Merge branch 'main' into users/markwallace/codeql-actions-v3

.github/workflows/python-test-coverage.yml

@@ -31,15 +31,15 @@ jobs:
       - name: Setup filename variables
         run: echo "FILE_ID=${{ github.event.number }}" >> $GITHUB_ENV
       - name: Download coverage
-        uses: dawidd6/action-download-artifact@v3
+        uses: dawidd6/action-download-artifact@v6
         with:
           name: python-coverage-${{ env.FILE_ID }}.txt
           github_token: ${{ secrets.GH_ACTIONS_PR_WRITE }}
           workflow: python-unit-tests.yml
           search_artifacts: true
           if_no_artifact_found: warn
       - name: Download pytest
-        uses: dawidd6/action-download-artifact@v3
+        uses: dawidd6/action-download-artifact@v6
         with:
           name: pytest-${{ env.FILE_ID }}.xml
           github_token: ${{ secrets.GH_ACTIONS_PR_WRITE }}

dotnet/Directory.Packages.props

@@ -81,8 +81,8 @@
     <PackageVersion Include="Microsoft.Extensions.Logging.Debug" Version="8.0.0" />
     <PackageVersion Include="Microsoft.Extensions.Options.DataAnnotations" Version="8.0.0" />
     <PackageVersion Include="Microsoft.Extensions.TimeProvider.Testing" Version="8.10.0" />
-    <PackageVersion Include="Microsoft.Extensions.FileProviders.Physical" Version="[8.0.0, 9.0.0)" />
-    <PackageVersion Include="Microsoft.Extensions.FileProviders.Embedded" Version="[8.0.0, 9.0.0)" />
+    <PackageVersion Include="Microsoft.Extensions.FileProviders.Physical" Version="8.0.0" />
+    <PackageVersion Include="Microsoft.Extensions.FileProviders.Embedded" Version="8.0.0" />
     <!-- Test -->
     <PackageVersion Include="Microsoft.NET.Test.Sdk" Version="17.11.1" />
     <PackageVersion Include="Moq" Version="[4.18.4]" />
@@ -102,13 +102,13 @@
     <PackageVersion Include="DuckDB.NET.Data.Full" Version="1.1.2.1" />
     <PackageVersion Include="DuckDB.NET.Data" Version="1.1.2.1" />
     <PackageVersion Include="MongoDB.Driver" Version="2.30.0" />
-    <PackageVersion Include="Microsoft.Graph" Version="[4.51.0, 5)" />
-    <PackageVersion Include="Microsoft.Identity.Client.Extensions.Msal" Version="[2.28.0, )" />
+    <PackageVersion Include="Microsoft.Graph" Version="4.51.0" />
+    <PackageVersion Include="Microsoft.Identity.Client.Extensions.Msal" Version="2.28.0" />
     <PackageVersion Include="Microsoft.OpenApi" Version="1.6.22" />
     <PackageVersion Include="Microsoft.OpenApi.Readers" Version="1.6.22" />
     <PackageVersion Include="Microsoft.OpenApi.ApiManifest" Version="0.5.6-preview" />
     <PackageVersion Include="Microsoft.Plugins.Manifest" Version="1.0.0-rc2" />
-    <PackageVersion Include="Google.Apis.CustomSearchAPI.v1" Version="[1.60.0.3001, )" />
+    <PackageVersion Include="Google.Apis.CustomSearchAPI.v1" Version="1.60.0.3001" />
     <PackageVersion Include="Grpc.Net.Client" Version="2.66.0" />
     <PackageVersion Include="protobuf-net" Version="3.2.45" />
     <PackageVersion Include="protobuf-net.Reflection" Version="3.2.12" />

dotnet/src/Connectors/Connectors.AzureOpenAI.UnitTests/Services/AzureOpenAIChatCompletionServiceTests.cs

@@ -1541,6 +1541,10 @@ public async Task GetStreamingChatMessageContentsWithFunctionCallAndEmptyArgumen
         { "V2024_10_01_PREVIEW", "2024-10-01-preview" },
         { "2024_10_01_Preview", "2024-10-01-preview" },
         { "2024-10-01-preview", "2024-10-01-preview" },
+        { "V2024_09_01_preview", "2024-09-01-preview" },
+        { "V2024_09_01_PREVIEW", "2024-09-01-preview" },
+        { "2024_09_01_Preview", "2024-09-01-preview" },
+        { "2024-09-01-preview", "2024-09-01-preview" },
         { "V2024_08_01_preview", "2024-08-01-preview" },
         { "V2024_08_01_PREVIEW", "2024-08-01-preview" },
         { "2024_08_01_Preview", "2024-08-01-preview" },
@@ -1549,6 +1553,7 @@ public async Task GetStreamingChatMessageContentsWithFunctionCallAndEmptyArgumen
         { "2024_06_01", "2024-06-01" },
         { "2024-06-01", "2024-06-01" },
         { AzureOpenAIClientOptions.ServiceVersion.V2024_10_01_Preview.ToString(), null },
+        { AzureOpenAIClientOptions.ServiceVersion.V2024_09_01_Preview.ToString(), null },
         { AzureOpenAIClientOptions.ServiceVersion.V2024_08_01_Preview.ToString(), null },
         { AzureOpenAIClientOptions.ServiceVersion.V2024_06_01.ToString(), null }
     };

dotnet/src/Connectors/Connectors.AzureOpenAI/Core/AzureClientCore.cs

@@ -136,7 +136,9 @@ internal static AzureOpenAIClientOptions GetAzureOpenAIClientOptions(HttpClient?
             {
                 "2024-06-01" or "V2024_06_01" or "2024_06_01" => AzureOpenAIClientOptions.ServiceVersion.V2024_06_01,
                 "2024-08-01-PREVIEW" or "V2024_08_01_PREVIEW" or "2024_08_01_PREVIEW" => AzureOpenAIClientOptions.ServiceVersion.V2024_08_01_Preview,
+                "2024-09-01-PREVIEW" or "V2024_09_01_PREVIEW" or "2024_09_01_PREVIEW" => AzureOpenAIClientOptions.ServiceVersion.V2024_09_01_Preview,
                 "2024-10-01-PREVIEW" or "V2024_10_01_PREVIEW" or "2024_10_01_PREVIEW" => AzureOpenAIClientOptions.ServiceVersion.V2024_10_01_Preview,
+
                 _ => throw new NotSupportedException($"The service version '{serviceVersion}' is not supported.")
             };
         }

dotnet/src/Connectors/Connectors.Memory.Kusto/KustoMemoryStore.cs

@@ -374,7 +374,7 @@ private static string GetCollectionName(string tableName)
     /// Returns base Kusto query.
     /// </summary>
     /// <remarks>
-    /// Kusto is an append-only store. Although deletions are possible, they are highly discourged,
+    /// Kusto is an append-only store. Although deletions are possible, they are highly discouraged,
     /// and should only be used in rare cases (see: https://learn.microsoft.com/en-us/azure/data-explorer/kusto/concepts/data-soft-delete#use-cases).
     /// As such, the recommended approach for dealing with row updates is versioning.
     /// An easy way to achieve this is by using the ingestion time of the record (insertion time).

dotnet/src/Connectors/Connectors.Memory.Qdrant/QdrantVectorStoreRecordCollection.cs

@@ -175,7 +175,20 @@ await this.RunOperationAsync(
         foreach (var dataProperty in dataProperties)
         {
             var storageFieldName = this._propertyReader.GetStoragePropertyName(dataProperty.DataModelPropertyName);
-            var schemaType = QdrantVectorStoreCollectionCreateMapping.s_schemaTypeMap[dataProperty.PropertyType!];
+
+            if (QdrantVectorStoreCollectionCreateMapping.s_schemaTypeMap.TryGetValue(dataProperty.PropertyType!, out PayloadSchemaType schemaType))
+            {
+                // Do nothing since schemaType is already set.
+            }
+            else if (VectorStoreRecordPropertyVerification.IsSupportedEnumerableType(dataProperty.PropertyType) && VectorStoreRecordPropertyVerification.GetCollectionElementType(dataProperty.PropertyType) == typeof(string))
+            {
+                // For enumerable of strings, use keyword schema type, since this allows tag filtering.
+                schemaType = PayloadSchemaType.Keyword;
+            }
+            else
+            {
+                throw new InvalidOperationException($"Property {nameof(VectorStoreRecordDataProperty.IsFilterable)} on {nameof(VectorStoreRecordDataProperty)} '{dataProperty.DataModelPropertyName}' is set to true, but the property type is not supported for filtering. The Qdrant VectorStore supports filtering on {string.Join(", ", QdrantVectorStoreCollectionCreateMapping.s_schemaTypeMap.Keys.Select(x => x.Name))} properties only.");
+            }
 
             await this.RunOperationAsync(
                 "CreatePayloadIndex",

dotnet/src/IntegrationTests/Connectors/Memory/Qdrant/QdrantVectorStoreFixture.cs

@@ -57,7 +57,7 @@ public QdrantVectorStoreFixture()
                 new VectorStoreRecordDataProperty("HotelCode", typeof(int)) { IsFilterable = true },
                 new VectorStoreRecordDataProperty("ParkingIncluded", typeof(bool)) { IsFilterable = true, StoragePropertyName = "parking_is_included" },
                 new VectorStoreRecordDataProperty("HotelRating", typeof(float)) { IsFilterable = true },
-                new VectorStoreRecordDataProperty("Tags", typeof(List<string>)),
+                new VectorStoreRecordDataProperty("Tags", typeof(List<string>)) { IsFilterable = true },
                 new VectorStoreRecordDataProperty("Description", typeof(string)),
                 new VectorStoreRecordVectorProperty("DescriptionEmbedding", typeof(ReadOnlyMemory<float>?)) { Dimensions = VectorDimensions, DistanceFunction = DistanceFunction.ManhattanDistance }
             }
@@ -146,11 +146,17 @@ await this.QdrantClient.CreateCollectionAsync(
 
         // Create test data common to both named and unnamed vectors.
         var tags = new ListValue();
-        tags.Values.Add("t1");
-        tags.Values.Add("t2");
+        tags.Values.Add("t11.1");
+        tags.Values.Add("t11.2");
         var tagsValue = new Value();
         tagsValue.ListValue = tags;
 
+        var tags2 = new ListValue();
+        tags2.Values.Add("t13.1");
+        tags2.Values.Add("t13.2");
+        var tagsValue2 = new Value();
+        tagsValue2.ListValue = tags2;
+
         // Create some test data using named vectors.
         var embedding = await this.EmbeddingGenerator.GenerateEmbeddingAsync("This is a great hotel.");
         var embeddingArray = embedding.ToArray();
@@ -183,7 +189,7 @@ await this.QdrantClient.CreateCollectionAsync(
             {
                 Id = 13,
                 Vectors = new Vectors { Vectors_ = namedVectors3 },
-                Payload = { ["HotelName"] = "My Hotel 13", ["HotelCode"] = 13, ["parking_is_included"] = false, ["Description"] = "This is a great hotel." }
+                Payload = { ["HotelName"] = "My Hotel 13", ["HotelCode"] = 13, ["parking_is_included"] = false, ["Tags"] = tagsValue2, ["Description"] = "This is a great hotel." }
             },
             new PointStruct
             {
@@ -214,7 +220,7 @@ await this.QdrantClient.CreateCollectionAsync(
             {
                 Id = 13,
                 Vectors = embeddingArray,
-                Payload = { ["HotelName"] = "My Hotel 13", ["HotelCode"] = 13, ["parking_is_included"] = false, ["Description"] = "This is a great hotel." }
+                Payload = { ["HotelName"] = "My Hotel 13", ["HotelCode"] = 13, ["parking_is_included"] = false, ["Tags"] = tagsValue2, ["Description"] = "This is a great hotel." }
             },
         ];
 
@@ -327,7 +333,7 @@ public record HotelInfo()
         [VectorStoreRecordData(IsFilterable = true, StoragePropertyName = "parking_is_included")]
         public bool ParkingIncluded { get; set; }
 
-        [VectorStoreRecordData]
+        [VectorStoreRecordData(IsFilterable = true)]
         public List<string> Tags { get; set; } = new List<string>();
 
         /// <summary>A data field.</summary>

dotnet/src/IntegrationTests/Connectors/Memory/Qdrant/QdrantVectorStoreRecordCollectionTests.cs

@@ -66,7 +66,7 @@ public async Task ItCanCreateACollectionUpsertGetAndSearchAsync(bool hasNamedVec
         var vector = await fixture.EmbeddingGenerator.GenerateEmbeddingAsync("A great hotel");
         var actual = await sut.VectorizedSearchAsync(
             vector,
-            new VectorSearchOptions { Filter = new VectorSearchFilter().EqualTo("HotelCode", 30) });
+            new VectorSearchOptions { Filter = new VectorSearchFilter().EqualTo("HotelCode", 30).AnyTagEqualTo("Tags", "t2") });
 
         // Assert
         var collectionExistResult = await sut.CollectionExistsAsync();
@@ -221,8 +221,8 @@ public async Task ItCanGetDocumentFromVectorStoreAsync(bool useRecordDefinition,
         Assert.True(getResult?.ParkingIncluded);
         Assert.Equal(4.5f, getResult?.HotelRating);
         Assert.Equal(2, getResult?.Tags.Count);
-        Assert.Equal("t1", getResult?.Tags[0]);
-        Assert.Equal("t2", getResult?.Tags[1]);
+        Assert.Equal("t11.1", getResult?.Tags[0]);
+        Assert.Equal("t11.2", getResult?.Tags[1]);
         Assert.Equal("This is a great hotel.", getResult?.Description);
         if (withEmbeddings)
         {
@@ -389,7 +389,7 @@ public async Task ItCanSearchWithFilterAsync(bool useRecordDefinition, string co
 
         // Act.
         var vector = await fixture.EmbeddingGenerator.GenerateEmbeddingAsync("A great hotel");
-        var filter = filterType == "equality" ? new VectorSearchFilter().EqualTo("HotelName", "My Hotel 11") : new VectorSearchFilter().AnyTagEqualTo("Tags", "t1");
+        var filter = filterType == "equality" ? new VectorSearchFilter().EqualTo("HotelName", "My Hotel 13") : new VectorSearchFilter().AnyTagEqualTo("Tags", "t13.2");
         var actual = await sut.VectorizedSearchAsync(
             vector,
             new()
@@ -402,12 +402,11 @@ public async Task ItCanSearchWithFilterAsync(bool useRecordDefinition, string co
         Assert.Single(searchResults);
 
         var searchResultRecord = searchResults.First().Record;
-        Assert.Equal(11ul, searchResultRecord?.HotelId);
-        Assert.Equal("My Hotel 11", searchResultRecord?.HotelName);
-        Assert.Equal(11, searchResultRecord?.HotelCode);
-        Assert.Equal(4.5f, searchResultRecord?.HotelRating);
-        Assert.Equal(true, searchResultRecord?.ParkingIncluded);
-        Assert.Equal(new string[] { "t1", "t2" }, searchResultRecord?.Tags.ToArray());
+        Assert.Equal(13ul, searchResultRecord?.HotelId);
+        Assert.Equal("My Hotel 13", searchResultRecord?.HotelName);
+        Assert.Equal(13, searchResultRecord?.HotelCode);
+        Assert.Equal(false, searchResultRecord?.ParkingIncluded);
+        Assert.Equal(new string[] { "t13.1", "t13.2" }, searchResultRecord?.Tags.ToArray());
         Assert.Equal("This is a great hotel.", searchResultRecord?.Description);
     }
 
@@ -448,7 +447,7 @@ public async Task ItCanUpsertAndRetrieveUsingTheGenericMapperAsync()
         Assert.Equal(11, baseSetGetResult.Data["HotelCode"]);
         Assert.True((bool)baseSetGetResult.Data["ParkingIncluded"]!);
         Assert.Equal(4.5f, baseSetGetResult.Data["HotelRating"]);
-        Assert.Equal(new[] { "t1", "t2" }, ((List<string>)baseSetGetResult.Data["Tags"]!).ToArray());
+        Assert.Equal(new[] { "t11.1", "t11.2" }, ((List<string>)baseSetGetResult.Data["Tags"]!).ToArray());
         Assert.Equal("This is a great hotel.", baseSetGetResult.Data["Description"]);
         Assert.NotNull(baseSetGetResult.Vectors["DescriptionEmbedding"]);
         Assert.IsType<ReadOnlyMemory<float>>(baseSetGetResult.Vectors["DescriptionEmbedding"]);

python/semantic_kernel/connectors/ai/open_ai/services/azure_chat_completion.py

@@ -2,18 +2,12 @@
 
 import json
 import logging
-import sys
-from collections.abc import AsyncGenerator, Mapping
+from collections.abc import Mapping
 from copy import deepcopy
 from typing import Any, TypeVar
 from uuid import uuid4
 
-if sys.version_info >= (3, 12):
-    from typing import override  # pragma: no cover
-else:
-    from typing_extensions import override  # pragma: no cover
-
-from openai import AsyncAzureOpenAI, AsyncStream
+from openai import AsyncAzureOpenAI
 from openai.lib.azure import AsyncAzureADTokenProvider
 from openai.types.chat.chat_completion import ChatCompletion, Choice
 from openai.types.chat.chat_completion_chunk import ChatCompletionChunk
@@ -23,24 +17,19 @@
 from semantic_kernel.connectors.ai.open_ai.prompt_execution_settings.azure_chat_prompt_execution_settings import (
     AzureChatPromptExecutionSettings,
 )
-from semantic_kernel.connectors.ai.open_ai.prompt_execution_settings.open_ai_prompt_execution_settings import (
-    OpenAIChatPromptExecutionSettings,
-)
 from semantic_kernel.connectors.ai.open_ai.services.azure_config_base import AzureOpenAIConfigBase
 from semantic_kernel.connectors.ai.open_ai.services.open_ai_chat_completion_base import OpenAIChatCompletionBase
 from semantic_kernel.connectors.ai.open_ai.services.open_ai_handler import OpenAIModelTypes
 from semantic_kernel.connectors.ai.open_ai.services.open_ai_text_completion_base import OpenAITextCompletionBase
 from semantic_kernel.connectors.ai.open_ai.settings.azure_open_ai_settings import AzureOpenAISettings
 from semantic_kernel.connectors.ai.prompt_execution_settings import PromptExecutionSettings
-from semantic_kernel.contents.chat_history import ChatHistory
 from semantic_kernel.contents.chat_message_content import ChatMessageContent
 from semantic_kernel.contents.function_call_content import FunctionCallContent
 from semantic_kernel.contents.function_result_content import FunctionResultContent
 from semantic_kernel.contents.streaming_chat_message_content import StreamingChatMessageContent
 from semantic_kernel.contents.text_content import TextContent
 from semantic_kernel.contents.utils.finish_reason import FinishReason
-from semantic_kernel.exceptions.service_exceptions import ServiceInitializationError, ServiceInvalidResponseError
-from semantic_kernel.utils.telemetry.model_diagnostics.decorators import trace_streaming_chat_completion
+from semantic_kernel.exceptions.service_exceptions import ServiceInitializationError
 
 logger: logging.Logger = logging.getLogger(__name__)
 
@@ -121,42 +110,6 @@ def __init__(
             client=async_client,
         )
 
-    @override
-    @trace_streaming_chat_completion(OpenAIChatCompletionBase.MODEL_PROVIDER_NAME)
-    async def _inner_get_streaming_chat_message_contents(
-        self,
-        chat_history: "ChatHistory",
-        settings: "PromptExecutionSettings",
-    ) -> AsyncGenerator[list["StreamingChatMessageContent"], Any]:
-        """Override the base method.
-
-        This is because the latest Azure OpenAI API GA version doesn't support `stream_option`
-        yet and it will potentially result in errors if the option is included.
-        This method will be called instead of the base method.
-        TODO: Remove this method when the `stream_option` is supported by the Azure OpenAI API.
-        GitHub Issue: https://github.com/microsoft/semantic-kernel/issues/8996
-        """
-        if not isinstance(settings, OpenAIChatPromptExecutionSettings):
-            settings = self.get_prompt_execution_settings_from_settings(settings)
-        assert isinstance(settings, OpenAIChatPromptExecutionSettings)  # nosec
-
-        settings.stream = True
-        settings.messages = self._prepare_chat_history_for_request(chat_history)
-        settings.ai_model_id = settings.ai_model_id or self.ai_model_id
-
-        response = await self._send_request(settings)
-        if not isinstance(response, AsyncStream):
-            raise ServiceInvalidResponseError("Expected an AsyncStream[ChatCompletionChunk] response.")
-        async for chunk in response:
-            if len(chunk.choices) == 0:
-                continue
-
-            assert isinstance(chunk, ChatCompletionChunk)  # nosec
-            chunk_metadata = self._get_metadata_from_streaming_chat_response(chunk)
-            yield [
-                self._create_streaming_chat_message_content(chunk, choice, chunk_metadata) for choice in chunk.choices
-            ]
-
     @classmethod
     def from_dict(cls, settings: dict[str, Any]) -> "AzureChatCompletion":
         """Initialize an Azure OpenAI service from a dictionary of settings.

python/tests/unit/connectors/ai/open_ai/services/test_azure_chat_completion.py

@@ -948,4 +948,8 @@ async def test_cmc_streaming(
         model=azure_openai_unit_test_env["AZURE_OPENAI_CHAT_DEPLOYMENT_NAME"],
         stream=True,
         messages=azure_chat_completion._prepare_chat_history_for_request(chat_history),
+        # NOTE: The `stream_options={"include_usage": True}` is explicitly enforced in
+        # `OpenAIChatCompletionBase._inner_get_streaming_chat_message_contents`.
+        # To ensure consistency, we align the arguments here accordingly.
+        stream_options={"include_usage": True},
     )