hh

app.py

@@ -38,11 +38,16 @@ def extract_text(doc):
         return doc.read().decode('utf-8')
 
     if doc.name.endswith(".pdf"):
-        raw = doc.read().decode('latin-1')
-        
-        with pdfplumber.open(io.BytesIO(raw)) as pdf:
-            pages = [page.extract_text() for page in pdf.pages]
-            return "\n".join(pages)
+        raw = doc.read()  
+
+        # Remove null bytes without decoding 
+        raw = raw.replace(b'\x00', b'')
+
+        pdf = pdfplumber.open(BytesIO(raw))
+        pages = [page.extract_text() for page in pdf.pages]      
+        return "\n".join(pages)
+    
+    
     if doc.name.endswith('.docx'):
         raw_text = doc.read()
         return docx2txt.process(raw_text)