search

tidyverse / multidplyr

A dplyr backend that partitions a data frame over multiple processes
https://multidplyr.tidyverse.org
Other
641 stars 75 forks source link

Issue attaching multiple libraries to each node #21

Closed Mullefa closed 7 years ago

Mullefa commented 8 years ago

The documentation infers you can attach multiple packages to each node with a single call to cluster_library(). However, this doesn't work, at least in my case:

get_default_cluster() %>%
  cluster_library(c("dplyr", "geosphere"))

Initialising 7 core cluster.
Error in library(packages, character.only = TRUE) : 
  'package' must be of length 1
puterleat commented 8 years ago

I think the whole issue of attaching packages is quite confusing at the moment... It took me a while in the docs to realise you basically have to recreate the namespace of functions within each node of the cluster... I wonder if there's a way to automate this? Could the current environment be copied to each node?

hadley commented 7 years ago

Fixed by #43