Java requirement

🎈 Using Streamlit
1

Hello,Preformatted text
I am building a streamlit app to extract/download Tables from PDF using tabula library.
Here is the code :

import tempfile
import streamlit as st
import tabula as tb 
import pandas as pd

uploaded_file2 = st.sidebar.file_uploader("Upload PDF", type=["pdf"], label_visibility="visible")

if uploaded_file2 :

    tables = tb.read_pdf(uploaded_file2, pages='all')

    colpi1_t, colpi2_t, colpi3_t = st.columns([40,20,40])

    with colpi2_t:
        tabe_idx = st.number_input(label=f"{len(tables)} Tables", value=1, min_value=1,
            max_value=len(tables), step=1, label_visibility="visible")

    temp = tempfile.NamedTemporaryFile(delete=True)
    temp_filename = temp.name + '.xlsx'

    with pd.ExcelWriter(temp_filename) as writer:
        for i in range (len(tables)):
            tables[i].to_excel(writer, sheet_name=f"table{i}")

    with open(writer, "rb") as f:
        binary_data = f.read()

    excel_namefile = f"Tables_{uploaded_file2.name[:-4]}.xlsx"

    st.download_button(
        label="Download",
        data=binary_data,
        file_name=excel_namefile)

   st.dataframe(tables[tabe_idx-1].style.format(na_rep='No Data', precision=1), use_container_width=True)

this app run correctly in local, but when hosted, i got this eror : tabula.errors.JavaNotFoundError: java command is not found from this Python process.Please ensure Java is installed and PATH is set for java

Can anyone help me fix this.

Thank you all.

2

Then you need to install java in your cloud environment. You use packages.txt to install dependencies (other than python packages). I think the package you need is default-jre-headless.

1 Like
3

Hello, Thank you for your answer.
Can you give me an example ?, i am new in IT development and i don’t know many things about cloud environment.

4

Create the file packages.txt, write default-jre-headless in it, that’s all. There is an example in the link I posted, just with two other packages that you don’t need.

2 Likes
5

Thank you so much, it works !