tables zip file download implemented

app.py

@@ -11,6 +11,7 @@ import shutil
 import zipfile
 from io import BytesIO
 temp_figure_dir="pdf_figures/"
+temp_table_dir="pdf_tables/"
 import time
 # get https://github.com/oschwartz10612/poppler-windows/releases/tag/v22.01.0-0
 # poppler-utils:
@@ -168,6 +169,7 @@ if 'page_count' in st.session_state:
         st.session_state.color_image_list = []
         st.session_state.gray_image_np_list = []
         st.session_state.pdf_figures_image_list=[]
+        st.session_state.pdf_tables_image_list = []
         pdf_tables_image_list=[]
         st.session_state.pdf_text_list=[]
 
@@ -185,20 +187,24 @@ if 'page_count' in st.session_state:
         st.session_state.extracted_text = ""
 
         manage_temp_to_be_zipped_directory(temp_figure_dir)
-
+        manage_temp_to_be_zipped_directory(temp_table_dir)
         for index, gray_pdf_image_np in enumerate(st.session_state.gray_image_np_list):
           print("index="+str(index))
 
           figures_image_list,tables_image_list,text=utils.gray_pdf_image_np_to_text(index,gray_pdf_image_np, debug=True)
           st.session_state.pdf_figures_image_list.append(figures_image_list)
+          st.session_state.pdf_tables_image_list.append(tables_image_list)
           if st.session_state.pdf_figures_image_list[index]:
               for pdf_figure_text_image in st.session_state.pdf_figures_image_list[index]:
                   raw_image_file_name = f"page_{index}_{pdf_figure_text_image[0]}.png"
                   cleaned_image_file_name = clean_filename(raw_image_file_name)
                   Image.fromarray(pdf_figure_text_image[1]).save(temp_figure_dir+cleaned_image_file_name)
+          if st.session_state.pdf_tables_image_list:
+              for pdf_table_text_image in st.session_state.pdf_tables_image_list[index]:
+                  raw_image_file_name = f"page_{index}_{pdf_table_text_image[0]}.png"
+                  cleaned_image_file_name = clean_filename(raw_image_file_name)
+                  Image.fromarray(pdf_table_text_image[1]).save(temp_table_dir + cleaned_image_file_name)
 
-
-          pdf_tables_image_list.append(tables_image_list)
           st.session_state.pdf_text_list.append(text)
           st.session_state.extracted_text=st.session_state.extracted_text+f"<Page {index+1} start>\n" + text + f"\n<Page {index+1} end>\n>"
           # st.write(text)
@@ -206,6 +212,7 @@ if 'page_count' in st.session_state:
           progress_percentage = (index) / (st.session_state.num_pages_to_extract - 1)
           read_pdf_progress_bar.progress(progress_percentage)
         st.session_state.figure_zip_bytes=zip_directory(temp_figure_dir)
+        st.session_state.table_zip_bytes = zip_directory(temp_table_dir)
         #add_animation_to_image()
         #st.session_state['video_generated'] = True
         st.rerun()
@@ -218,7 +225,7 @@ if 'page_count' in st.session_state:
                            mime="text/plain")
 
         download_figure_zip_file_name = uploaded_locked_pdf_file.name.replace(".pdf", "_figures.zip")
-
+        download_table_zip_file_name = uploaded_locked_pdf_file.name.replace(".pdf", "_tables.zip")
         st.download_button(
             label="Download Figures ZIP",
             data=st.session_state.figure_zip_bytes,
@@ -226,7 +233,12 @@ if 'page_count' in st.session_state:
             mime="application/zip"
         )
 
-
+        st.download_button(
+            label="Download Tables ZIP",
+            data=st.session_state.table_zip_bytes,
+            file_name=download_table_zip_file_name,
+            mime="application/zip"
+        )
 
         #  st.image(Image.fromarray(bgr_image))
         # for index,pdf_text in enumerate(st.session_state.pdf_text_list):
@@ -239,6 +251,12 @@ if 'page_count' in st.session_state:
                 for pdf_figure_text_image in st.session_state.pdf_figures_image_list[index]:
                     st.write(pdf_figure_text_image[0])
                     st.image(Image.fromarray(pdf_figure_text_image[1]))
+            if not st.session_state.pdf_tables_image_list[index]:
+                st.write("no tables")
+            else:
+                for pdf_table_text_image in st.session_state.pdf_tables_image_list[index]:
+                    st.write(pdf_table_text_image[0])
+                    st.image(Image.fromarray(pdf_table_text_image[1]))