Bagikan melalui


Tutorial: Mengembangkan aplikasi Databricks dengan Streamlit

Tutorial ini menunjukkan cara membuat aplikasi Databricks menggunakan Konektor SQL Databricks untuk Python dan Streamlit. Anda akan mempelajari cara mengembangkan aplikasi yang melakukan hal berikut:

  • Membaca tabel Unity Catalog dan menampilkannya di antarmuka Streamlit.
  • Mengedit data dan menulisnya kembali ke tabel.

Langkah 1: Mengonfigurasi hak istimewa

Contoh-contoh ini mengasumsikan bahwa aplikasi Anda menggunakan otorisasi aplikasi. Perwakilan layanan aplikasi Anda harus memiliki:

  • SELECT hak akses pada tabel Unity Catalog
  • MODIFY hak akses pada tabel Unity Catalog
  • CAN USE hak istimewa pada gudang data SQL

Untuk informasi selengkapnya, lihat Referensi hak istimewa Katalog Unity dan ACL gudang SQL.

Langkah 2: Menginstal dependensi

Buat requirements.txt file dan sertakan paket berikut:

databricks-sdk
databricks-sql-connector
streamlit
pandas

Langkah 3: Mengonfigurasi eksekusi aplikasi

app.yaml Buat file untuk menentukan bagaimana aplikasi Anda dimulai di Azure Databricks Apps.

command: ['streamlit', 'run', 'app.py']

Langkah 4: Membaca tabel Katalog Unity

Contoh kode ini menunjukkan cara membaca data dari tabel Unity Catalog dan menampilkannya menggunakan Streamlit. Buat app.py file yang memenuhi tujuan berikut:

  • Menggunakan autentikasi prinsipal layanan aplikasi.
  • Meminta pengguna untuk jalur HTTP gudang SQL dan nama tabel Katalog Unity.
  • SELECT * Menjalankan kueri pada tabel yang ditentukan.
  • Menampilkan hasil dalam Streamlit st.dataframe.

app.py

import pandas as pd
import streamlit as st
from databricks import sql
from databricks.sdk.core import Config
import os

cfg = Config()

# Use app service principal authentication
def get_connection(http_path):
    server_hostname = cfg.host
    if server_hostname.startswith('https://'):
        server_hostname = server_hostname.replace('https://', '')
    elif server_hostname.startswith('http://'):
        server_hostname = server_hostname.replace('http://', '')
    return sql.connect(
        server_hostname=server_hostname,
        http_path=http_path,
        credentials_provider=lambda: cfg.authenticate,
        _use_arrow_native_complex_types=False,
    )

# Read data from a Unity Catalog table and return it as a pandas DataFrame
def read_table(table_name: str, conn) -> pd.DataFrame:
    with conn.cursor() as cursor:
        cursor.execute(f"SELECT * FROM {table_name}")
        return cursor.fetchall_arrow().to_pandas()

# Use Streamlit input fields to accept user input
http_path_input = st.text_input(
    "Enter your Databricks HTTP Path:", placeholder="/sql/1.0/warehouses/xxxxxx"
)
table_name = st.text_input(
    "Specify a Unity Catalog table name:", placeholder="catalog.schema.table"
)

# Display the result in a Streamlit DataFrame
if http_path_input and table_name:
    conn = get_connection(http_path_input)
    df = read_table(table_name, conn)
    st.dataframe(df)
else:
    st.warning("Provide both the warehouse path and a table name to load data.")

Langkah 5: Edit tabel Katalog Unity

Contoh kode ini memungkinkan pengguna membaca, mengedit, dan menulis perubahan pada tabel Unity Catalog menggunakan fitur pengeditan data Streamlit. Tambahkan fungsionalitas berikut ke app.py file:

  • Gunakan INSERT OVERWRITE untuk menulis data yang diperbarui kembali ke tabel.

app.py

import pandas as pd
import streamlit as st
from databricks import sql
from databricks.sdk.core import Config
import math

cfg = Config()

# Use app service principal authentication
def get_connection(http_path):
    server_hostname = cfg.host
    if server_hostname.startswith('https://'):
        server_hostname = server_hostname.replace('https://', '')
    elif server_hostname.startswith('http://'):
        server_hostname = server_hostname.replace('http://', '')
    return sql.connect(
        server_hostname=server_hostname,
        http_path=http_path,
        credentials_provider=lambda: cfg.authenticate,
        _use_arrow_native_complex_types=False,
    )

# Read data from a Unity Catalog table and return it as a pandas DataFrame
def read_table(table_name: str, conn) -> pd.DataFrame:
    with conn.cursor() as cursor:
        cursor.execute(f"SELECT * FROM {table_name}")
        return cursor.fetchall_arrow().to_pandas()

# Format values for SQL, handling NaN/None as NULL
def format_value(val):
    if val is None or (isinstance(val, float) and math.isnan(val)):
        return 'NULL'
    else:
        return repr(val)

# Use `INSERT OVERWRITE` to update existing rows and insert new ones
def insert_overwrite_table(table_name: str, df: pd.DataFrame, conn):
    progress = st.empty()
    with conn.cursor() as cursor:
        rows = list(df.itertuples(index=False))
        values = ",".join([f"({','.join(map(format_value, row))})" for row in rows])
        with progress:
            st.info("Calling Databricks SQL...")
        cursor.execute(f"INSERT OVERWRITE {table_name} VALUES {values}")
    progress.empty()
    st.success("Changes saved")

# Use Streamlit input fields to accept user input
http_path_input = st.text_input(
    "Enter your Databricks HTTP Path:", placeholder="/sql/1.0/warehouses/xxxxxx"
)
table_name = st.text_input(
    "Specify a Unity Catalog table name:", placeholder="catalog.schema.table"
)

# Display the result in a Streamlit DataFrame
if http_path_input and table_name:
    conn = get_connection(http_path_input)
    if conn:
        st.success("✅ Connected successfully!")
        original_df = read_table(table_name, conn)
        edited_df = st.data_editor(original_df, num_rows="dynamic", hide_index=True)
        df_diff = pd.concat([original_df, edited_df]).drop_duplicates(keep=False)
        if not df_diff.empty:
            st.warning(f"⚠️ You have {len(df_diff) // 2} unsaved changes")
            if st.button("Save changes"):
                insert_overwrite_table(table_name, edited_df, conn)
                st.rerun()
else:
    st.warning("Provide both the warehouse path and a table name to load data.")

Langkah selanjutnya