import mario

profile = mario.get_database_profile("gloria")
profile.keys()

Result · An offline metadata dictionary. No database files are downloaded or parsed, and no settings are changed.

Family and model profiles#

Profile contents#

Section

Contains

Next action

ghg

Native accounts, factors and profile selection

Calculate GHG

compute

Parser preferences for numerical calculations

Configure calculations

concordances

Variants, target editions, sources and hierarchy levels

db.get_concordances() or db.get_aggregation()

Family support is not model availability. A family can have concordances while the imported release has none. On a parsed model, inspect:

profile = db.database_profile
profile["concordances"]["available"]
profile["concordances"]["unavailable_reason"]
GHG profile selection and existing CO2eq totals

ghg["selected_profile"] follows source/name metadata. It does not check that all configured gas accounts exist, or record an explicit override made in a previous calc_ghg(profile=...) call.

API defaults are AR6/GWP100. Already characterized CO2eq totals keep their original factors; selecting another report does not reconstruct their gases. BONSAI’s default AR6 recipe also differs from the explorer’s AR5 comparison.

Preferred solver versus active session settings

compute["parser_presets"] describes a parser’s preferred settings. The current configuration is mario.download_settings()["compute"].

BONSAI, GLORIA and EMERGING prefer solve/direct with optional UMFPACK and a SciPy fallback. An empty preset leaves the session unchanged. Apply manual overrides after parsing.

Versions, provenance and copies

Definitions are packaged with MARIO. as_of describes the classification metadata snapshot, not the economic reference year or a fresh online query. Available editions can still have incomplete coverage at item level.

Returned dictionaries are independent copies. Editing them does not change settings, register GHG profiles or update the packaged concordances. Availability is independent across sections: a source may support GHG characterization but have no bundled classification mapping, or vice versa.