Update app.py

app.py

@@ -4,7 +4,6 @@ import gradio as gr
 from transformers import pipeline
 from sentence_transformers import SentenceTransformer, util
 import PyPDF2
-import re
 
 # Set up logging with immediate writing
 logging.basicConfig(
@@ -19,35 +18,6 @@ logger = logging.getLogger()
 qa_model = pipeline("question-answering", model="distilbert-base-uncased-distilled-squad")
 embedder = SentenceTransformer('all-MiniLM-L6-v2')
 
-def preprocess_qa_format(text):
-    """
-    Detects Q/A pairs in different formats and returns only the answer text.
-    
-    Supported formats:
-    1. "Question X: ..." followed by "Answer: ..."
-    2. "Q X: ..." followed by "A: ..."
-    3. "Q X: ..." followed by an inferred answer line.
-    """
-    # Pattern for explicit "Question ..." and "Answer ..." pairs
-    pattern1 = re.compile(r"(?i)question\s*\d*\s*:\s*(.+?)\n\s*answer[:]*\s*(.+?)(?:\n|$)")
-    # Pattern for shorthand "Q ..." and "A: ..." pairs
-    pattern2 = re.compile(r"(?i)Q\s*\d*\s*:\s*(.+?)\n\s*A[:]*\s*(.+?)(?:\n|$)")
-    # Pattern for "Q ..." followed by an inferred answer (starting with a capital letter and ending with a period)
-    pattern3 = re.compile(r"(?i)Q\s*\d*\s*:\s*(.+?)\n\s*([A-Z][^.]*\..+?)(?:\n|$)")
-    
-    def replacer(match):
-        # We ignore the question text entirely and keep only the answer
-        answer_text = match.group(2).strip()
-        return f"{answer_text}\n"
-
-    # Apply the transformations
-    text = pattern1.sub(replacer, text)
-    text = pattern2.sub(replacer, text)
-    text = pattern3.sub(replacer, text)
-    return text
-
-
-
 # Helper function to extract text from PDF
 def extract_text_from_pdf(file_path):
     text = ""