dbplyr standardises all the ways of referring to a table (i.e. a single
string, a string wrapped in I()
, a DBI::Id()
and the results of
in_schema()
and in_catalog()
) into a table "path" of the form
table
, schema.table
, or catalog.schema.path
. A table path is
always suitable for inlining into a query, so user input is quoted unless
it is wrapped in I()
.
This is primarily for internal usage, but you may need to work with it if you're implementing a backend, and you need to compute with the table path, not just pass it on unchanged to some other dbplyr function.
is_table_path()
returnsTRUE
if the object is atable_path
.as_table_path()
coerces known table identifiers to atable_path
.check_table_path()
throws an error if the object is not atable_path
.table_path_name()
returns the last component of the table path (i.e. the name of the table).table_path_components()
returns a list containing the components of each table path.
A table_path
object can technically be a vector of table paths, but
you will never see this in table paths constructed from user inputs.