app.py

import streamlit as st
import json
from datasets import load_dataset

st.set_page_config(page_title="Large GitHub Issues", layout="wide")
st.title("Issues with large text")

@st.cache()
def load_data():
    ds = load_dataset("loubnabnl/large-text-issues", split="train")
    return ds

def print_issue(events):
    for event in events:
        st.markdown("""---""")
        masked_author = f"masked as {event['masked_author']}" if "masked_author" in event else ""
        st.markdown(f"**Author:** {event['author']} {masked_author}, {event['action']} {event['type']} with title: {event['title']}.\
        Text size is: **{event['size']}** and Number of lines is: **{event['nb_lines']}**")
        st.code(f"{event['text']}", language="none")

samples = load_data()
col1, _ = st.columns([2, 4])
with col1:
    index_example = st.number_input(f"Index of the chosen conversation from the existing {len(samples)}", min_value=0, max_value=len(samples)-1, value=0, step=1)

st.write(f"Issue size: {samples[index_example]['text_size_no_bots']}\n\n")
print_issue(samples[index_example]["events"])