retrieve GA4 data using API with UTF-8 metrics name. Save to JSON file using Pandas DataFrame

ga4_reports.py

import os
import pandas as pd


from google.analytics.data_v1beta import BetaAnalyticsDataClient
from google.analytics.data_v1beta.types import (
    DateRange,
    Dimension,
    Metric,
    MetricType,
    RunReportRequest,
    OrderBy,
    FilterExpression,
    Filter,
    FilterExpressionList,
)


property_list = [
    "YOUR-PROPERTY-ID",
    "YOUR-ANOTHER-PROPERTY-ID",
    "AND-THE-NEXT-ONE",
]
properties = {
    "YOUR-PROPERTY-ID": "domain-url.com",
    "YOUR-ANOTHER-PROPERTY-ID": "second-domain.com",
    "AND-THE-NEXT-ONE": "anither-domain.com",
}
report_type_list = [
    'basic',
    'conversions',
]

start_date = '2024-02-01'
end_date = '2024-02-18'

dimensions_dict = {
    'basic': ['date', 'sessionSourceMedium', 'sessionCampaignName'],
    'conversions': ['date', 'sessionSourceMedium', 'sessionCampaignName', 'eventName'],
}
metrics_dict = {
    'basic': [
        'totalUsers',
        'newUsers',
        'sessions',
        'bounceRate',
        'averageSessionDuration',
        'screenPageViewsPerSession',
    ],
    'conversions': [
        'conversions',
    ],
}


def create_dimensions(report_type):
    dimensions = []
    for dimension in dimensions_dict[report_type]:
        dimensions.append(Dimension(name=dimension))
    return dimensions


def create_metrics(report_type):
    metrics = []
    for metric in metrics_dict[report_type]:
        metrics.append(Metric(name=metric))
    return metrics


def run_sample():
    for property_key in property_list:
        property_id = property_key
        for report_type in report_type_list:
            run_report(property_id, report_type)


def run_report(property_id, report_type):
    client = BetaAnalyticsDataClient()

    request = RunReportRequest(
        property=f"properties/{property_id}",
        dimensions=create_dimensions(report_type),
        metrics=create_metrics(report_type),
        date_ranges=[DateRange(start_date=start_date, end_date=end_date)],
        order_bys=[
            OrderBy(dimension=OrderBy.DimensionOrderBy(dimension_name="date"))
        ],
    )

    response = client.run_report(request)

    print_run_report_response(response)

    property_value = properties.get(property_id)
    print_run_report_response_rows_table(response, property_value, report_type)


def print_run_report_response(response):
    print(f"{response.row_count} rows received")
    for dimensionHeader in response.dimension_headers:
        print(f"Dimension header name: {dimensionHeader.name}")
    for metricHeader in response.metric_headers:
        metric_type = MetricType(metricHeader.type_).name
        print(f"Metric header name: {metricHeader.name} ({metric_type})")


def print_run_report_response_rows_table(response, property_value, report_type):
    data = []
    for row in response.rows:
        dimension_values = [value.value for value in row.dimension_values]
        metric_values = [value.value for value in row.metric_values]
        data.append(dimension_values + metric_values)

    columns = [dimension.name for dimension in response.dimension_headers] + [metric.name for metric in response.metric_headers]
    df = pd.DataFrame(data, columns=columns)
    # print(df)
    response_to_json(df, property_value, report_type)
    return df


def response_to_json(data, property_value, report_type):
    response_files_path = f"ga4_reports/{property_value}/original/{start_date}___{end_date}"
    report_file = f'{response_files_path}/{report_type}.json'
    os.makedirs(f'{response_files_path}', exist_ok=True)
    return data.to_json(report_file, orient='table')


if __name__ == "__main__":
    run_sample()