fastgcp 0.1.1

LEGAL DISCLAIMER

Neither this package nor the author, tomathon, are affiliated with or endorsed by Google. The inclusion of Google trademark(s), if any, upon this webpage is solely to identify Google contributors, goods, or services, and not for commercial purposes.

---

🚀 FastGCP is the faster, more intuitive way of working with Google Cloud Platform services and frameworks.

The fastgcp package is built on top of canonical Google Python packages(s) without any alteration to Google's base code.

Full documentation can be found here (WIP)

---

Installation

fastgcp is installed using pip with the command:

$ pip install fastgcp

---

The BigQuery class

One of the main FastGCP classes is BigQuery. It allows users to work with BQ database tables and objects in a more Pythonic way.

Quickstart

Import BigQuery:

from fastgcp import BigQuery

Initialize BigQuery using your GCP Project name:

bq = BigQuery(project_id = "gcp-project-name")

Alternatively, to initialize using a specific service account key, pass the full key path to your key.json:

bq = BigQuery(
    project_id = "gcp-project-name",
    service_acct_key_path = "/full/path/to/your/service/account/key.json"
)

Common BigQuery Methods:

bq.read_bigquery

Read an existing BigQuery table into a DataFrame

read_bigquery(bq_dataset_dot_table = None, date_cols = [], preview_top = None, to_verbose = True)

• bq_dataset_dot_table : the "dataset-name.table-name" path of the existing BigQuery table
• date_cols : [optional] column(s) passed inside a list that should be parsed as dates
• preview_top : [optional] only read in the top N rows
• to_verbose : should info be printed? defaults to True
• use_polars : should a polars DataFrame be returned instead of a pandas DataFrame? Defaults to True

EX:

my_table = bq.read_bigquery("my_bq_dataset.my_bq_table")

my_table = bq.read_bigquery("my_bq_dataset.my_bq_table", date_cols = ['date'])

-----------

bq.read_custom_query

Read an existing BigQuery table into a DataFrame, using a custom SQL expression

read_custom_query(custom_query, to_verbose = True)

• custom_query : the custom BigQuery SQL query that will produce a table to be read into a DataFrame
• to_verbose : should info be printed? defaults to True
• use_polars : should a polars DataFrame be returned instead of a pandas DataFrame? Defaults to True

EX:

my_custom_table = bq.read_custom_query("""
    SELECT
        date,
        sales,
        products
    FROM
        my_bq_project_id.my_bq_dataset.my_bq_table
    WHERE
        sales_month = 'June'
""")

-----------

bq.write_bigquery

Write a DataFrame to BigQuery

write_bigquery(df, bq_dataset_dot_table = None, use_schema = None, append_to_existing = False, to_verbose = True)

• df : the DataFrame to be written to a BigQuery table
• bq_dataset_dot_table : the "dataset-name.table-name" path of the existing BigQuery table
• use_schema : [optional] a custom schema for the BigQuery table. NOTE: see bq.guess_schema below
• append_to_existing : should the DataFrame be appended to an existing BigQuery table? defaults to False (create new / overwrite)
• to_verbose : should info be printed? defaults to True

EX:

bq.write_bigquery(my_df, "my_bq_dataset.my_data")

bq.write_bigquery(my_df, "my_bq_dataset.my_data", append_to_existing = True)

-----------

bq.send_query

Send a custom SQL query to BigQuery. Process is carried out within BigQuery. Nothing is returned

send_query(que, to_verbose = True)

• que : the custom SQL query to be sent and carried out within BigQuery
• to_verbose : should info be printed? defaults to True

EX:

bq.send_query("""
    CREATE TABLE my_bq_project_id.my_bq_dataset.my_new_bq_table AS 
    (
        SELECT
            date,
            sales,
            products
        FROM
            my_bq_project_id.my_bq_dataset.my_bq_table
        WHERE
            sales_month = 'June'
    )
""")

-----------