Update main.py

main.py

@@ -2,7 +2,6 @@ from fastapi import FastAPI, UploadFile, HTTPException
 import subprocess
 import tempfile
 import os
-import uuid
 import pdfplumber
 import io
 
@@ -12,26 +11,14 @@ app = FastAPI()
 async def health_check():
     return {"status": "ok"}
 
-def convert_with_libreoffice(input_path: str, output_dir: str) -> str:
-    # بروفايل منفصل لكل تحويل، عشان نتجنب تعارض الـ lock بين الطلبات المتزامنة
-    unique_profile = f"/tmp/lo_profile_{uuid.uuid4().hex}"
-
-    subprocess.run(
-        [
-            "libreoffice", "--headless", "--norestore", "--nolockcheck",
-            f"-env:UserInstallation=file://{unique_profile}",
-            "--convert-to", "txt:Text",
-            "--outdir", output_dir, input_path
-        ],
-        check=True,
-        timeout=120  # رفعناها من 60 لـ 120 ثانية
-    )
-    txt_path = os.path.join(
-        output_dir,
-        os.path.splitext(os.path.basename(input_path))[0] + ".txt"
+def convert_with_pandoc(input_path: str) -> str:
+    result = subprocess.run(
+        ["pandoc", input_path, "-t", "plain"],
+        capture_output=True, text=True, timeout=20  # عادي جدًا 20 ثانية كافية
     )
-    with open(txt_path, "r", encoding="utf-8") as f:
-        return f.read()
+    if result.returncode != 0:
+        raise Exception(result.stderr)
+    return result.stdout
 
 @app.post("/extract-text")
 async def extract_text(file: UploadFile):
@@ -44,9 +31,11 @@ async def extract_text(file: UploadFile):
             with open(input_path, "wb") as f:
                 f.write(content)
             try:
-                text = convert_with_libreoffice(input_path, tmp)
+                text = convert_with_pandoc(input_path)
             except subprocess.TimeoutExpired:
-                raise HTTPException(504, "LibreOffice conversion timed out. Try again (cold start issue).")
+                raise HTTPException(504, "Pandoc conversion timed out.")
+            except Exception as e:
+                raise HTTPException(500, f"Pandoc error: {str(e)}")
 
     elif filename.endswith('.pdf'):
         text_parts = []