whackthejacker/DataHubHub

1import streamlit as st2import pandas as pd3import json4 5def render_dataset_preview(dataset, dataset_type):6    """7    Renders a preview of the dataset with pagination options.8    9    Args:10        dataset: The dataset to preview (pandas DataFrame)11        dataset_type: The type of dataset (csv, json, etc.)12    """13    if dataset is None:14        st.warning("No dataset to preview.")15        return16    17    st.markdown(f"<h3>Dataset Preview: {st.session_state.dataset_name}</h3>", unsafe_allow_html=True)18    19    # Show basic info20    col1, col2, col3 = st.columns(3)21    with col1:22        st.metric("Rows", f"{dataset.shape[0]:,}")23    with col2:24        st.metric("Columns", f"{dataset.shape[1]:,}")25    with col3:26        st.metric("Type", dataset_type.upper())27    28    # Preview options29    col1, col2 = st.columns([1, 3])30    with col1:31        num_rows = st.number_input("Rows to display", min_value=5, max_value=100, value=10, step=5)32    with col2:33        preview_mode = st.radio("Preview mode", ["Head", "Tail", "Sample"], horizontal=True)34    35    # Display dataset preview36    st.markdown("<div class='dataset-preview'>", unsafe_allow_html=True)37    38    if preview_mode == "Head":39        st.dataframe(dataset.head(num_rows), use_container_width=True)40    elif preview_mode == "Tail":41        st.dataframe(dataset.tail(num_rows), use_container_width=True)42    else:  # Sample43        st.dataframe(dataset.sample(min(num_rows, len(dataset))), use_container_width=True)44    45    st.markdown("</div>", unsafe_allow_html=True)46    47    # Show dataset schema48    with st.expander("Dataset Schema"):49        col1, col2 = st.columns(2)50        51        with col1:52            st.markdown("**Column Types**")53            type_df = pd.DataFrame({54                'Column': dataset.dtypes.index,55                'Type': dataset.dtypes.values.astype(str)56            })57            st.dataframe(type_df, use_container_width=True)58        59        with col2:60            st.markdown("**Missing Values**")61            missing_df = pd.DataFrame({62                'Column': dataset.columns,63                'Missing': dataset.isna().sum().values,64                'Percentage': dataset.isna().sum().values / len(dataset) * 10065            })66            st.dataframe(missing_df.style.format({67                'Percentage': '{:.2f}%'68            }), use_container_width=True)69    70    # Raw data71    with st.expander("Raw Data (First 5 records)"):72        if dataset_type == 'csv':73            st.code(dataset.head(5).to_csv(index=False), language="text")74        else:  # json or jsonl75            st.code(dataset.head(5).to_json(orient='records', indent=2), language="json")76