Update app.py

app.py

@@ -2,58 +2,41 @@ import gradio as gr
 from docx import Document
 from docx.shared import RGBColor
 from difflib import get_close_matches
-from utils import extract_text_from_pdf, parse_pdf_to_dict
+from pdf_parser import extract_text_from_pdf, parse_data_blocks
 
-# Improved red color detection
 def is_red_color(run):
     color = run.font.color
-    if color is None:
+    if not color or not color.rgb:
         return False
-    if color.rgb:
-        # Try detecting red tones even if slightly off
-        red_shades = [
-            RGBColor(255, 0, 0),
-            RGBColor(200, 0, 0),
-            RGBColor(255, 20, 20),
-            RGBColor(192, 0, 0),
-            RGBColor(220, 20, 60),
-        ]
-        return any(color.rgb == shade for shade in red_shades)
-    return False
+    r, g, b = color.rgb[0], color.rgb[1], color.rgb[2]
+    return r >= 200 and g <= 100 and b <= 100  # red-dominant
 
 def replace_red_text_with_data(doc_path, data_dict):
     doc = Document(doc_path)
-    matched = 0
 
     for para in doc.paragraphs:
         for run in para.runs:
             if is_red_color(run):
-                red_text = run.text.strip()
-                print(f"Found red text: {red_text}")
-                if not red_text:
-                    continue
-                # Use fuzzy matching
-                key_match = get_close_matches(red_text.lower(), [k.lower() for k in data_dict], n=1, cutoff=0.6)
-                if key_match:
+                original_text = run.text.strip()
+                # Try exact or close match
+                match = get_close_matches(original_text.lower(), [k.lower() for k in data_dict.keys()], n=1, cutoff=0.6)
+                if match:
                     for key in data_dict:
-                        if key.lower() == key_match[0]:
-                            print(f"Replacing '{red_text}' with '{data_dict[key]}'")
+                        if key.lower() == match[0]:
                             run.text = data_dict[key]
-                            matched += 1
                             break
-    print(f"Total replacements: {matched}")
     return doc
 
-def process_files(pdf_file, template_docx):
-    pdf_path = pdf_file
-    doc_path = template_docx
-    output_path = "filled_output.docx"
-
-    raw_text = extract_text_from_pdf(pdf_path)
-    data_dict = parse_pdf_to_dict(raw_text)
+def process_files(pdf_file, word_template):
+    # Extract data from PDF
+    raw_text = extract_text_from_pdf(pdf_file)
+    data_dict = parse_data_blocks(raw_text)
 
-    final_doc = replace_red_text_with_data(doc_path, data_dict)
+    # Replace red text in Word
+    final_doc = replace_red_text_with_data(word_template, data_dict)
 
+    # Save and return output
+    output_path = "filled_output.docx"
     final_doc.save(output_path)
     return output_path
 
@@ -63,9 +46,9 @@ demo = gr.Interface(
         gr.File(label="Upload PDF Report", file_types=[".pdf"]),
         gr.File(label="Upload Word Template (.docx)", file_types=[".docx"])
     ],
-    outputs=gr.File(label="Download Filled Report (.docx)"),
-    title="Audit Report Generator",
-    description="Upload a PDF and a Word template. This tool will auto-fill red-colored outdated text with data from the PDF."
+    outputs=gr.File(label="Download Updated Word (.docx)"),
+    title="Audit Report Auto-Filler",
+    description="Replaces outdated red text in Word using updated values from a PDF report."
 )
 
 if __name__ == "__main__":