Github developmental version of magrittr includes some cool new function for piping but I do not exactly catch de difference between %>% and %,%. Is this only formal with %>% for value and %,% for functions, or there is some specific peculiarity?
The normal piping operator is %>%. You can use %,% to create a reusable pipe, a pipe without data. Then later you can use the same pipe with various data sets. Here is an example.
library(magrittr)
library(dplyr)
library(Lahman)
Suppose you want to calculate the top 5 baseball players, according to total hits. Then you can do something like this (taken from the magrittr README):
Batting %>%
group_by(playerID) %>%
summarise(total = sum(G)) %>%
arrange(desc(total)) %>%
head(5)
# Source: local data frame [5 x 2]
#
# playerID total
# 1 rosepe01 3562
# 2 yastrca01 3308
# 3 aaronha01 3298
# 4 henderi01 3081
# 5 cobbty01 3035
So far so good. Now let's assume that you have several data sets in the same format as Batting, so you could just reuse the same pipe again. %,% helps you create, save and reuse the pipe:
top_total <- group_by(playerID) %,%
summarise(total = sum(G)) %,%
arrange(desc(total)) %,%
head(5)
top_total(Batting)
# Source: local data frame [5 x 2]
#
# playerID total
# 1 rosepe01 3562
# 2 yastrca01 3308
# 3 aaronha01 3298
# 4 henderi01 3081
# 5 cobbty01 3035
Of course you could also create a function the regular R way, i.e. top_total <- function(...) ..., but %,% is a more concise way.