Update app.py

app.py

@@ -244,9 +244,6 @@ else:
 
 
     # Step 4: Summarization (Using full text)
-    st.subheader("Summarization")
-
-    # Display full extracted text (not just first 1000 characters)
     input_text = st.text_area(
         "Enter text to summarize",
         height=400,
@@ -256,31 +253,34 @@ else:
     if st.button("Summarize"):
         text_to_summarize = st.session_state["pdf_text"].strip() if st.session_state["pdf_text"] else input_text.strip()
     
-        # Debugging: Print input length
-        st.write(f"**Text Length:** {len(text_to_summarize.split())} words")
-        
-        # Handle empty text case
+        # Debugging: Print text length before processing
+        st.write(f"Original text length: {len(text_to_summarize.split())} words")
+    
         if not text_to_summarize:
-            st.error("No text provided for summarization!")
-        elif len(text_to_summarize.split()) < 20:
-            st.error("The text is too short! Provide at least 20 words.")
+            st.error("No text provided for summarization.")
         else:
             try:
                 with st.spinner("Generating summary..."):
-                    max_input_length = 1024  # Adjust max input length for BART
-                    text_to_summarize = " ".join(text_to_summarize.split()[:max_input_length])  #Trim long text
+                    # Tokenize input and truncate properly
+                    input_tokens = tokenizer.encode(text_to_summarize, truncation=True, max_length=1024)
+                    truncated_text = tokenizer.decode(input_tokens)
+    
+                    # Debugging: Check tokenized text length
+                    st.write(f"Tokenized length: {len(input_tokens)} tokens")
     
                     summary = summarizer(
-                        text_to_summarize,
+                        truncated_text,
                         max_length=256,  # Keep max summary length reasonable
                         min_length=50,
                         do_sample=False
                     )
-                    st.write("**Summary:**")
+    
+                    st.write("Summary:")
                     st.success(summary[0]["summary_text"])
-            
-            except IndexError as e:
-                st.error(f"**Summarization failed: Index out of range.**")
-                st.write(f"Debugging Info:\n- Text Length: {len(text_to_summarize.split())} words\n- Model: {summarizer.model}")
+    
+            except IndexError:
+                st.error("Summarization failed: Index out of range.")
+                st.write(f"Debugging Info:\n- Original text length: {len(text_to_summarize.split())} words\n- Tokenized length: {len(input_tokens)} tokens\n- Model: {summarizer.model}")
             except Exception as e:
-                st.error(f"**Summarization failed:** {e}")
+                st.error(f"Summarization failed: {e}")
+