Update update_docx_with_pdf.py

update_docx_with_pdf.py

@@ -1,19 +1,14 @@
 import openai
 import json
+import os
 
-# Set your OpenAI API key here
-OPENAI_API_KEY = "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"
+def update_json_with_pdf(word_json_file, pdf_txt_file, output_file):
+    # --- Load files ---
+    with open(word_json_file, "r", encoding="utf-8") as f:
+        word_json = f.read()
+    with open(pdf_txt_file, "r", encoding="utf-8") as f:
+        pdf_txt = f.read()
 
-# Load PDF text
-WORD_JSON_FILE = "word_red_data.json"
-PDF_TEXT_FILE = "pdf_all_text_full.txt"
-OUTPUT_FILE = "updated_word_data1.json"
-
-# --- Load files ---
-with open(WORD_JSON_FILE, "r", encoding="utf-8") as f:
-    word_json = f.read()
-with open(PDF_TEXT_FILE, "r", encoding="utf-8") as f:
-    pdf_txt = f.read()
 
 # --- Build prompt ---
 user_prompt = f"""
@@ -30,17 +25,20 @@ Instructions:
 - Make sure the JSON is valid and ready to use.
 """
 
-# --- Call OpenAI API (no env var needed) ---
-client = openai.OpenAI(api_key=OPENAI_API_KEY)
-response = client.chat.completions.create(
-    model="gpt-4o",
-    messages=[
-        {"role": "system", "content": "You are a data extraction assistant. Only reply with valid JSON. Do not add any extra text or formatting. Do NOT use markdown/code blocks, just output JSON."},
-        {"role": "user", "content": user_prompt}
-    ],
-    max_tokens=4096,
-    temperature=0
-)
+ # --- Call OpenAI API ---
+    api_key = os.environ.get("OPENAI_API_KEY")
+    if not api_key:
+        raise RuntimeError("OPENAI_API_KEY not found in environment variables!")
+    client = openai.OpenAI(api_key=api_key)
+    response = client.chat.completions.create(
+        model="gpt-4o",
+        messages=[
+            {"role": "system", "content": "You are a data extraction assistant. Only reply with valid JSON. Do not add any extra text or formatting. Do NOT use markdown/code blocks, just output JSON."},
+            {"role": "user", "content": user_prompt}
+        ],
+        max_tokens=4096,
+        temperature=0
+    )
 
 updated_json_str = response.choices[0].message.content.strip()
 
@@ -53,4 +51,11 @@ try:
 except Exception as e:
     print("⚠️ Model did not return valid JSON. Raw output below:\n")
     print(updated_json_str)
-    print("\n❌ Failed to parse updated JSON:", e)
+    print("\n❌ Failed to parse updated JSON:", e)
+
+if __name__ == "__main__":
+    import sys
+    if len(sys.argv) != 4:
+        print("Usage: python update_docx_with_pdf.py <word_json_file> <pdf_txt_file> <output_json_file>")
+        exit(1)
+    update_json_with_pdf(sys.argv[1], sys.argv[2], sys.argv[3])