Find not-fitting element (culprit) in vector

Find the element(s) in a vector, whose data type potentially deviates from the majority data type or a specified one.

Usage

find_culprit(
  vec,
  should_be = c("majority", "numeric", "double", "integer", "character", "logical"),
  return_index = FALSE,
  na.rm = TRUE,
  verbose = TRUE,
  ...
)

Arguments

vec

Data vector to find the not-fitting element in.

should_be

Expected data type of vec. Numeric is indifferent if double or integer is provided. If set to majority, expect the data type of most of the elements.

return_index

Return index of culprit elements rather than their value.

na.rm

Do not treat NA values as culprits, i.e. discard them from the culprit return.

verbose

If guessed data type should be printed if should_be is majority.

...

Arguments passed on to readr::guess_parser

na

Character vector of strings to interpret as missing values. Set this option to character() to indicate no missing values.

locale

The locale controls defaults that vary from place to place. The default locale is US-centric (like R), but you can use locale() to create your own locale that controls things like the default time zone, encoding, decimal mark, big mark, and day/month names.

Value

A named vector, with either the element values (default) or their index (return_index) and the guessed data types as names.

Details

This function can be used to find the culprit element, which prevents clean data type conversion. Utilizes readr::guess_parser() to guess type of every vector element. User can specify which data type the vector should_be, or simply rely on the majority guess. Ties in the majority guess are resolved lexicographically by data type name, i.e. character > double > integer > logical. If should_be is numeric, the data type is interpreted as double for lexicographic sorting.

Examples

long_data <- c(1, 2.2, "4,", 5)

# standard type conversion would convert "<4" to NA
if (FALSE) { # \dontrun{
as.numeric(long_data)
} # }

# better for maximizing information content: find culprit element and treat it appropriately
find_culprit(long_data, should_be = "numeric")
#> number 
#>   "4," 
as.numeric(gsub(",", "", long_data))
#> [1] 1.0 2.2 4.0 5.0