Overview

Most of the configurations are set through Flask Config Class.

BADGES

In order to add a badge to a resource you should first add the combination of badge name and category to the in WHITELIST_BADGES Config Class.

Example:

WHITELIST_BADGES: List[Badge] = [Badge(badge_name='beta',
                                 category='table_status')]

Once this is done users will be able to add badge the badges in the whitelist by running:

curl -X PUT https://{amundsen metadata url}/table/"{table key}"/badge/{badge_name}?category={category}

USER_DETAIL_METHOD OPTIONAL

This is a method that can be used to get the user details from any third-party or custom system. This custom function takes user_id as a parameter, and returns a dictionary consisting user details’ fields defined in UserSchema.

Example:

def get_user_details(user_id):
    user_info = {
        'email': 'test@email.com',
        'user_id': user_id,
        'first_name': 'Firstname',
        'last_name': 'Lastname',
        'full_name': 'Firstname Lastname',
    }
    return user_info

USER_DETAIL_METHOD = get_user_details

STATISTICS_FORMAT_SPEC OPTIONAL

This is a variable enabling possibility to reformat statistics displayed in UI.

The key is name of statistic and a value is a dictionary with optional keys: * new_name - how to rename statistic (if absent proxy should default to old name) * format - how to format numerical statistics (if absent, proxy should default to original format) * drop - should given statistic not be displayed in UI (if absent, proxy should keep it)

Example (if you’re using deeque library), you might want to:

STATISTICS_FORMAT_SPEC = {
        'stdDev': dict(new_name='standard deviation', format='{:,.2f}'),
        'mean': dict(format='{:,.2f}'),
        'maximum': dict(format='{:,.2f}'),
        'minimum': dict(format='{:,.2f}'),
        'completeness': dict(format='{:.2%}'),
        'approximateNumDistinctValues': dict(new_name='distinct values', format='{:,.0f}', ),
        'sum': dict(drop=True)
}