Writing an agent or informant to disk with x_write_disk() can be useful
for keeping data validation intel or table information close at hand for
later retrieval (with x_read_disk()). By default, any data table that the
agent or informant may have held before being committed to disk will be
expunged. This behavior can be changed by setting keep_tbl to TRUE but
this only works in the case where the table is not of the tbl_dbi or the
tbl_spark class.
x_write_disk(
x,
filename,
path = NULL,
keep_tbl = FALSE,
keep_extracts = FALSE,
quiet = FALSE
)An agent object of class ptblank_agent, or, an informant of
class ptblank_informant.
The filename to create on disk for the agent or
informant.
An optional path to which the file should be saved (this is
automatically combined with filename).
An option to keep a data table that is associated with the
agent or informant (which is the case when the agent, for example, is
created using create_agent(tbl = <data table, ...)). The default is
FALSE where the data table is removed before writing to disk. For
database tables of the class tbl_dbi and for Spark DataFrames
(tbl_spark) the table is always removed (even if keep_tbl is set to
TRUE).
An option to keep any collected extract data for failing
rows. By default, this is FALSE.
Should the function not inform when the file is written? By
default this is FALSE.
Invisibly returns TRUE if the file has been written.
9-1
It is recommended to set up a table-prep formula so that the agent and
informant can access refreshed data after being read from disk through
x_read_disk(). This can be done initially with the read_fn argument of
create_agent()/create_informant() or, later, with set_read_fn().
Alternatively, we can reintroduce the agent or informant to a data table
with the set_tbl() function.
Other Object Ops:
activate_steps(),
deactivate_steps(),
remove_read_fn(),
remove_steps(),
remove_tbl(),
set_read_fn(),
set_tbl(),
x_read_disk()