col_spec: Get column types for use as 'col_types' in...

In jedwards24/jemisc: Some miscellaneous functions for personal use

Get column types for use as col_types in readr::read_csv()

Description

Returns the compact string representation for the col_types argument in readr::read_csv() and related functions of the supplied data. The intended use is to record the types after an initial load (with any further processing), so it can be used for future reads of the same data.

Usage

col_spec(data)

Arguments

data	A data frame.

Details

The column types will be one of:

• c = character

• i = integer

• d = double

• l = logical

• f = factor

• D = date

• T = date time

• t = time

These types are read directly from data - there is no guessing or coercion. The first four are identified using typeof(). The remainder overwrite the type if one of the following is true for the column: is.factor(), lubridate::is.Date(), lubridate::is.POSIXt(), or "hms" %in% class(.).

This function is unnecessary because you can use the spec() directly as col_types or extract the string using as.character(spec(x)). Also this function will not capture the col_number() specification which reads values such as "£10.50" or "1,234" correctly.

jedwards24/jemisc documentation built on Oct. 23, 2022, 10:17 p.m.