fix multipartform issue (#44123)

* fix multipartform

* fix pylint

* update patch

Author: amber-yujueWang

sdk/cognitiveservices/azure-ai-transcription/README.md

@@ -165,7 +165,7 @@ with open(audio_file_path, "rb") as audio_file:
     options = TranscriptionOptions(locales=["en-US"])  # Specify the language
 
     # Create the request content
-    request_content = TranscriptionContent(options=options, audio=audio_file)
+    request_content = TranscriptionContent(definition=options, audio=audio_file)
 
     # Transcribe the audio
     result = client.transcribe(request_content)
@@ -204,7 +204,6 @@ client = TranscriptionClient(endpoint=endpoint, credential=AzureKeyCredential(ap
 
 # URL to your audio file (must be publicly accessible)
 audio_url = "https://example.com/path/to/audio.wav"
-
 # Configure transcription options
 options = TranscriptionOptions(locales=["en-US"])
 
@@ -266,7 +265,7 @@ with open(audio_file_path, "rb") as audio_file:
     options = TranscriptionOptions(locales=["en-US"], enhanced_mode=enhanced_mode)
 
     # Create the request content
-    request_content = TranscriptionContent(options=options, audio=audio_file)
+    request_content = TranscriptionContent(definition=options, audio=audio_file)
 
     # Transcribe the audio with enhanced mode
     result = client.transcribe(request_content)
@@ -312,7 +311,7 @@ async with TranscriptionClient(endpoint=endpoint, credential=AzureKeyCredential(
         options = TranscriptionOptions(locales=["en-US"])  # Specify the language
 
         # Create the request content
-        request_content = TranscriptionContent(options=options, audio=audio_file)
+        request_content = TranscriptionContent(definition=options, audio=audio_file)
 
         # Transcribe the audio
         result = await client.transcribe(request_content)

sdk/cognitiveservices/azure-ai-transcription/azure/ai/transcription/_operations/_operations.py

@@ -109,7 +109,7 @@ def transcribe(self, body: Union[_models.TranscriptionContent, JSON], **kwargs:
 
         _body = body.as_dict() if isinstance(body, _Model) else body
         _file_fields: list[str] = ["audio"]
-        _data_fields: list[str] = ["options"]
+        _data_fields: list[str] = ["definition"]
         _files, _data = prepare_multipart_form_data(_body, _file_fields, _data_fields)
 
         _request = build_transcription_transcribe_request(

sdk/cognitiveservices/azure-ai-transcription/azure/ai/transcription/_operations/_patch.py

@@ -1,3 +1,4 @@
+# pylint: disable=line-too-long,useless-suppression
 # coding=utf-8
 # --------------------------------------------------------------------------
 # Copyright (c) Microsoft Corporation. All rights reserved.
@@ -9,11 +10,15 @@
 """
 from collections.abc import MutableMapping
 from typing import Any, Optional
+import json
 from azure.core.tracing.decorator import distributed_trace
+from azure.core.exceptions import map_error, HttpResponseError, ClientAuthenticationError, ResourceNotFoundError, ResourceExistsError, ResourceNotModifiedError
 
 from .. import models as _models
+from .._utils.model_base import _deserialize, SdkJSONEncoder
 from ._operations import (
     _TranscriptionClientOperationsMixin as _TranscriptionClientOperationsMixinGenerated,
+    build_transcription_transcribe_request,
 )
 
 JSON = MutableMapping[str, Any]
@@ -56,11 +61,47 @@ def transcribe_from_url(
         else:
             options.audio_url = audio_url
 
-        # Create request content without audio file (service will fetch from URL)
-        body = _models.TranscriptionContent(options=options, audio=None)
-
-        # Call the underlying protocol method
-        return super().transcribe(body, **kwargs)
+        # Send as multipart request with only definition (no audio file)
+        error_map: MutableMapping = {
+            401: ClientAuthenticationError,
+            404: ResourceNotFoundError,
+            409: ResourceExistsError,
+            304: ResourceNotModifiedError,
+        }
+        error_map.update(kwargs.pop("error_map", {}) or {})
+
+        _headers = kwargs.pop("headers", {}) or {}
+        _params = kwargs.pop("params", {}) or {}
+
+        _params["api-version"] = self._config.api_version
+        _headers["Accept"] = "application/json"
+
+        # Serialize definition as JSON string for multipart
+        definition_json = json.dumps(options.as_dict(), cls=SdkJSONEncoder, exclude_readonly=True)
+
+        # Build multipart request - pass definition through files to ensure multipart encoding
+        # The definition needs to be in files list with explicit content-type to trigger multipart/form-data
+        _request = build_transcription_transcribe_request(
+            api_version=self._config.api_version,
+            files=[("definition", (None, definition_json, "application/json"))],
+            headers=_headers,
+            params=_params,
+        )
+
+        path_format_arguments = {
+            "endpoint": self._serialize.url("self._config.endpoint", self._config.endpoint, "str", skip_quote=True),
+        }
+        _request.url = self._client.format_url(_request.url, **path_format_arguments)
+
+        pipeline_response = self._client._pipeline.run(_request, stream=False, **kwargs)  # pylint: disable=protected-access
+        response = pipeline_response.http_response
+
+        if response.status_code not in [200]:
+            map_error(status_code=response.status_code, response=response, error_map=error_map)
+            raise HttpResponseError(response=response)
+
+        deserialized = _deserialize(_models.TranscriptionResult, response.json())
+        return deserialized
 
 
 __all__: list[str] = [

sdk/cognitiveservices/azure-ai-transcription/azure/ai/transcription/aio/_operations/_operations.py

@@ -89,7 +89,7 @@ async def transcribe(
 
         _body = body.as_dict() if isinstance(body, _Model) else body
         _file_fields: list[str] = ["audio"]
-        _data_fields: list[str] = ["options"]
+        _data_fields: list[str] = ["definition"]
         _files, _data = prepare_multipart_form_data(_body, _file_fields, _data_fields)
 
         _request = build_transcription_transcribe_request(

sdk/cognitiveservices/azure-ai-transcription/azure/ai/transcription/aio/_operations/_patch.py

@@ -1,3 +1,4 @@
+# pylint: disable=line-too-long,useless-suppression
 # coding=utf-8
 # --------------------------------------------------------------------------
 # Copyright (c) Microsoft Corporation. All rights reserved.
@@ -9,9 +10,13 @@
 """
 from collections.abc import MutableMapping
 from typing import Any, Optional
+import json
 from azure.core.tracing.decorator_async import distributed_trace_async
+from azure.core.exceptions import map_error, HttpResponseError, ClientAuthenticationError, ResourceNotFoundError, ResourceExistsError, ResourceNotModifiedError
 
 from ... import models as _models
+from ..._utils.model_base import _deserialize, SdkJSONEncoder
+from ..._operations._operations import build_transcription_transcribe_request
 from ._operations import (
     _TranscriptionClientOperationsMixin as _TranscriptionClientOperationsMixinGenerated,
 )
@@ -56,11 +61,45 @@ async def transcribe_from_url(
         else:
             options.audio_url = audio_url
 
-        # Create request content without audio file (service will fetch from URL)
-        body = _models.TranscriptionContent(options=options, audio=None)
-
-        # Call the underlying protocol method
-        return await super().transcribe(body, **kwargs)
+        # Send as multipart request with only definition (no audio file)
+        error_map: MutableMapping = {
+            401: ClientAuthenticationError,
+            404: ResourceNotFoundError,
+            409: ResourceExistsError,
+            304: ResourceNotModifiedError,
+        }
+        error_map.update(kwargs.pop("error_map", {}) or {})
+
+        _headers = kwargs.pop("headers", {}) or {}
+        _params = kwargs.pop("params", {}) or {}
+
+        _headers["Accept"] = "application/json"
+
+        # Serialize definition as JSON string for multipart
+        definition_json = json.dumps(options.as_dict(), cls=SdkJSONEncoder, exclude_readonly=True)
+
+        # Build multipart request - pass definition through files to ensure multipart encoding
+        # The definition needs to be in files list with explicit content-type to trigger multipart/form-data
+        _request = build_transcription_transcribe_request(
+            api_version=self._config.api_version,
+            files=[("definition", (None, definition_json, "application/json"))],
+            headers=_headers,
+        )
+
+        path_format_arguments = {
+            "endpoint": self._serialize.url("self._config.endpoint", self._config.endpoint, "str", skip_quote=True),
+        }
+        _request.url = self._client.format_url(_request.url, **path_format_arguments)
+
+        pipeline_response = await self._client._pipeline.run(_request, stream=False, **kwargs)  # pylint: disable=protected-access
+        response = pipeline_response.http_response
+
+        if response.status_code not in [200]:
+            map_error(status_code=response.status_code, response=response, error_map=error_map)
+            raise HttpResponseError(response=response)
+
+        deserialized = _deserialize(_models.TranscriptionResult, response.json())
+        return deserialized
 
 
 __all__: list[str] = [

sdk/cognitiveservices/azure-ai-transcription/azure/ai/transcription/models/_models.py

@@ -247,20 +247,18 @@ def __init__(self, *args: Any, **kwargs: Any) -> None:
 class TranscriptionContent(_Model):
     """Request model for transcription operation.
 
-    :ivar options: Metadata for a transcription request. This field contains a JSON-serialized
-     object of type ``TranscriptionOptions``.
-    :vartype options: ~azure.ai.transcription.models.TranscriptionOptions
+    :ivar definition: Metadata for a transcription request. This field contains a JSON-serialized
+     object of type ``TranscriptionOptions``. Required.
+    :vartype definition: ~azure.ai.transcription.models.TranscriptionOptions
     :ivar audio: The content of the audio file to be transcribed. The audio file must be shorter
      than 2 hours in audio duration and smaller than 250 MB in size. Optional if audioUrl is
      provided in the definition.
     :vartype audio: ~azure.ai.transcription._utils.utils.FileType
     """
 
-    options: Optional["_models.TranscriptionOptions"] = rest_field(
-        visibility=["read", "create", "update", "delete", "query"]
-    )
+    definition: "_models.TranscriptionOptions" = rest_field(visibility=["read", "create", "update", "delete", "query"])
     """Metadata for a transcription request. This field contains a JSON-serialized object of type
-     ``TranscriptionOptions``."""
+     ``TranscriptionOptions``. Required."""
     audio: Optional[FileType] = rest_field(
         visibility=["read", "create", "update", "delete", "query"], is_multipart_file_input=True
     )
@@ -272,7 +270,7 @@ class TranscriptionContent(_Model):
     def __init__(
         self,
         *,
-        options: Optional["_models.TranscriptionOptions"] = None,
+        definition: "_models.TranscriptionOptions",
         audio: Optional[FileType] = None,
     ) -> None: ...
 

sdk/cognitiveservices/azure-ai-transcription/generated_tests/test_transcription.py

@@ -18,8 +18,7 @@ def test_transcribe(self, transcription_endpoint):
         client = self.create_client(endpoint=transcription_endpoint)
         response = client.transcribe(
             body={
-                "audio": "filetype",
-                "options": {
+                "definition": {
                     "audioUrl": "str",
                     "channels": [0],
                     "diarization": {"enabled": bool, "maxSpeakers": 0},
@@ -29,6 +28,7 @@ def test_transcribe(self, transcription_endpoint):
                     "phraseList": {"biasingWeight": 0.0, "phrases": ["str"]},
                     "profanityFilterMode": "str",
                 },
+                "audio": "filetype",
             },
         )
 

sdk/cognitiveservices/azure-ai-transcription/generated_tests/test_transcription_async.py

@@ -19,8 +19,7 @@ async def test_transcribe(self, transcription_endpoint):
         client = self.create_async_client(endpoint=transcription_endpoint)
         response = await client.transcribe(
             body={
-                "audio": "filetype",
-                "options": {
+                "definition": {
                     "audioUrl": "str",
                     "channels": [0],
                     "diarization": {"enabled": bool, "maxSpeakers": 0},
@@ -30,6 +29,7 @@ async def test_transcribe(self, transcription_endpoint):
                     "phraseList": {"biasingWeight": 0.0, "phrases": ["str"]},
                     "profanityFilterMode": "str",
                 },
+                "audio": "filetype",
             },
         )
 

sdk/cognitiveservices/azure-ai-transcription/samples/async_samples/sample_transcribe_audio_file_async.py

@@ -47,7 +47,7 @@ async def sample_transcribe_audio_file_async():
             options = TranscriptionOptions(locales=["en-US"])  # Specify the language
 
             # Create the request content
-            request_content = TranscriptionContent(options=options, audio=audio_file)
+            request_content = TranscriptionContent(definition=options, audio=audio_file)
 
             # Transcribe the audio
             result = await client.transcribe(request_content)

sdk/cognitiveservices/azure-ai-transcription/samples/async_samples/sample_transcribe_multiple_languages_async.py

@@ -49,7 +49,7 @@ async def sample_transcribe_multiple_languages_async():
             options = TranscriptionOptions(locales=["en-US", "es-ES", "fr-FR", "de-DE"])  # Multiple candidates
 
             # Create the request content
-            request_content = TranscriptionContent(options=options, audio=audio_file)
+            request_content = TranscriptionContent(definition=options, audio=audio_file)
 
             # Transcribe the audio
             result = await client.transcribe(request_content)