Reputation: 60
So I have created an environment (which I am trying to use as a hashtable).
To clarify I'm accessing the values stored in the environment with this:
hash[["uniqueIDString"]] ## hash takes a uniqueID and returns a
## dataframe subset that is precomputed
I also have a function called func which returns some subset of the rows returned by hash. It works fine for single calls but it isn't vectorized so I can't use it in a transform which is kind of vital.
The following doesn't work:
df <- transform(df,FOO = func(hash[[ID]])$FOO)
It gives me an error about having the wrong number of arguments for the hash which I presume is because it's passing a vector of IDs to my environment and the environment doesn't know what to do.
EDIT: The exact error is:
Error in hash[[ID]] :
wrong arguments for subsetting an environment
EDIT: With Rob's suggestion I receive the following error:
Error in hash[ID] :
object of type 'environment' is not subsettable
EDIT: For clarification I'm attempting to use the hash lookup in the context of a transform where the values in the ID column are looked up in the hashtable and passed to func so that the output can become a new column.
Upvotes: 1
Views: 3171
Reputation: 4359
I use environments as hash tables a lot, the way to retrieve values corresponding to multiple keys is to usemget
:
hash <- new.env()
hash[['one']] <- 'this is one'
hash[['two']] <- 'this is two'
mget(c('one', 'two'), envir = hash)
which returns a list
$one
[1] "this is one"
$two
[1] "this is two"
If you need the output as a vector, use unlist
:
unlist(mget(c('one', 'two'), envir = hash))
gives you
one two
"this is one" "this is two"
UPDATE If your IDs come from a data frame, you'd need to use unlist
to convert that column to vector:
df <- data.frame(id=c('one', 'two'))
unlist(mget(unlist(df$id), envir = hash))
Upvotes: 4