Update app.py

app.py

@@ -1,4 +1,4 @@
-from transformers import AutoProcessor, AutoModelForImageClassification
+from transformers import BlipProcessor, BlipForConditionalGeneration
 from PIL import Image
 import gradio as gr
 import torch
@@ -12,6 +12,7 @@ import os
 from dotenv import load_dotenv
 import base64
 import io
+import shutil
 import concurrent.futures
 
 # Load environment variables from .env file
@@ -29,24 +30,27 @@ except Exception as e:
     sf = None
     print(f"Failed to connect to Salesforce: {str(e)}")
 
-# Load ViT model and processor (generic ImageNet pretrained)
-processor = AutoProcessor.from_pretrained("google/vit-base-patch16-224")
-model = AutoModelForImageClassification.from_pretrained("google/vit-base-patch16-224")
+# Load BLIP model and processor
+processor = BlipProcessor.from_pretrained("Salesforce/blip-image-captioning-base")
+model = BlipForConditionalGeneration.from_pretrained("Salesforce/blip-image-captioning-base")
 model.eval()
 device = "cuda" if torch.cuda.is_available() else "cpu"
 model.to(device)
 
-# Inference function to classify image and get predicted label
+# Inference function to generate captions dynamically based on image content
 def generate_captions_from_image(image):
     if image.mode != "RGB":
         image = image.convert("RGB")
-    inputs = processor(images=image, return_tensors="pt").to(device)
-    with torch.no_grad():
-        outputs = model(**inputs)
-    logits = outputs.logits
-    predicted_class_idx = logits.argmax(-1).item()
-    predicted_label = model.config.id2label[predicted_class_idx]
-    return predicted_label
+    
+    # Resize image for faster processing (use smaller resolution to speed up inference)
+    image = image.resize((320, 320))  # Reduced size for faster processing
+    
+    # Preprocess the image and generate a caption
+    inputs = processor(image, return_tensors="pt").to(device, torch.float16)
+    output = model.generate(**inputs, max_new_tokens=50)
+    caption = processor.decode(output[0], skip_special_tokens=True)
+    
+    return caption
 
 # Function to save DPR text to a PDF file
 def save_dpr_to_pdf(dpr_text, image_paths, captions, filename):
@@ -89,14 +93,16 @@ def save_dpr_to_pdf(dpr_text, image_paths, captions, filename):
             else:
                 flowables.append(Spacer(1, 12))
         
-        # Add images and captions in the correct order
+        # Add images and captions in the correct order (no need to add description to dpr_text again)
         for img_path, caption in zip(image_paths, captions):
             try:
+                # Add image first
                 img = PDFImage(img_path, width=200, height=150)  # Adjust image size if needed
                 flowables.append(img)
+                # Add description below the image
                 description = f"Description: {caption}"
                 flowables.append(Paragraph(description, body_style))
-                flowables.append(Spacer(1, 12))
+                flowables.append(Spacer(1, 12))  # Add some space between images
             except Exception as e:
                 flowables.append(Paragraph(f"Error loading image: {str(e)}", body_style))
         
@@ -106,15 +112,18 @@ def save_dpr_to_pdf(dpr_text, image_paths, captions, filename):
     except Exception as e:
         return f"Error saving PDF: {str(e)}", None
 
-# Function to upload file to Salesforce as ContentVersion
+# Function to upload a file to Salesforce as ContentVersion
 def upload_file_to_salesforce(file_path, filename, sf_connection, file_type):
     try:
+        # Read file content and encode in base64
         with open(file_path, 'rb') as f:
             file_content = f.read()
         file_content_b64 = base64.b64encode(file_content).decode('utf-8')
         
+        # Set description based on file type
         description = "Daily Progress Report PDF" if file_type == "pdf" else "Site Image"
         
+        # Create ContentVersion
         content_version = sf_connection.ContentVersion.create({
             'Title': filename,
             'PathOnClient': filename,
@@ -122,60 +131,68 @@ def upload_file_to_salesforce(file_path, filename, sf_connection, file_type):
             'Description': description
         })
         
+        # Get ContentDocumentId
         content_version_id = content_version['id']
         content_document = sf_connection.query(
             f"SELECT ContentDocumentId FROM ContentVersion WHERE Id = '{content_version_id}'"
         )
         content_document_id = content_document['records'][0]['ContentDocumentId']
         
+        # Generate a valid Salesforce URL for the ContentDocument
         content_document_url = f"https://{sf_connection.sf_instance}/sfc/servlet.shepherd/version/download/{content_version_id}"
-    
         return content_document_id, content_document_url, f"File {filename} uploaded successfully"
     except Exception as e:
         return None, None, f"Error uploading {filename} to Salesforce: {str(e)}"
 
-# Generate DPR, save PDF, upload to Salesforce
+# Function to generate the daily progress report (DPR), save as PDF, and upload to Salesforce
 def generate_dpr(files):
     dpr_text = []
     captions = []
     image_paths = []
     current_time = datetime.now().strftime("%Y-%m-%d %H:%M:%S")
     
+    # Add header to the DPR
     dpr_text.append(f"Daily Progress Report\nGenerated on: {current_time}\n")
     
+    # Process images in parallel for faster performance
     with concurrent.futures.ThreadPoolExecutor() as executor:
         results = list(executor.map(lambda file: generate_captions_from_image(Image.open(file.name)), files))
     
     for i, file in enumerate(files):
         caption = results[i]
         captions.append(caption)
+        
+        # Generate DPR section for this image with dynamic caption
         dpr_section = f"\nImage: {file.name}\nDescription: {caption}\n"
         dpr_text.append(dpr_section)
+        
+        # Save image path for embedding in the report
         image_paths.append(file.name)
     
+    # Combine DPR text
     dpr_output = "\n".join(dpr_text)
+    
+    # Generate PDF filename with timestamp
     pdf_filename = f"DPR_{datetime.now().strftime('%Y-%m-%d_%H-%M-%S')}.pdf"
     
+    # Save DPR text to PDF
     pdf_result, pdf_filepath = save_dpr_to_pdf(dpr_output, image_paths, captions, pdf_filename)
     
-    salesforce_result = ""
-    pdf_content_document_id = None
-    pdf_url = None
-    
     if sf and pdf_filepath:
         try:
-            report_description = "; ".join(captions)[:255]
+            # Create Daily_Progress_Reports__c record
+            report_description = "; ".join(captions)[:255]  # Concatenate captions, limit to 255 chars
             dpr_record = sf.Daily_Progress_Reports__c.create({
-                'Detected_Activities__c': report_description
+                'Detected_Activities__c': report_description  # Store in Detected_Activities__c field
             })
             dpr_record_id = dpr_record['id']
-            salesforce_result += f"Created Daily_Progress_Reports__c record with ID: {dpr_record_id}\n"
             
+            # Upload PDF to Salesforce
             pdf_content_document_id, pdf_url, pdf_upload_result = upload_file_to_salesforce(
                 pdf_filepath, pdf_filename, sf, "pdf"
             )
-            salesforce_result += pdf_upload_result + "\n"
             
+            # Link PDF to DPR record
             if pdf_content_document_id:
                 sf.ContentDocumentLink.create({
                     'ContentDocumentId': pdf_content_document_id,
@@ -183,38 +200,53 @@ def generate_dpr(files):
                     'ShareType': 'V'
                 })
             
+            # Update the DPR record with the PDF URL
             if pdf_url:
                 sf.Daily_Progress_Reports__c.update(dpr_record_id, {
-                    'PDF_URL__c': pdf_url
+                    'PDF_URL__c': pdf_url  # Storing the PDF URL correctly
                 })
-                salesforce_result += f"Updated PDF URL for record ID {dpr_record_id}\n"
             
+            # Upload images to Salesforce and link them to DPR record
             for file in files:
                 image_filename = os.path.basename(file.name)
                 image_content_document_id, image_url, image_upload_result = upload_file_to_salesforce(
                     file.name, image_filename, sf, "image"
                 )
+                
                 if image_content_document_id:
+                    # Link image to the Daily Progress Report record (DPR) using ContentDocumentLink
                     sf.ContentDocumentLink.create({
                         'ContentDocumentId': image_content_document_id,
-                        'LinkedEntityId': dpr_record_id,
-                        'ShareType': 'V'
+                        'LinkedEntityId': dpr_record_id,  # Link image to DPR record
+                        'ShareType': 'V'  # 'V' means Viewer access
                     })
+                    
+                    # Now, update the DPR record with the ContentDocumentId in the Site_Images field (if it's a text or URL field)
                     sf.Daily_Progress_Reports__c.update(dpr_record_id, {
-                        'Site_Images__c': image_content_document_id
+                        'Site_Images__c': image_content_document_id  # Storing the ContentDocumentId directly
                     })
-                    salesforce_result += image_upload_result + "\n"
                         
         except Exception as e:
-            salesforce_result += f"Error interacting with Salesforce: {str(e)}\n"
-    else:
-        salesforce_result = "Salesforce connection not available or PDF generation failed.\n"
+            pass  # No output for Salesforce errors now
     
-    return (
-        dpr_output + f"\n\n{pdf_result}\n\nSalesforce Upload Status:\n{salesforce_result}",
-        pdf_filepath
-    )
+    # Return the PDF file for Gradio download (using shutil to copy and return the file)
+    if pdf_filepath:
+        # Copy the PDF file to a temporary directory for Gradio to serve it
+        temp_pdf_path = "/tmp/" + os.path.basename(pdf_filepath)
+        shutil.copy(pdf_filepath, temp_pdf_path)
+        
+        # Only return the DPR output and the PDF file path, excluding Salesforce upload details
+        return (
+            dpr_output + f"\n\n{pdf_result}",  # Removed Salesforce upload status
+            temp_pdf_path  # Returning the file path for download
+        )
+    else:
+        return (
+            dpr_output + f"\n\n{pdf_result}",  # Removed Salesforce upload status
+            None
+        )
 
+# Gradio interface for uploading multiple files, displaying DPR, and downloading PDF
 iface = gr.Interface(
     fn=generate_dpr,
     inputs=gr.Files(type="filepath", label="Upload Site Photos"),