def get_df()

in mozetl/taar/taar_ensemble.py [0:0]

• 26 lines of code
• 1 McCabe index (conditional complexity)

def get_df(spark, date_from):
    df = (
        spark.sql("SELECT * FROM clients_daily")
        .where("active_addons IS NOT null")
        .where("size(active_addons) > 2")
        .where("size(active_addons) < 100")
        .where("channel = 'release'")
        .where("app_name = 'Firefox'")
        .where("submission_date_s3 >= {}".format(date_from))
        .selectExpr(
            "client_id as client_id",
            "active_addons as active_addons",
            "city as geo_city",
            "subsession_hours_sum as subsession_length",
            "locale as locale",
            "os as os",
            "row_number() OVER (PARTITION BY client_id ORDER BY submission_date_s3 desc) as rn",
            "places_bookmarks_count_mean AS bookmark_count",
            "scalar_parent_browser_engagement_tab_open_event_count_sum AS tab_open_count",
            "scalar_parent_browser_engagement_total_uri_count_sum AS total_uri",
            "scalar_parent_browser_engagement_unique_domains_count_max AS unique_tlds",
        )
        .where("rn = 1")
        .drop("rn")
    )
    return df