Update main.py

main.py

@@ -1,5 +1,7 @@
 from fastapi import FastAPI, UploadFile, HTTPException
-from docx import Document
+import subprocess
+import tempfile
+import os
 import pdfplumber
 import io
 
@@ -7,7 +9,22 @@ app = FastAPI()
 
 @app.get("/")
 async def health_check():
-    return {"status": "ok", "service": "extract-text"}
+    return {"status": "ok"}
+
+def convert_with_libreoffice(input_path: str, output_dir: str) -> str:
+    subprocess.run(
+        [
+            "libreoffice", "--headless", "--convert-to", "txt:Text",
+            "--outdir", output_dir, input_path
+        ],
+        check=True, timeout=60
+    )
+    txt_path = os.path.join(
+        output_dir,
+        os.path.splitext(os.path.basename(input_path))[0] + ".txt"
+    )
+    with open(txt_path, "r", encoding="utf-8") as f:
+        return f.read()
 
 @app.post("/extract-text")
 async def extract_text(file: UploadFile):
@@ -15,8 +32,11 @@ async def extract_text(file: UploadFile):
     filename = file.filename.lower()
 
     if filename.endswith('.docx'):
-        doc = Document(io.BytesIO(content))
-        text = "\n".join(p.text for p in doc.paragraphs if p.text.strip())
+        with tempfile.TemporaryDirectory() as tmp:
+            input_path = os.path.join(tmp, file.filename)
+            with open(input_path, "wb") as f:
+                f.write(content)
+            text = convert_with_libreoffice(input_path, tmp)
 
     elif filename.endswith('.pdf'):
         text_parts = []
@@ -25,12 +45,17 @@ async def extract_text(file: UploadFile):
                 page_text = page.extract_text()
                 if page_text:
                     text_parts.append(page_text)
+                for table in page.extract_tables():
+                    for row in table:
+                        row_text = " | ".join(str(c) for c in row if c)
+                        if row_text:
+                            text_parts.append(row_text)
         text = "\n".join(text_parts)
 
     else:
         raise HTTPException(400, "Unsupported file type. Only .docx and .pdf allowed.")
 
     if not text.strip():
-        raise HTTPException(422, "Could not extract any text from file (possibly scanned/image-based).")
+        raise HTTPException(422, "Could not extract any text from file.")
 
     return {"filename": file.filename, "extractedText": text}