Update backend/main.py

backend/main.py

@@ -6,6 +6,8 @@ from pathlib import Path
 from typing import List
 
 import fitz
+import pytesseract
+from pdf2image import convert_from_path
 from groq import Groq
 from fastapi import FastAPI, File, UploadFile, HTTPException
 from fastapi.middleware.cors import CORSMiddleware
@@ -87,10 +89,22 @@ Return ONLY valid JSON (no markdown fences, no extra text) using this schema:
 def extract_text_from_pdf(pdf_path: str) -> str:
     text = ""
 
-    doc = fitz.open(pdf_path)
+    # ---------- Try PyMuPDF extraction ----------
+    try:
+        doc = fitz.open(pdf_path)
+        for page in doc:
+            text += page.get_text()
+    except Exception:
+        pass
 
-    for page in doc:
-        text += page.get_text()
+    # ---------- OCR fallback if no text ----------
+    if not text.strip():
+        try:
+            images = convert_from_path(pdf_path)
+            for img in images:
+                text += pytesseract.image_to_string(img)
+        except Exception:
+            raise ValueError("PDF text extraction failed.")
 
     if not text.strip():
         raise ValueError("PDF text extraction failed.")
@@ -98,7 +112,7 @@ def extract_text_from_pdf(pdf_path: str) -> str:
     if len(text.strip()) < 500:
         text = text[:2000]
 
-    # Try to isolate Methods section
+    # ---------- Try to isolate Methods section ----------
     methods_match = re.search(
         r'(?:^\s*(?:\d+\.?\d*\.?\s+)?'
         r'(?:materials?\s+and\s+methods?|experimental\s+procedures?'
@@ -113,7 +127,7 @@ def extract_text_from_pdf(pdf_path: str) -> str:
         text = methods_match.group(2)
     else:
         chars = len(text)
-        text  = text[chars // 10: chars // 10 + 15000]
+        text = text[chars // 10: chars // 10 + 15000]
 
     words = text.split()
     if len(words) > 1500: