Get the cell-status (missing-cell) table

The one accessor for "which of the cells this table does not carry are missing rather than zero, and why", whichever form the table is in: an ivt object from read_ivt(missing = TRUE), a Parquet path, or the Arrow connection get_statcan_ivt() returns.

Usage

ivt_missing(x)

Arguments

x

An ivt object, a .parquet path, or an Arrow dataset / dplyr query from get_statcan_ivt().

Value

For an ivt, the missing tibble (member-id coordinates); for a Parquet source, a lazy arrow::open_dataset() connection to the _missing.parquet sidecar with the same labelled coordinates as the data. NULL when the table carries none – i.e. it was decoded without missing = TRUE.

Details

Every row is a cell the file marks as not a zero, with the reason the file states (symbol / status), or NA where the page carries only the bare absent mask. On the published table (read_ivt()'s default complete = TRUE) these are simply its is.na(value) rows, so this is a convenience view; under complete = FALSE it is the only thing separating a suppressed cell from a published zero, since neither is stored.

See also

read_ivt(), ivt_tidy_missing(), ivt_write_parquet()

Examples

path <- system.file("extdata", "98100044.ivt", package = "canivt")
ivt_missing(read_ivt(path, missing = TRUE))
#> # A tibble: 0 × 5
#> # ℹ 5 variables: geo <int>, type <int>, collective <int>, symbol <chr>,
#> #   status <chr>