/README.md
+++ b/ai/select-algorithm-typescript/README.md
@@ -50,9 +50,10 @@ Compare IVF, HNSW, and DiskANN vector index algorithms in Azure DocumentDB using
     | `EMBEDDED_FIELD` | Field name containing the vector (default: `DescriptionVector`) |
     | `EMBEDDING_DIMENSIONS` | Vector dimensions (default: `1536`) |
     | `LOAD_SIZE_BATCH` | Batch size for data insertion |
-    | `SIMILARITY` | Not used in compare-all mode — all 3 similarity metrics (COS, L2, IP) are tested automatically |
 
-4. **Create a `data/` directory and copy the data file:**
+    `compare-all.ts` always runs all 9 algorithm/metric combinations. It does not read `ALGORITHM` or `SIMILARITY` environment variables.
+
+    4. **Create a `data/` directory and copy the data file:**
 
     ```bash
     mkdir -p data

From 912b511afb77078f0c1d76754b9e8a0c46217b65 Mon Sep 17 00:00:00 2001
From: "Dina Berry (She/her)" <diberry@microsoft.com>
Date: Fri, 22 May 2026 08:58:47 -0700
Subject: [PATCH 08/10] Fix README phantom config docs, align .NET dimensions
 with appsettings

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
---
 ai/select-algorithm-dotnet/CompareAll.cs    | 10 +++++-----
 ai/select-algorithm-dotnet/README.md        |  7 ++-----
 ai/select-algorithm-dotnet/appsettings.json |  3 +--
 ai/select-algorithm-java/README.md          |  5 +----
 ai/select-algorithm-python/README.md        |  5 ++---
 5 files changed, 11 insertions(+), 19 deletions(-)

diff --git a/ai/select-algorithm-dotnet/CompareAll.cs b/ai/select-algorithm-dotnet/CompareAll.cs
index 2cf3d10..562d8c1 100644
--- a/ai/select-algorithm-dotnet/CompareAll.cs
+++ b/ai/select-algorithm-dotnet/CompareAll.cs
@@ -65,7 +65,7 @@ public static void Run(AppConfiguration appConfig)
             Console.WriteLine("Embedding generated (reused for all searches)\n");
 
             // Define 9 index configurations
-            var configs = BuildIndexConfigs(dimensions);
+            var configs = BuildIndexConfigs();
 
             // Run each config sequentially: drop→create→wait→search
             // DocumentDB doesn't allow multiple vector indexes of the same kind on the same field
@@ -77,7 +77,7 @@ public static void Run(AppConfiguration appConfig)
                 DropVectorIndexes(collection, vectorField);
 
                 // 2. Create this specific index
-                CreateIndex(collection, vectorField, config);
+                CreateIndex(collection, vectorField, dimensions, config);
                 Console.WriteLine($"  ✓ {config.Name} created");
 
                 // 3. Search with retries while the index becomes available
@@ -139,7 +139,7 @@ public static void Run(AppConfiguration appConfig)
         }
     }
 
-    private static List<IndexConfig> BuildIndexConfigs(int dimensions)
+    private static List<IndexConfig> BuildIndexConfigs()
     {
         string[] metrics = ["COS", "L2", "IP"];
         var configs = new List<IndexConfig>();
@@ -177,7 +177,7 @@ private static void DropVectorIndexes(IMongoCollection<BsonDocument> collection,
         catch { }
     }
 
-    private static void CreateIndex(IMongoCollection<BsonDocument> collection, string vectorField, IndexConfig config)
+    private static void CreateIndex(IMongoCollection<BsonDocument> collection, string vectorField, int dimensions, IndexConfig config)
     {
         // Drop existing index with same name if present
         try
@@ -192,7 +192,7 @@ private static void CreateIndex(IMongoCollection<BsonDocument> collection, strin
         var cosmosSearchOptions = new BsonDocument
         {
             { "kind", config.Kind },
-            { "dimensions", int.Parse(Environment.GetEnvironmentVariable("EMBEDDING_DIMENSIONS") ?? "1536") },
+            { "dimensions", dimensions },
             { "similarity", config.Similarity }
         };
 
diff --git a/ai/select-algorithm-dotnet/README.md b/ai/select-algorithm-dotnet/README.md
index 01d3345..b750f20 100644
--- a/ai/select-algorithm-dotnet/README.md
+++ b/ai/select-algorithm-dotnet/README.md
@@ -42,8 +42,7 @@ Demonstrates three vector index algorithms available in Azure DocumentDB:
      },
      "Embedding": {
        "EmbeddedField": "DescriptionVector",
-       "Dimensions": 1536,
-       "EmbeddingSizeBatch": 16
+       "Dimensions": 1536
      },
      "DataFiles": {
        "WithVectors": "data/Hotels_Vector.json"
@@ -92,13 +91,12 @@ dotnet run
 |---------------------------|---------|-------------|
 | `DocumentDB:ClusterName` | (required) | DocumentDB cluster name |
 | `AzureOpenAI:Endpoint` | (required) | Azure OpenAI endpoint |
-| `AzureOpenAI:EmbeddingModel` | (required) | Embedding model deployment name |
+| `AzureOpenAI:EmbeddingModel` | `text-embedding-3-small` | Embedding model deployment name |
 | `DataFiles:WithVectors` | `data/Hotels_Vector.json` | Path to vectors JSON file |
 | `Embedding:EmbeddedField` | `DescriptionVector` | Field name containing embeddings |
 | `Embedding:Dimensions` | `1536` | Vector dimensions |
 | `DocumentDB:DatabaseName` | `Hotels` | Target database name |
 | `DocumentDB:LoadBatchSize` | `100` | Batch size for data loading |
-| `Embedding:EmbeddingSizeBatch` | `16` | Batch size for embedding requests |
 
 **Additional environment variables for compare mode:**
 
@@ -106,7 +104,6 @@ dotnet run
 |----------|---------|-------------|
 | `QUERY_TEXT` | `luxury hotel near the beach` | Search query text |
 | `TOP_K` | `5` | Number of results per search |
-| `VERBOSE` | `false` | Show detailed per-result output |
 
 ## How It Works
 
diff --git a/ai/select-algorithm-dotnet/appsettings.json b/ai/select-algorithm-dotnet/appsettings.json
index 472d7ff..b6634fb 100644
--- a/ai/select-algorithm-dotnet/appsettings.json
+++ b/ai/select-algorithm-dotnet/appsettings.json
@@ -18,7 +18,6 @@
   },
   "Embedding": {
     "EmbeddedField": "DescriptionVector",
-    "Dimensions": 1536,
-    "EmbeddingSizeBatch": 16
+    "Dimensions": 1536
   }
 }
diff --git a/ai/select-algorithm-java/README.md b/ai/select-algorithm-java/README.md
index 85361f7..ca2648c 100644
--- a/ai/select-algorithm-java/README.md
+++ b/ai/select-algorithm-java/README.md
@@ -66,16 +66,13 @@ mvn exec:java -Pcompare
 |----------|---------|-------------|
 | `DOCUMENTDB_CLUSTER_NAME` | (required) | DocumentDB cluster name |
 | `AZURE_OPENAI_EMBEDDING_ENDPOINT` | (required) | Azure OpenAI endpoint |
-| `AZURE_OPENAI_EMBEDDING_MODEL` | (required) | Embedding model deployment name |
+| `AZURE_OPENAI_EMBEDDING_MODEL` | `text-embedding-3-small` | Embedding model deployment name |
 | `DATA_FILE_WITH_VECTORS` | `data/Hotels_Vector.json` | Path to vectors JSON file |
 | `EMBEDDED_FIELD` | `DescriptionVector` | Field name containing embeddings |
 | `EMBEDDING_DIMENSIONS` | `1536` | Vector dimensions |
 | `AZURE_DOCUMENTDB_DATABASENAME` | `Hotels` | Target database name |
-| `LOAD_SIZE_BATCH` | `100` | Batch size for data loading |
-| `EMBEDDING_SIZE_BATCH` | `16` | Batch size for embedding requests |
 | `QUERY_TEXT` | `luxury hotel near the beach` | Search query text |
 | `TOP_K` | `5` | Number of results per search |
-| `VERBOSE` | `false` | Print detailed per-index results |
 
 `CompareAll.java` always runs all 9 algorithm/metric combinations. It does not read `ALGORITHM` or `SIMILARITY` environment variables.
 
diff --git a/ai/select-algorithm-python/README.md b/ai/select-algorithm-python/README.md
index e867fbc..8d5da02 100644
--- a/ai/select-algorithm-python/README.md
+++ b/ai/select-algorithm-python/README.md
@@ -80,15 +80,14 @@ The script creates a single `hotels` collection, loads data once, then for each
 |----------|---------|-------------|
 | `DOCUMENTDB_CLUSTER_NAME` | (required) | DocumentDB cluster name |
 | `AZURE_OPENAI_EMBEDDING_ENDPOINT` | (required) | Azure OpenAI endpoint |
-| `AZURE_OPENAI_EMBEDDING_MODEL` | (required) | Embedding model deployment name |
+| `AZURE_OPENAI_EMBEDDING_MODEL` | `text-embedding-3-small` | Embedding model deployment name |
+| `AZURE_OPENAI_EMBEDDING_API_VERSION` | `2023-05-15` | Azure OpenAI API version |
 | `DATA_FILE_WITH_VECTORS` | `data/Hotels_Vector.json` | Path to vectors JSON file |
 | `EMBEDDED_FIELD` | `DescriptionVector` | Field name containing embeddings |
 | `EMBEDDING_DIMENSIONS` | `1536` | Vector dimensions |
 | `AZURE_DOCUMENTDB_DATABASENAME` | `Hotels` | Target database name |
 | `LOAD_SIZE_BATCH` | `100` | Batch size for data loading |
-| `EMBEDDING_SIZE_BATCH` | `16` | Batch size for embedding requests |
 | `QUERY_TEXT` | `luxury hotel near the beach` | Search query text |
 | `TOP_K` | `5` | Number of results per search |
-| `VERBOSE` | `false` | Show all k results per combo |
 
 `compare_all.py` always runs all 9 algorithm/metric combinations. It does not read `ALGORITHM` or `SIMILARITY` environment variables.

From 2cfa69d88a60e65a70e83d59210690f712959836 Mon Sep 17 00:00:00 2001
From: "Dina Berry (She/her)" <diberry@microsoft.com>
Date: Fri, 22 May 2026 09:10:02 -0700
Subject: [PATCH 09/10] Fix Java batch size: read LOAD_SIZE_BATCH env var
 (cross-language consistency)

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
---
 .../java/com/azure/documentdb/selectalgorithm/CompareAll.java  | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/ai/select-algorithm-java/src/main/java/com/azure/documentdb/selectalgorithm/CompareAll.java b/ai/select-algorithm-java/src/main/java/com/azure/documentdb/selectalgorithm/CompareAll.java
index 2d22e64..ee8733b 100644
--- a/ai/select-algorithm-java/src/main/java/com/azure/documentdb/selectalgorithm/CompareAll.java
+++ b/ai/select-algorithm-java/src/main/java/com/azure/documentdb/selectalgorithm/CompareAll.java
@@ -55,7 +55,8 @@ public static void run() {
 
             collection.drop();
             System.out.println("  Collection reset.");
-            Utils.insertData(collection, data, 100);
+            int batchSize = Integer.parseInt(System.getenv().getOrDefault("LOAD_SIZE_BATCH", "100"));
+            Utils.insertData(collection, data, batchSize);
 
             // Generate ONE embedding for the query (reused for all 9 searches)
             OpenAIClient aiClient = Utils.getOpenAIClient();

From 6394c978eb31bbb1e7a9303236537d8b013df548 Mon Sep 17 00:00:00 2001
From: "Dina Berry (She/her)" <diberry@microsoft.com>
Date: Fri, 22 May 2026 09:18:39 -0700
Subject: [PATCH 10/10] Fix Python README: run from sample root (consistent
 with other languages)

Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
---
 ai/select-algorithm-python/README.md | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

diff --git a/ai/select-algorithm-python/README.md b/ai/select-algorithm-python/README.md
index 8d5da02..b89a800 100644
--- a/ai/select-algorithm-python/README.md
+++ b/ai/select-algorithm-python/README.md
@@ -48,8 +48,7 @@ Compare IVF, HNSW, and DiskANN vector index algorithms in Azure DocumentDB. Each
 
 2. Install dependencies:
    ```bash
-   cd src
-   pip install -r ../requirements.txt
+   pip install -r requirements.txt
    ```
 
 3. Copy the shared data file into the local `data/` directory:
@@ -68,8 +67,7 @@ Compare IVF, HNSW, and DiskANN vector index algorithms in Azure DocumentDB. Each
 Compare all 9 combinations (3 algorithms × 3 similarity metrics) in a single invocation:
 
 ```bash
-cd src
-python compare_all.py
+python src/compare_all.py
 ```
 
 The script creates a single `hotels` collection, loads data once, then for each of the 9 algorithm/metric combinations: creates the index → searches → drops the index. DocumentDB only allows one vector index per kind per field, so indexes are created sequentially.