wip

src/backend/core/utils.py

@@ -401,6 +401,8 @@ def _get_s3_client(*, override_domain: bool = True):
             config=botocore.client.Config(
                 region_name=settings.AWS_S3_REGION_NAME,
                 signature_version=settings.AWS_S3_SIGNATURE_VERSION,
+                request_checksum_calculation=settings.AWS_REQUEST_CHECKSUM_CALCULATION,
+                response_checksum_validation=settings.AWS_RESPONSE_CHECKSUM_VALIDATION,
             ),
         )
     return default_storage.connection.meta.client

src/summary/summary/core/celery_worker.py

@@ -442,6 +442,19 @@ def call_webhook_v2_task(
         payload=webhook_payload_adapter.validate_python(payload), tenant_id=tenant_id
     )
 
+MOCK_TRANSCRIPT_JSON = """
+{
+  "segments": [
+    {"start": 0.0, "end": 0.9, "text": "Bonjour à tous.", "speaker": "SPEAKER_00"}
+  ],
+  "word_segments": [
+    {"word": "Bonjour", "start": 0.0, "end": 0.42, "score": 0.98},
+    {"word": "à", "start": 0.45, "end": 0.52, "score": 0.95},
+    {"word": "tous.", "start": 0.55, "end": 0.9, "score": 0.97}
+  ]
+}
+"""
+
 
 @celery.task(
     bind=True,
@@ -474,6 +487,13 @@ def process_audio_transcribe_v2_task(
 
     job_id = self.request.id
 
+    transcription_res = WhisperXResponse.model_validate_json(MOCK_TRANSCRIPT_JSON)
+
+    file_service.store_transcript(
+        transcript=transcription_res,
+        job_id=job_id,
+    )
+
     try:
         transcription_res = WhisperXResponse(
             **transcribe_audio(  # type: ignore

src/summary/summary/core/config.py

@@ -84,6 +84,8 @@ class Settings(BaseSettings):
     aws_s3_secret_access_key: SecretStr
     aws_s3_secure_access: bool = True
     aws_s3_region_name: str | None = None
+    aws_request_checksum_calculation: str = "when_required"
+    aws_response_checksum_validation: str = "when_required"
     aws_transcript_path: str = "transcripts"
     aws_summary_path: str = "summaries"