Fixed json format

project/backend/exxetaGPT-service/app.py

@@ -25,7 +25,7 @@ def extract_text_from_ocr_json():
     }
 
     requests.post(VALIDATE_SERVICE_URL, json=validate_payload, timeout=600)
-    return jsonify({"Sent to validate-service"}), 200
+    return jsonify("Sent to validate-service"), 200
 
 
 if __name__ == "__main__":

project/backend/exxetaGPT-service/extractExxeta.py

@@ -3,11 +3,10 @@ import json
 import os
 import time
 import logging
-
 from dotenv import load_dotenv
 
 MODEL = "gpt-4o-mini"
-EXXETA_BASE_URL= "https://ai.exxeta.com/api/v2/azure/openai"
+EXXETA_BASE_URL = "https://ai.exxeta.com/api/v2/azure/openai"
 load_dotenv()
 EXXETA_API_KEY = os.getenv("API_KEY")
 
@@ -19,13 +18,15 @@ logger = logging.getLogger(__name__)
 
 def extract_with_exxeta(pages_json):
     results = []
+
     if not EXXETA_API_KEY:
         logger.warning("EXXETA_API_KEY nicht gesetzt. Rückgabe eines leeren JSON.")
         return json.dumps(results, indent=2, ensure_ascii=False)
 
     for page_data in pages_json:
         page_num = page_data.get("page")
-        text = page_data.get("text", "").strip()
+        page_data.get("page")
+        text = page_data.get("text", "")
 
         if not text:
             continue
@@ -105,7 +106,7 @@ def extract_with_exxeta(pages_json):
         payload = {
             "model": MODEL,
             "messages": [
-                {"role": "system", "content": "Du bist ein Finanzanalyst, der Fondsprofile auswertet. Antworte nur mit validen JSON-Arrays."},
+                {"role": "system", "content": "Du bist ein Finanzanalyst. Antworte ausschließlich mit einem validen JSON-Array."},
                 {"role": "user", "content": prompt}
             ],
             "temperature": 0.0
@@ -143,5 +144,4 @@ def extract_with_exxeta(pages_json):
                 if attempt == MAX_RETRIES:
                     results.extend([])
 
-    json_result = json.dumps(results, indent=2, ensure_ascii=False)
-    return json_result
+    return json.dumps(results, indent=2, ensure_ascii=False)

project/backend/ocr-service/app.py

@@ -16,7 +16,7 @@ def convert_extract_text_from_pdf():
         return {"error": "No file"}, 400
 
     file = request.files["file"]
-    pitchbook_id = request.form.get("id")
+    pitchbook_id = request.form.get("pitchbook_id")
 
     if not pitchbook_id:
         return {"error": "No ID"}, 400

project/backend/spacy-service/app.py

@@ -25,7 +25,7 @@ def extract_pdf():
     }
 
     requests.post(VALIDATE_SERVICE_URL, json=validate_payload, timeout=600)
-    return jsonify({"Sent to validate-service"}), 200
+    return jsonify("Sent to validate-service"), 200
 
 
 if __name__ == "__main__":

project/backend/spacy-service/extractSpacy.py

@@ -6,15 +6,14 @@ current_dir = os.path.dirname(os.path.abspath(__file__))
 model_path = os.path.join(current_dir, "spacy_training/output/model-last")
 nlp = spacy.load(model_path)
 
-
 def extract(pages_json):
-
     results = []
 
     for page in pages_json:
-        text = page.get("text", "").strip()
-        page_num = page.get("page")
+        text = page.get("text", "")
+        text = text.strip()
 
+        page_num = page.get("page")
         if not text:
             continue
 
@@ -26,5 +25,4 @@ def extract(pages_json):
                 "page": page_num
             })
 
-    json_result = json.dumps(results, indent=2, ensure_ascii=False)
-    return json_result
+    return json.dumps(results, indent=2, ensure_ascii=False)

project/docker-compose.yml

@@ -29,9 +29,9 @@ services:
         condition: service_healthy
     healthcheck:
       test:  wget --spider --no-verbose http://127.0.0.1:5000/health || exit 1
-      interval: 10s
-      timeout: 5s
-      retries: 5
+      interval: 20s
+      timeout: 10s
+      retries: 10
     ports:
       - 5050:5000