Compare validation check state between two runs — ks_compare_check

Compares current results against a previous snapshot and adds a status column indicating whether each row is new, changed, or unchanged relative to the previous run.

Usage

ks_compare_check_state(current, previous = NULL, key_cols, compare_cols)

Arguments

current: A data frame from the most recent ks_check_rules run.
previous: A data frame from a previous run, or NULL (default). If NULL or empty, all rows in current receive status = "new".
key_cols: Character vector of column names used to match rows across runs. Must be provided explicitly.
compare_cols: Character vector of column names whose values are used to detect changes. Must be provided explicitly.

Value

current with an additional character column status:

"new": Row key not found in previous.
"changed": Row key found but comparison signature differs.
"": Row is unchanged.

Examples

if (FALSE) { # \dontrun{
result_new <- ks_check_rules(data_new, rules)
result_old <- readRDS("previous_check.rds")
ks_compare_check_state(result_new, result_old)
} # }