内容

bigquery, MySQLからpandas dataframeを作る関数と、dataframeからBigqueryへ保存とGCSへのCSV保存をやりたく、ついにまとめたので、ちと記事作成。

コード

sample.py

import pandas as pd
import os
import pymysql
import dsclient

# init client with project id and creadential
client = dsclient.Client(
    "YOUR_GCP_PROJECT_NAME",
    "/hogehoge/hogehoge/credentials.json"
)


def bq2df(query):
    # read data with big query
    return client.query(query)


def mysql2df(query):
    # read data from MySQL with Google Cloud SQL
    connection = pymysql.connect(
        host='XXXXXXX',
        user='YOUR_USERNAME',
        password='YOUR_PASSWORD',
        db='YOUR_DBNAME',
        charset='utf8',
        cursorclass=pymysql.cursors.DictCursor
    )
    return pd.read_sql(sql, connection)


def df2bq(df, dataset, table):
    # store data with big query
    tb_name = dataset + "." + table
    client.load(df, tb_name)


def df2gcs(df, folder, file_name, bucket="gs://hogehoge/"):
    # store data to Google Cloud Storage as csv
    buket_name = os.path.join(bucket, folder, file_name)
    client.write_csv(df, buket_name)


if __name__ == "__main__":
    df = pd.DataFrame([i for i in range(100)], columns=["test"])
    folder = "write_test"
    file_name = "test1.csv"
    df2gcs(df, folder, file_name)

英語のコメントアウトは気にしないでちょ
各hogehoge等を変更してガムバッテクレ
これ超便利。

参考サイト