data.table (version 1.17.0)

measure: Specify measure.vars via regex or separator

Description

These functions compute an integer vector or list for use as the measure.vars argument to melt. Each measured variable name is converted into several groups that occupy different columns in the output melted data. measure allows specifying group names/conversions in R code (each group and conversion specified as an argument) whereas measurev allows specifying group names/conversions using data values (each group and conversion specified as a list element). See vignette("datatable-reshape") for more info.

Usage

measure(..., sep, pattern, cols, multiple.keyword="value.name")
measurev(fun.list, sep, pattern, cols, multiple.keyword="value.name")

Arguments

...

One or more (1) symbols (without argument name; symbol is used for group name) or (2) functions to convert the groups (with argument name that is used for group name). Must have same number of arguments as groups that are specified by either sep or pattern arguments.

fun.list

Named list which must have the same number of elements as groups that are specified by either sep or pattern arguments. Each name used for a group name, and each value must be either a function (to convert the group from a character vector to an atomic vector of the same size) or NULL (no conversion).

sep

Separator to split each element of cols into groups. Columns that result in the maximum number of groups are considered measure variables.

pattern

Perl-compatible regex with capture groups to match to cols. Columns that match the regex are considered measure variables.

cols

A character vector of column names.

multiple.keyword

A string, if used as a group name, then measure returns a list and melt returns multiple value columns (with names defined by the unique values in that group). Otherwise if the string not used as a group name, then measure returns a vector and melt returns a single value column.

See Also

melt, https://github.com/Rdatatable/data.table/wiki/Getting-started

Examples

(two.iris = data.table(datasets::iris)[c(1,150)])
# melt into a single value column.
melt(two.iris, measure.vars = measure(part, dim, sep="."))
# do the same, programmatically with measurev
my.list = list(part=NULL, dim=NULL)
melt(two.iris, measure.vars=measurev(my.list, sep="."))
# melt into two value columns, one for each part.
melt(two.iris, measure.vars = measure(value.name, dim, sep="."))
# melt into two value columns, one for each dim.
melt(two.iris, measure.vars = measure(part, value.name, sep="."))
# melt using sep, converting child number to integer.
(two.families = data.table(sex_child1="M", sex_child2="F", age_child1=10, age_child2=20))
print(melt(two.families, measure.vars = measure(
  value.name, child=as.integer,
  sep="_child"
)), class=TRUE)
# same melt using pattern.
print(melt(two.families, measure.vars = measure(
  value.name, child=as.integer,
  pattern="(.*)_child(.)"
)), class=TRUE)
# same melt with pattern and measurev function list.
print(melt(two.families, measure.vars = measurev(
  list(value.name=NULL, child=as.integer),
  pattern="(.*)_child(.)"
)), class=TRUE)
# inspired by data(who, package="tidyr")
(who <- data.table(id=1, new_sp_m5564=2, newrel_f65=3))
# melt to three variable columns, all character.
melt(who, measure.vars = measure(diagnosis, gender, ages, pattern="new_?(.*)_(.)(.*)"))
# melt to five variable columns, two numeric (with custom conversion).
print(melt(who, measure.vars = measure(
  diagnosis, gender, ages,
  ymin=as.numeric,
  ymax=function(y)ifelse(y=="", Inf, as.numeric(y)),
  pattern="new_?(.*)_(.)(([0-9]{2})([0-9]{0,2}))"
)), class=TRUE)