Update app.py

app.py

@@ -1,34 +1,11 @@
 import os
 from paddleocr import PaddleOCR
+from PIL import Image
 import gradio as gr
 import re
 from simple_salesforce import Salesforce
 import pandas as pd
 
-# ---------------------------
-# Salesforce Credentials Setup
-# Replace these with your actual credentials or set them as environment variables
-SALESFORCE_USERNAME = os.getenv("SALESFORCE_USERNAME", "[email protected]")
-SALESFORCE_PASSWORD = os.getenv("SALESFORCE_PASSWORD", "Hms@2025")
-SALESFORCE_SECURITY_TOKEN = os.getenv("SALESFORCE_SECURITY_TOKEN", "2TM7U3O9HGSiJuFMUbVd3xuY")
-
-print(f"Using Salesforce credentials:")
-print(f"Username: {SALESFORCE_USERNAME}")
-print(f"Password set: {'Yes' if SALESFORCE_PASSWORD else 'No'}")
-print(f"Security Token set: {'Yes' if SALESFORCE_SECURITY_TOKEN else 'No'}")
-
-# Test Salesforce login once at startup to catch invalid credentials early
-try:
-    sf_test = Salesforce(
-        username=SALESFORCE_USERNAME,
-        password=SALESFORCE_PASSWORD,
-        security_token=SALESFORCE_SECURITY_TOKEN,
-        domain="login"  # Use "test" if sandbox
-    )
-    print(f"✅ Successfully connected to Salesforce as {SALESFORCE_USERNAME}")
-except Exception as e:
-    print(f"❌ Salesforce login test failed: {e}")
-
 # Attribute mappings: readable names to Salesforce API names
 ATTRIBUTE_MAPPING = {
     "Name": "Patient_Name__c",
@@ -37,17 +14,30 @@ ATTRIBUTE_MAPPING = {
     "Phone Number": "Phone_Number__c"
 }
 
+# Desired order of attributes for display
 ATTRIBUTE_ORDER = ["Name", "Age", "Gender", "Phone Number"]
 
+# Mapping for Gender__c picklist values
 GENDER_MAPPING = {
     "Male": "Male",
     "Female": "Female",
     "Other": "Others"
 }
 
+# Salesforce credentials from environment variables (removed hardcoded defaults)
+SALESFORCE_USERNAME = os.getenv("[email protected]")
+SALESFORCE_PASSWORD = os.getenv("Hms@2025")
+SALESFORCE_SECURITY_TOKEN = os.getenv("jxumt3OHbuMTJjMyFnnG2v3e")
+
+# Log the credentials being used (for debugging)
+print(f"Using Salesforce credentials - Username: {SALESFORCE_USERNAME}")
+print(f"Password set: {'Yes' if SALESFORCE_PASSWORD else 'No'}")
+print(f"Security token set: {'Yes' if SALESFORCE_SECURITY_TOKEN else 'No'}")
+
 # Initialize PaddleOCR
 ocr = PaddleOCR(use_angle_cls=True, lang='en')
 
+# Function to extract text from an image using PaddleOCR
 def extract_text(image):
     result = ocr.ocr(image)
     extracted_text = []
@@ -55,92 +45,123 @@ def extract_text(image):
         extracted_text.append(line[1][0])
     return "\n".join(extracted_text)
 
+# Function to extract attributes using regex
 def extract_attributes(extracted_text):
     attributes = {}
+
+    # Patterns for extracting personal information
     patterns = {
         "Name": r"Name[:\-]?\s*([A-Za-z\s]+?)(?=\s*(?:Age|Gender|Phone Number|Phone|Mobile|$))",
         "Age": r"Age[:\-]?\s*(\d{1,3})",
         "Gender": r"Gender[:\-]?\s*(Male|Female|Other)",
         "Phone Number": r"(?:(?:Phone Number)|Phone|Mobile|Phonenumber)[:\-]?\s*(?:\+91)?([6-9]\d{9})"
     }
+
     for readable_attr, pattern in patterns.items():
         match = re.search(pattern, extracted_text, re.IGNORECASE)
         if match:
             attributes[readable_attr] = match.group(1).strip()
+
     if "Gender" in attributes:
         attributes["Gender"] = GENDER_MAPPING.get(attributes["Gender"], attributes["Gender"])
+
     return attributes
 
+# Function to filter attributes for valid Salesforce fields
 def filter_valid_attributes(attributes, valid_fields):
-    return {ATTRIBUTE_MAPPING[key]: value for key, value in attributes.items() if ATTRIBUTE_MAPPING[key] in valid_fields}
+    filtered = {ATTRIBUTE_MAPPING[key]: value for key, value in attributes.items() if ATTRIBUTE_MAPPING[key] in valid_fields}
+    return filtered
 
+# Function to create a record in Salesforce
 def interact_with_salesforce(attributes):
     try:
-        # Validate credentials presence
+        # Validate that credentials are not empty
         if not all([SALESFORCE_USERNAME, SALESFORCE_PASSWORD, SALESFORCE_SECURITY_TOKEN]):
-            raise ValueError("Salesforce credentials are missing or incomplete.")
+            raise ValueError("One or more Salesforce credentials are missing. Check environment variables.")
 
+        # Initialize Salesforce connection
         sf = Salesforce(
             username=SALESFORCE_USERNAME,
             password=SALESFORCE_PASSWORD,
             security_token=SALESFORCE_SECURITY_TOKEN,
-            domain="login"
+            domain="login",  # Changed from "test" to "login" for production/developer org
+            version="60.0"
         )
+        print(f"Successfully connected to Salesforce as {SALESFORCE_USERNAME}")
+
+        # Reference the Patient_Registration__c object
         object_name = "Patient_Registration__c"
         sf_object = sf.__getattr__(object_name)
 
-        # Describe object to get valid fields and picklist values
+        # Get the object's schema to validate fields
         schema = sf_object.describe()
         valid_fields = {field["name"] for field in schema["fields"]}
+        print(f"Valid fields for {object_name}: {valid_fields}")
 
+        # Check field permissions and picklist values for Gender__c
         field_details = {field["name"]: {
             "createable": field["createable"],
             "required": not field["nillable"] and not field["defaultedOnCreate"],
             "picklist_values": [val["value"] for val in field.get("picklistValues", [])] if field.get("picklistValues") else None
         } for field in schema["fields"]}
+        print(f"Field details: {field_details}")
 
-        # Filter attributes for valid Salesforce fields
+        # Filter attributes to match valid Salesforce fields
         filtered_attributes = filter_valid_attributes(attributes, valid_fields)
 
-        # Check required field Patient_Name__c
+        # Ensure Patient_Name__c is provided (likely required)
         if "Patient_Name__c" not in filtered_attributes or not filtered_attributes["Patient_Name__c"]:
-            raise ValueError("Patient_Name__c is required but not provided.")
+            raise ValueError("Patient_Name__c is required but was not provided.")
+
+        # Log the attributes being sent for debugging
+        print(f"Attributes being sent to Salesforce: {filtered_attributes}")
 
-        # Convert Age__c to int if present
+        # Ensure Age__c is a number
         if "Age__c" in filtered_attributes:
             filtered_attributes["Age__c"] = int(filtered_attributes["Age__c"])
 
-        # Validate Gender__c against picklist
+        # Validate Gender__c against picklist values
         if "Gender__c" in filtered_attributes:
             gender_values = field_details.get("Gender__c", {}).get("picklist_values", [])
             if gender_values and filtered_attributes["Gender__c"] not in gender_values:
-                raise ValueError(f"Invalid Gender__c value '{filtered_attributes['Gender__c']}'. Allowed: {gender_values}")
+                raise ValueError(f"Invalid value for Gender__c: '{filtered_attributes['Gender__c']}'. Allowed values: {gender_values}")
 
-        # Create Salesforce record
+        # Create the record
         result = sf_object.create(filtered_attributes)
         return f"✅ Successfully created Patient Registration record with ID: {result['id']}."
 
     except Exception as e:
         return f"❌ Error interacting with Salesforce: {str(e)}"
 
+# Function to process the image and extract attributes
 def process_image(image):
     extracted_text = extract_text(image)
     if not extracted_text:
         return "No text detected in the image.", None, None
 
     attributes = extract_attributes(extracted_text)
+
+    # Ensure all attributes are present, even if empty, in the desired order
     ordered_attributes = {attr: attributes.get(attr, "") for attr in ATTRIBUTE_ORDER}
+
+    # Convert attributes to DataFrame for display
     df = pd.DataFrame(list(ordered_attributes.items()), columns=["Attribute", "Value"])
     return f"Extracted Text:\n{extracted_text}", df, None
 
+# Function to handle edited attributes and export to Salesforce
 def export_to_salesforce(edited_df):
     try:
+        # Convert edited DataFrame back to dictionary
         edited_attributes = dict(zip(edited_df["Attribute"], edited_df["Value"]))
+
+        # Export to Salesforce
         message = interact_with_salesforce(edited_attributes)
         return message
+
     except Exception as e:
         return f"❌ Error exporting to Salesforce: {str(e)}"
 
+# Gradio Interface
 def app():
     with gr.Blocks() as demo:
         with gr.Tab("📥 OCR Processing"):
@@ -152,6 +173,7 @@ def app():
             ok_button = gr.Button("OK")
             result_output = gr.Text(label="🚀 Result")
 
+        # Define button actions
         extract_button.click(
             fn=process_image,
             inputs=[image_input],
@@ -166,4 +188,4 @@ def app():
     return demo
 
 if __name__ == "__main__":
-    app().launch(share=True)
+    app().launch(share=True)