recode_conditional: Conditionally recode values within a data frame column based...
In epicentre-msf/llutils: Linelist Utilities

Description Usage Arguments Value Examples

Conditionally recode values within a data frame column based on one or more dictionaries containing columns to match against, and corresponding replacement values to insert where a match is made

1	recode_conditional(df, dict, col_recode, flag_recoded = FALSE)

`df`	A `data.frame`-like object
`dict`	The dictionary. A data frame, or list of data frames, each containing a column of replacement values, and one or more columns to match against `df` (matching via `dplyr::left_join`).
`col_recode`	Name (character) of the column to be recoded. This column must be present both within `df` and `dict`.
`flag_recoded`	Logical indicating whether to add a column to the output indicating whether the given row has been matched by a dictionary entry and therefore recoded. Defaults to FALSE. If TRUE, the new column will be called `<col_recode>_is_recoded`.

A data.frame corresponding to df, but with values in column col_recode recoded based on the given dictionary.

If flag_recoded is TRUE, the output will contain an additional column <col_recode>_is_recoded, indicating whether the given row has been matched by a dictionary entry and therefore recoded.

# dictionaries
dict1 <- data.frame(
  x1 = c("a", "b", "c", "d"),
  y = c("alpha", "bravo", "charlie", "delta"),
  stringsAsFactors = FALSE
)

dict2 <- data.frame(
  x1 = c("0", "1", "2", "3"),
  y = c("zero", "one", "two", "three"),
  stringsAsFactors = FALSE
)

# data frame with column to be recoded ('y')
dat <- data.frame(
  x1 = c("a", "c", "c", "d", "b", "0"),
  y = c("???", "???", "???", "???", "???", "???"),
  stringsAsFactors = FALSE
)

# recode column "y" based on dictionary 'dict1'
recode_conditional(dat, dict = dict1, col_recode = "y")

# recode column "y" based on dictionaries 'dict1' and 'dict2'
recode_conditional(dat, dict = list(dict1, dict2), col_recode = "y")