Read TDA's output files

tda_read_table reads one table, taking column names from the comment header where they can be matched to the columns. tda_read_blocks splits a file holding several tables, which is what a life table writes when there are groups. tda_file and tda_blocks do the same relative to a run's directory, and report what TDA said if the file is missing.

Usage

tda_read_table(file, col_names = NULL)

tda_read_blocks(file, col_names = NULL)

tda_blocks(x, file, ...)

tda_file(x, file, ...)

Arguments

file

a file name.

col_names

column names to apply; a list for tda_read_blocks, one entry per block.

x

a run or fitted model.

...

passed on.

Value

A data frame, or a list of them.

See also

Other TDA infrastructure: tda_ccnt(), tda_help(), tda_output(), tda_run(), tda_write_data()

Examples

set.seed(39)
d <- data.frame(income = round(rlnorm(30, 8, 0.5)))
fit <- tda_ineq(d)
# tda_file() reads a named output file relative to the run's directory
tda_file(fit, "out.txt", col_names = c("index", "cases", "min", "max",
                                       "mean", "sd", "vcoeff", "gini"))
#>   index cases  min  max     mean      sd    vcoeff      gini
#> 1     1    30 1256 5459 2929.233 1099.43 0.3753304 0.2099765

# tda_blocks(): a life table with groups writes several tables into one
# file -- counts, then a survivor/density/rate table, once per group
set.seed(1)
d2 <- data.frame(t = c(4, 3, 1, 5, 8, 2), s = c(1, 1, 0, 1, 1, 1),
                 g = c(1, 1, 1, 2, 2, 2))
lt <- tda_ltb(Surv(t, s) ~ as.factor(g), d2, tp = seq(0, 10, 2))
length(tda_blocks(lt, "out.ltb"))  # 4: two groups, two tables each
#> [1] 4

# tda_read_blocks(): the same thing from a raw file path, when there is
# no fit object to resolve one from
path <- file.path(lt$run$dir, "out.ltb")
length(tda_read_blocks(path))
#> [1] 4