lglass has several database backends for persisting and accessing the registry objects. The API is consistent through the several database backends. At the moment the following database backends are supported:
The SQLite3 database is really simple designed and not very performant. Furthermore, its database schema is completely broken but it uses less memory than our Filesystem database.
whois+lglass.database.sqlite3+sqlite3:{PATH}
The Filesystem database is our original database implementation and uses directories to represent the object type while using the filename to represent the primary key. Using this schema it is simple to use by humans but also performant.
whois+lglass.database.file+file:{PATH}
lglass also provides many pseudo databases which wraps around existing real databases and provide additional information like inverse related objects and IP range lookups. At the moment, there are two important pseudo databases:
The CIDR Database resolves requests for IP addresses and ASNs. It transforms the request to a request for all possible supernets (cidr requests) and ranges (range requests), and therefore returns also supernets and ranges.
whois+lglass.database.cidr+cidr:?{params}
Possible parameters are:
| Parameter | Description |
|---|---|
| range-slice | Python slice expression to slice the range request results |
| cidr-slice | Python slice expression to slice the cidr request results |
The Schema Database provides schema validation and inverse relation resolution.
whois+lglass.database.schema+schema:?{params}
Possible parameters are:
| Parameter | Description |
|---|---|
| types-include | Comma-separated list of types for inverse relation resolution |