Update app.py

app.py

@@ -73,16 +73,18 @@ def main():
         st.session_state.file1 = None
     if "file2" not in st.session_state:
         st.session_state.file2 = None
+    if "extracted_texts" not in st.session_state:
+        st.session_state.extracted_texts = {"file1": None, "file2": None, "extracted_text_1": [], "extracted_text_2": []}
     st.markdown('<div style="text-align: center;">' + '<h1>PDF Upload and Compare App</h1>' + '</div>', unsafe_allow_html=True)
     
     # Create columns for side-by-side buttons
     col1, col2 = st.columns([2, 2])
         
     # Add upload button to left column
-    st.session_state.file1 = col1.file_uploader("**Text 1**", type="pdf")
+    st.session_state.file1 = col1.file_uploader("**PDF 1**", type="pdf")
     
     # Add upload button to right column
-    st.session_state.file2 = col2.file_uploader("**Text 2**", type="pdf")
+    st.session_state.file2 = col2.file_uploader("**PDF 2**", type="pdf")
 
     
     # Check if both files are uploaded
@@ -93,10 +95,11 @@ def main():
         
         if st.session_state.display_data["file1"] != st.session_state.file1 or st.session_state.display_data["file2"] != st.session_state.file2:
             st.session_state.display_data = {"file1": st.session_state.file1, "file2": st.session_state.file2, "i": 0}
+            st.session_state.extracted_texts = {"file1": st.session_state.file1, "file2": st.session_state.file2, "extracted_text_1": extract_text_with_pypdf(st.session_state.file1), "extracted_text_2": extract_text_with_pypdf(st.session_state.file2)}
         
         try:
-            extracted_text1 = extract_text_with_pypdf(st.session_state.file1)
-            extracted_text2 = extract_text_with_pypdf(st.session_state.file2)
+            extracted_text1 = st.session_state.extracted_texts["extracted_text_1"]
+            extracted_text2 = st.session_state.extracted_texts["extracted_text_2"]
             with col1.expander(filename1):
                 st.write("\n\n".join("\n\n------------------------\n\n".join(extracted_text1).splitlines()))
             
@@ -137,7 +140,7 @@ def main():
                         reverse_pdata = [{'text1': d['text2'], 'text2': d['text1'], 'explanation': d['explanation']} for d in pdata]
                         st.session_state.differences_data.append({"etext1": etext2, "etext2": etext1, "pdata": reverse_pdata, "dext1": dext2, "dext2": dext1})
                     
-                    display_text = "\n\n\n".join([f"**Text1:**\n\n{d['text1']}\n\n**Text2:**\n\n{d['text2']}\n\n**Explanation:**\n\n{d['explanation']}\n\n----------------------\n" for d in pdata])
+                    display_text = "\n\n\n".join([f"**PDF 1:**\n\n{d['text1']}\n\n**PDF 2:**\n\n{d['text2']}\n\n**Explanation:**\n\n{d['explanation']}\n\n----------------------\n" for d in pdata])
     
                     with st.expander(f"**Page {i+1}** - {filename1}"):
                         st.markdown("\n\n".join(dext1.splitlines()), unsafe_allow_html=True)