Reference catalogs¶
Reference data lives in YAML under cloudg/inventory/catalogs/, loaded by
load_catalog(name) (cached per process):
| File | Keys | Used by |
|---|---|---|
aws_cloudcontrol.yaml |
covered_types (mapping of group → CloudFormation types a deep collector already covers), global_prefixes (types listed once per account), asset_types (CloudFormation type → AssetType), sensitive_fields (type → top-level properties to drop) |
Cloud Control sweep |
aws_arn_types.yaml |
arn_types (service:resource-type or service → AssetType) |
asset_type_from_arn (Tagging API and Cloud Control hits) |
azure_arm_types.yaml |
arm_types (lower-case ARM type → AssetType) |
asset_type_from_arm |
gcp_asset_types.yaml |
base (standard collector), deep (inventory, overrides base) |
GCP collectors |
Overlays: put a file with the same name in $CLOUDG_CATALOG_DIR; mappings merge key by key
(recursively), lists are extended without duplicates. AssetType names are validated by
asset_type_map() at import time, so a typo fails loudly with the catalog, key and value.
Because catalogs are read when the modules are imported, set $CLOUDG_CATALOG_DIR before
importing cloudg.
When you add a deep collector for a resource type that Cloud Control also lists, add the type
to covered_types so the sweep skips it.