protect_integer_cols: Protect column names that are bare integers from being...

In JGCRI/gcamdata: Build the GCAM Input Datasets

Protect column names that are bare integers from being interpreted as integers

Description

protect_integer_cols will prepend an 'X' to column names that are bare integers to protect them from being misinterpreted during sensitive operations (see below). unprotect_integer_cols will reverse the effect, reverting the column names to their original form.

Usage

protect_integer_cols(d)

unprotect_integer_cols(d)

Arguments

d	The data to have integer column names protected or unprotected

Details

Some of our data frames have column names that are years, such as "2005". In some cases a column name of this form can be misinterpreted as a column index, the (likely nonexistent) 2005th column in this example. The *_if functions in dplyr are examples of such circumstances. Protecting the column names with a leading character allows these functions to perform normally.

Much of the data system is expecting years in column names to be bare integers; therefore, you should unprotect the column names as soon as the sensitive operations are complete.

Value

Tibble with the integer column names protected

Examples

library(magrittr)
df <- dplyr::tibble(iso=c('bad','dum'), `2005`=c(123.45, NA), `2050`=c(867, 5309))
protect_integer_cols(df) %>%
    dplyr::select_if(function(col) {!any(is.na(col))}) %>%
    unprotect_integer_cols

JGCRI/gcamdata documentation built on March 21, 2023, 2:19 a.m.