reproducible (version 0.2.5)

pipe2: Pipe that is Cache-aware, being deprecated

Description

STILL EXPERIMENTAL. THIS MAY NOT WORK AS ANTICIPATED.

Usage

lhs %>% rhs

Arguments

lhs

A value or the magrittr placeholder.

rhs

A function call using the magrittr semantics.

Details

This pipe will likely be deprecated, as it masks other pipes in the R ecosystem. This is fine, except to work, the reproducible package must be guaranteed to be first on the search path, which is almost impossible in any realistic project. Please see the %C% function ?pipe

A pipe that works with Cache. The code for this is based on a verbatim copy from https://github.com/tidyverse/magrittr/blob/master/R/pipe.R on Sept 8, 2017, based on commit 81c2e2410ebb7c560a2b4a8384ef5c20946373c3, with enhancements to make it Cache-aware. This version is a drop-in replacement for %>% and will work identically when there is no Cache. To use this, simply add %>% Cache() to a pipe sequence. This can be in the middle or at the end. See examples. It has been tested with multiple Cache calls within the same (long) pipe.

If there is a Cache in the pipe, the behaviour of the pipe is altered. In the magrittr pipe, each step of the pipe chain is evaluated one at a time, in sequence. This will not allow any useful type of caching. Here, if there is a call to Cache in the pipe sequence, the entire pipe chain before the call to Cache will have its arguments examined and digested. This digest is compared to the cache repository database. If there is an identical pipe sequence in the Cache repository, then it will return the final result of the entire pipe up to the Cache call. If there is no identical copy in the cache repository, then it will evaluate the pipe sequence as per normal, caching the return value at the point of the Cache call into the cache repository for later use.

See Also

pipe

Examples

# NOT RUN {
tmpdir <- file.path(tempdir(), "testCache")
checkPath(tmpdir, create = TRUE)
try(detach("package:magrittr", unload = TRUE)) # magrittr, if loaded, gives an error below
a <- rnorm(10, 16) %>% mean() %>% prod(., 6)
b <- rnorm(10, 16) %>% mean() %>% prod(., 6) %>% Cache(cacheRepo = tmpdir)
d <- rnorm(10, 16) %>% mean() %>% prod(., 6) %>% Cache(cacheRepo = tmpdir)
all.equal(b,d) # TRUE
all.equal(a,d) # different because 'a' uses a unique rnorm, 'd' uses the Cached rnorm

# Can put Cache in the middle of a pipe -- f2 and f4 use "cached result" until Cache
f1 <- rnorm(10, 16) %>% mean() %>% prod(., runif(1)) %>% Cache(cacheRepo = tmpdir)
f2 <- rnorm(10, 16) %>% mean() %>% prod(., runif(1)) %>% Cache(cacheRepo = tmpdir)
f3 <- rnorm(10, 16) %>% mean() %>% Cache(cacheRepo = tmpdir) %>% prod(., runif(1))
f4 <- rnorm(10, 16) %>% mean() %>% Cache(cacheRepo = tmpdir) %>% prod(., runif(1))
all.equal(f1, f2) # TRUE because the runif is before the Cache
all.equal(f3, f4) # different because the runif is after the Cache

unlink(tmpdir, recursive = TRUE)
# }