High-Performance Open-Source Archive
describe() to describe a datasetcheck_data() now includes schema checks in the output
by default (check_type as first result column), including
explicit rows for column existence and declared type checksstop_on_schema_fail to check_data() to
optionally stop when schema checks failfilter_fails() to ignore schema/reference rows
and only process row rules from check_data() resultsdetect_backend()
fallback to dplyr when input is a data.frame
and data.table is unavailabledata_column(), rule_meta()) and reference
checks (reference_rule())ruleset(), check_data(),
read_rules(), and write_rules() for v1
schema-aware workflows; keep rule() as row-level API (no
col_rule())sample_data dataset (mixed types, NAs,
datetime) for examples and testsreference_rule() and extend examples in
ruleset(), check_data(),
reference_rule(), and data_column() to show
combined schema + relational workflows>= 1.5.1.9002 in all
DuckDB-backed tests via
skip_if_not_installed("duckdb", "1.5.1.9002")fail_on_X to stop_on_Xstop_on_fail to [check_data()], so
that the examples using read_custom() make sense (eg in the
Readme); thanks FedericoComoglio for
pointing it out!detect_backend()] to allow user the check which
backend is usedfilter_fails() allows the first argument to be a
ruleset and not only a result of
check_data()NEWS.md file to track changes to the
package.
Need mirroring services?
Contact our team at info@vpspulse.com.
Mirror powered by VPSpulse
Infrastructure sponsored by VPSPulse & Secure Payments by ArionPay.