Fix identation

vertex_api_invoice_extractor.py

@@ -11,31 +11,31 @@ from remittance_pdf_processing_utils import remove_duplicate_lists
 
 # Set up authentication
 def initialize_vertexai():
- # Get the base64-encoded service account JSON from an environment variable
- encoded_sa_json = os.environ.get('VERTEX_AI_SERVICE_ACCOUNT_JSON')
- 
- if not encoded_sa_json:
- raise ValueError("VERTEX_AI_SERVICE_ACCOUNT_JSON environment variable is not set")
- 
- try:
- # Decode the base64 string to get the JSON content
- sa_json_str = base64.b64decode(encoded_sa_json).decode('utf-8')
- sa_info = json.loads(sa_json_str)
- 
- # Create credentials object from the decoded JSON
- credentials = service_account.Credentials.from_service_account_info(
- sa_info,
- scopes=['https://www.googleapis.com/auth/cloud-platform']
- )
- 
- # Initialize Vertex AI with the credentials
- vertexai.init(project="saltech-ai-sandbox", location="us-central1", credentials=credentials)
- 
- print("Vertex AI initialized successfully.")
- except json.JSONDecodeError:
- raise ValueError("Invalid JSON format in the decoded service account information")
- except Exception as e:
- raise Exception(f"Error initializing Vertex AI: {str(e)}")
+ # Get the base64-encoded service account JSON from an environment variable
+ encoded_sa_json = os.environ.get('VERTEX_AI_SERVICE_ACCOUNT_JSON')
+ 
+ if not encoded_sa_json:
+ raise ValueError("VERTEX_AI_SERVICE_ACCOUNT_JSON environment variable is not set")
+ 
+ try:
+ # Decode the base64 string to get the JSON content
+ sa_json_str = base64.b64decode(encoded_sa_json).decode('utf-8')
+ sa_info = json.loads(sa_json_str)
+ 
+ # Create credentials object from the decoded JSON
+ credentials = service_account.Credentials.from_service_account_info(
+ sa_info,
+ scopes=['https://www.googleapis.com/auth/cloud-platform']
+ )
+ 
+ # Initialize Vertex AI with the credentials
+ vertexai.init(project="saltech-ai-sandbox", location="us-central1", credentials=credentials)
+ 
+ print("Vertex AI initialized successfully.")
+ except json.JSONDecodeError:
+ raise ValueError("Invalid JSON format in the decoded service account information")
+ except Exception as e:
+ raise Exception(f"Error initializing Vertex AI: {str(e)}")
 
 
 # Call this function at the start of your script or in your main function
@@ -84,14 +84,14 @@ def extract_invoice_numbers_with_vertex_ai_single_hop(base64_image: str) -> list
  "temperature": 0.1,
  "top_p": 0.95,
  }
- 
+ 
  safety_settings = {
  generative_models.HarmCategory.HARM_CATEGORY_HATE_SPEECH: generative_models.HarmBlockThreshold.BLOCK_MEDIUM_AND_ABOVE,
  generative_models.HarmCategory.HARM_CATEGORY_DANGEROUS_CONTENT: generative_models.HarmBlockThreshold.BLOCK_MEDIUM_AND_ABOVE,
  generative_models.HarmCategory.HARM_CATEGORY_SEXUALLY_EXPLICIT: generative_models.HarmBlockThreshold.BLOCK_MEDIUM_AND_ABOVE,
  generative_models.HarmCategory.HARM_CATEGORY_HARASSMENT: generative_models.HarmBlockThreshold.BLOCK_MEDIUM_AND_ABOVE,
  }
- safety_settings = {}
+ safety_settings = {}
 
  responses = model.generate_content(
  [image_part, text_prompt],
@@ -142,7 +142,7 @@ def extract_column_headers(base64_image: str) -> list[str]:
  generative_models.HarmCategory.HARM_CATEGORY_SEXUALLY_EXPLICIT: generative_models.HarmBlockThreshold.BLOCK_MEDIUM_AND_ABOVE,
  generative_models.HarmCategory.HARM_CATEGORY_HARASSMENT: generative_models.HarmBlockThreshold.BLOCK_MEDIUM_AND_ABOVE,
  }
- safety_settings = {}
+ safety_settings = {}
 
  responses = model.generate_content(
  [image_part, text_prompt],
@@ -193,7 +193,7 @@ def extract_invoice_numbers_for_column(base64_image: str, column_name: str) -> I
  generative_models.HarmCategory.HARM_CATEGORY_SEXUALLY_EXPLICIT: generative_models.HarmBlockThreshold.BLOCK_MEDIUM_AND_ABOVE,
  generative_models.HarmCategory.HARM_CATEGORY_HARASSMENT: generative_models.HarmBlockThreshold.BLOCK_MEDIUM_AND_ABOVE,
  }
- safety_settings = {}
+ safety_settings = {}
 
  responses = model.generate_content(
  [image_part, text_prompt],
@@ -358,7 +358,7 @@ def extract_invoice_numbers_from_text_with_vertex_ai_single_hop(text: str) -> li
  generative_models.HarmCategory.HARM_CATEGORY_SEXUALLY_EXPLICIT: generative_models.HarmBlockThreshold.BLOCK_MEDIUM_AND_ABOVE,
  generative_models.HarmCategory.HARM_CATEGORY_HARASSMENT: generative_models.HarmBlockThreshold.BLOCK_MEDIUM_AND_ABOVE,
  }
- safety_settings = {}
+ safety_settings = {}
 
  responses = model.generate_content(
  prompt,
@@ -434,7 +434,7 @@ def extract_column_headers_from_text(text: str) -> list[str]:
  generative_models.HarmCategory.HARM_CATEGORY_SEXUALLY_EXPLICIT: generative_models.HarmBlockThreshold.BLOCK_MEDIUM_AND_ABOVE,
  generative_models.HarmCategory.HARM_CATEGORY_HARASSMENT: generative_models.HarmBlockThreshold.BLOCK_MEDIUM_AND_ABOVE,
  }
- safety_settings = {}
+ safety_settings = {}
 
  response = model.generate_content(
  prompt,
@@ -479,7 +479,7 @@ def extract_invoice_numbers_for_column_from_text(text: str, column_name: str) ->
  generative_models.HarmCategory.HARM_CATEGORY_SEXUALLY_EXPLICIT: generative_models.HarmBlockThreshold.BLOCK_MEDIUM_AND_ABOVE,
  generative_models.HarmCategory.HARM_CATEGORY_HARASSMENT: generative_models.HarmBlockThreshold.BLOCK_MEDIUM_AND_ABOVE,
  }
- safety_settings = {}
+ safety_settings = {}
 
  response = model.generate_content(
  prompt,
@@ -516,7 +516,7 @@ def extract_payment_amounts_with_vertex_ai(base64_image: str) -> list[PaymentAmo
  generative_models.HarmCategory.HARM_CATEGORY_SEXUALLY_EXPLICIT: generative_models.HarmBlockThreshold.BLOCK_MEDIUM_AND_ABOVE,
  generative_models.HarmCategory.HARM_CATEGORY_HARASSMENT: generative_models.HarmBlockThreshold.BLOCK_MEDIUM_AND_ABOVE,
  }
- safety_settings = {}
+ safety_settings = {}
 
  responses = model.generate_content(
  [image_part, text_prompt],
@@ -560,14 +560,14 @@ def extract_payment_amounts_from_text_with_vertex_ai(text: str) -> list[PaymentA
  "temperature": 0.1,
  "top_p": 0.95,
  }
- 
+ 
  safety_settings = {
  generative_models.HarmCategory.HARM_CATEGORY_HATE_SPEECH: generative_models.HarmBlockThreshold.BLOCK_MEDIUM_AND_ABOVE,
  generative_models.HarmCategory.HARM_CATEGORY_DANGEROUS_CONTENT: generative_models.HarmBlockThreshold.BLOCK_MEDIUM_AND_ABOVE,
  generative_models.HarmCategory.HARM_CATEGORY_SEXUALLY_EXPLICIT: generative_models.HarmBlockThreshold.BLOCK_MEDIUM_AND_ABOVE,
  generative_models.HarmCategory.HARM_CATEGORY_HARASSMENT: generative_models.HarmBlockThreshold.BLOCK_MEDIUM_AND_ABOVE,
- }
- safety_settings = {}
+ }
+ safety_settings = {}
 
  response = model.generate_content(
  prompt,