replace_values: replace symbols or other character flags
In PKPDmisc: Pharmacokinetic and Pharmacodynamic Data Management Functions

Description Usage Arguments Details See Also Examples

replace symbols or other character flags

1	replace_values(x, flag_df, nonflag = NULL, as_numeric = TRUE)

`x`	vector to replace
`flag_df`	dataframe with first column being the flag and second the replacement value
`nonflag`	what to do with non-flagged elements
`as_numeric`	whether to return column as numeric

because of R's coercion rules being somewhat unpredictable regarding character vs factor the behavior of replace_values is to always treat as character data, even if passed in as a factor

unique_non_numerics: to identify which non-numeric values must be replaced before column can be safely coerced to a numeric

df <- data.frame(ID = 1, DV = c(1, "BQL", ".", 5))
rflags <- data.frame(flag = c("BQL", "."), replacement = -99)
df$DVR <- replace_values(df$DV, rflags)

library(dplyr)
df <- df %>% mutate(DVR2 = replace_values(DV, rflags))

# powerful with unique_non_numerics
df <- df %>% mutate(DVR3 = replace_values(DV, 
                                 data.frame(values = unique_non_numerics(DV), 
                                            replacement = NA)))