Update app.py

app.py

@@ -1,61 +1,46 @@
-# app.py
 import gradio as gr
 import tempfile
-from pdf_extractor import extract_label_value_pairs
-from word_extractor import extract_red_text_with_labels, is_red_font
-from docx import Document
-from docx.shared import RGBColor
-import difflib
-
-
-def find_best_match_label(target_label, pdf_data):
-    keys = list(pdf_data.keys())
-    match = difflib.get_close_matches(target_label.lower(), keys, n=1, cutoff=0.4)
-    return match[0] if match else None
-
-
-def replace_red_text_by_label(word_path, label_value_map):
-    doc = Document(word_path)
-
-    for table in doc.tables:
-        for row in table.rows:
-            cells = row.cells
-            if len(cells) >= 2:
-                label = cells[0].text.strip().replace(":", "").replace("\n", " ")
-                matched_label = find_best_match_label(label, label_value_map)
-                if not matched_label:
-                    continue
-
-                new_value = label_value_map[matched_label]
-
-                for para in cells[1].paragraphs:
-                    for run in para.runs:
-                        if is_red_font(run):
-                            run.text = new_value
-                            run.font.color.rgb = RGBColor(0, 0, 0)  # make black
-
-    temp_dir = tempfile.mkdtemp()
-    updated_path = f"{temp_dir}/updated.docx"
-    doc.save(updated_path)
-    return updated_path
-
+import os
+import subprocess
 
 def process_files(pdf_file, word_file):
-    pdf_path = pdf_file.name
-    word_path = word_file.name
-
-    pdf_data = extract_label_value_pairs(pdf_path)           # {label: value}
-    word_data = extract_red_text_with_labels(word_path)      # {label: [red_texts]}
-
-    updated_doc_path = replace_red_text_by_label(word_path, pdf_data)
-    return updated_doc_path
-
+    # 1. Save the uploaded files to temp
+    temp_dir = tempfile.mkdtemp()
+    pdf_path = os.path.join(temp_dir, "input.pdf")
+    word_path = os.path.join(temp_dir, "input.docx")
+    with open(pdf_path, "wb") as f:
+        f.write(pdf_file.read())
+    with open(word_path, "wb") as f:
+        f.write(word_file.read())
+
+    # 2. Step 1: Extract PDF data to txt
+    pdf_txt_path = os.path.join(temp_dir, "pdf_data.txt")
+    subprocess.run(["python", "extract_pdf_data.py", pdf_path, pdf_txt_path], check=True)
+
+    # 3. Step 2: Extract red text from Word to JSON
+    word_json_path = os.path.join(temp_dir, "word_data.json")
+    subprocess.run(["python", "extract_red_text.py", word_path, word_json_path], check=True)
+
+    # 4. Step 3: Update docx JSON with PDF txt, output updated JSON
+    updated_json_path = os.path.join(temp_dir, "updated_word_data.json")
+    subprocess.run([
+        "python", "update_docx_with_pdf.py", word_json_path, pdf_txt_path, updated_json_path
+    ], check=True)
+
+    # 5. Step 4: Compare word file with updated JSON and update docx
+    final_docx_path = os.path.join(temp_dir, "updated.docx")
+    subprocess.run([
+        "python", "updated_word.py", word_path, updated_json_path, final_docx_path
+    ], check=True)
+
+    # 6. Return final docx file
+    return final_docx_path
 
 gr.Interface(
     fn=process_files,
     inputs=[
-        gr.File(label="Upload PDF File", type="filepath"),
-        gr.File(label="Upload Word File", type="filepath")
+        gr.File(label="Upload PDF File"),
+        gr.File(label="Upload Word File")
     ],
     outputs=gr.File(label="Download Updated Word File"),
     title="Red Text Replacer",