fix: Notion Notes Update

docker.md

@@ -93,7 +93,7 @@ This guide will walk you through setting up and deploying the project using the
      docker run -p 8080:8080 \
         -e STORAGE_PATH=GCP \
         -e API_KEY=<set-to-any-value> \
-        -e GCP_BUCKET_NAME='<name-of-gcp-storage></name-of-gcp-storage> \
+        -e GCP_BUCKET_NAME='<name-of-gcp-storage>' \
         -e GCP_SA_CREDENTIALS=<contents-of-json-file-downloaded> \
         no-code-architects-toolkit
      ```
@@ -109,8 +109,8 @@ This guide will walk you through setting up and deploying the project using the
      docker run -p 8080:8080 \
         -e STORAGE_PATH=DRIVE \
         -e GDRIVE_USER='<email address>' \
-        -e GDRIVE_FOLDER_ID=<Folder-ID>
-        -e API_KEY=<set-to-any-value> \
+        -e GDRIVE_FOLDER_ID='<Folder-ID>'
+        -e API_KEY='<set-to-any-value>' \
         -e GCP_SA_CREDENTIALS=<contents-of-json-file-downloaded> \
         no-code-architects-toolkit
      ```

routes/transcribe.py

@@ -1,20 +1,82 @@
-from flask import Blueprint, request, jsonify
+from flask import Blueprint, request, jsonify, after_this_request
 import uuid
+import threading
+import logging
 from services.transcription import process_transcription
 from services.authentication import authenticate
+from services.webhook import send_webhook
 
 transcribe_bp = Blueprint('transcribe', __name__)
+logging.basicConfig(level=logging.INFO)
+logger = logging.getLogger(__name__)
 
 @transcribe_bp.route('/transcribe', methods=['POST'])
 @authenticate
-def transcribe_media():
+def transcribe():
     data = request.json
-    job_id = str(uuid.uuid4())  # Generate a job ID for tracking (optional)
-    print(f"Processing Job ID: {job_id}")
-
-    try:
-        # Call the transcription process directly
-        result = process_transcription(data['media_url'], data['output'])
-        return jsonify({"job_id": job_id, "result": result}), 200
-    except Exception as e:
-        return jsonify({"job_id": job_id, "error": str(e)}), 500
+    media_url = data.get('media_url')
+    output = data.get('output', 'transcript').lower()
+    webhook_url = data.get('webhook_url')
+    id = data.get('id')
+
+    logger.info(f"Received transcription request: media_url={media_url}, output={output}, webhook_url={webhook_url}, id={id}")
+
+    if not media_url:
+        logger.error("Missing media_url parameter in request")
+        return jsonify({"error": "Missing media_url parameter"}), 400
+
+    # Check if either webhook_url or id is provided without the other
+    if (webhook_url and not id) or (id and not webhook_url):
+        logger.warning("Either webhook_url or id is missing in the request")
+        return jsonify({"response": "transcript or srt text is missing"}), 200
+
+    job_id = str(uuid.uuid4())
+    logger.info(f"Generated job_id: {job_id}")
+
+    def process_and_notify(media_url, output, webhook_url, id, job_id):
+        try:
+            logger.info(f"Job {job_id}: Starting transcription process for {media_url}")
+            result = process_transcription(media_url, output)
+            logger.info(f"Job {job_id}: Transcription process completed successfully")
+
+            if webhook_url:
+                logger.info(f"Job {job_id}: Sending success webhook to {webhook_url}")
+                send_webhook(webhook_url, {
+                    "endpoint": "/transcribe",
+                    "id": id,
+                    "response": result,
+                    "code": 200,
+                    "message": "success"
+                })
+        except Exception as e:
+            logger.error(f"Job {job_id}: Error during transcription - {e}")
+            if webhook_url:
+                logger.info(f"Job {job_id}: Sending failure webhook to {webhook_url}")
+                send_webhook(webhook_url, {
+                    "endpoint": "/transcribe",
+                    "id": id,
+                    "response": None,
+                    "code": 500,
+                    "message": str(e)
+                })
+
+    @after_this_request
+    def start_background_processing(response):
+        logger.info(f"Job {job_id}: Starting background processing thread")
+        thread = threading.Thread(target=process_and_notify, args=(media_url, output, webhook_url, id, job_id))
+        thread.start()
+        return response
+
+    # If webhook_url and id are provided, return 202 Accepted
+    if webhook_url and id:
+        logger.info(f"Job {job_id}: Returning 202 Accepted response and processing in background")
+        return jsonify({"message": "processing"}), 202
+    else:
+        try:
+            logger.info(f"Job {job_id}: No webhook provided, processing synchronously")
+            result = process_transcription(media_url, output)
+            logger.info(f"Job {job_id}: Returning transcription result")
+            return jsonify({"response": result}), 200
+        except Exception as e:
+            logger.error(f"Job {job_id}: Error during synchronous transcription - {e}")
+            return jsonify({"message": str(e)}), 500

services/whisper_transcription.py

@@ -6,45 +6,28 @@
 from services.webhook import send_webhook
 from services.file_management import download_file, STORAGE_PATH
 
-def process_transcription(media_url, output_type, webhook_url):
+def process_transcription(media_url, output_type):
     try:
-        # Generate a unique job ID using UUID
-        job_id = str(uuid.uuid4())
-        input_filename = download_file(media_url, os.path.join(STORAGE_PATH, f"{job_id}_input"))
-
+        input_filename = download_file(media_url, os.path.join(STORAGE_PATH, 'input_media'))
+
         model = whisper.load_model("base")
         result = model.transcribe(input_filename)
 
-        os.remove(input_filename)  # Clean up the input file after transcription
-
         if output_type == 'transcript':
             output = result['text']
-        else:  # SRT format
+        elif output_type == 'srt':
             srt_subtitles = []
             for i, segment in enumerate(result['segments'], start=1):
                 start = timedelta(seconds=segment['start'])
                 end = timedelta(seconds=segment['end'])
                 text = segment['text'].strip()
                 srt_subtitles.append(srt.Subtitle(i, start, end, text))
             output = srt.compose(srt_subtitles)
+        else:
+            raise ValueError("Invalid output type. Must be 'transcript' or 'srt'.")
 
-        if webhook_url:
-            send_webhook(webhook_url, {
-                "endpoint": "/transcribe",
-                "job_id": job_id,
-                "response": output,
-                "code": 200,
-                "message": "success"
-            })
-
-        return job_id  # Return the generated job ID
+        os.remove(input_filename)
+        return output
     except Exception as e:
-        if webhook_url:
-            send_webhook(webhook_url, {
-                "endpoint": "/transcribe",
-                "job_id": None,
-                "response": None,
-                "code": 500,
-                "message": str(e)
-            })
+        print(f"Transcription failed: {str(e)}")
         raise