Repository navigation
[Request] data.table join nomatch=quote(stop()) #940
Description
Feature Request.
Related to #857.
We have 2 dts:
library(data.table)
dict <- data.table(symbol = c("a","b","c"), ratio = c(5,10,15), key = "symbol")
dt <- data.table(symbol = c("a","b","c","d"), value = rnorm(4))Which we want join:
dict[dt][,list(symbol, new_value = ratio * value)]Consider the case when we expect to have match on all data in the join.
In case of nomatch we want to raise error.
Currently it is possible using:
new_dt <- dict[dt]
if(any(is.na(new_dt$ratio))) stop("missing data in the reference dictionary")Which makes us unable to make chaining in such case.
If we could use something like this:
dict[dt,nomatch=quote(stop("missing data in the reference dictionary"))
][,list(symbol, new_value = ratio * value)]just to stop process and raise the error in case of any nomatch.
Activity
[-]data.table join nomatch=quote(stop())[/-][+][Request] data.table join nomatch=quote(stop())[/+]arunsrinivasan commentedon Nov 21, 2014
Or
nomatch = "error"like #655. I'd say this should be added to #857, and closed, as this is very related to #857.jangorecki commentedon Dec 13, 2014
merged with #857.