The purpose of cmdfun
is to significantly reduce the overhead involved in wrapping shell programs in R. The tools are intended to be intuitive and lightweight enough to use for data scientists trying to get things done quickly, but robust and full-fledged enough for developers to extend them to more advanced use cases.
Install the development version of cmdfun
with:
if (!requireNamespace("remotes"))
install.packages("remotes")
remotes::install_github("snystrom/cmdfun")
The cmdfun
framework provides mechanisms for capturing function arguments:
cmd_args_dots()
captures all arguments passed to ...
cmd_args_named()
captures all keyword arguments defined by the usercmd_args_all()
captures both named + dot argumentslibrary(cmdfun)
myFunction <- function(input, ...){
cmd_args_all()
}
(argsList <- myFunction(input = "test", boolean_flag = TRUE))
## $input
## [1] "test"
##
## $boolean_flag
## [1] TRUE
cmd_list_interp
converts the captured argument list to a parsed list of flag/value pairs.
## $input
## [1] "test"
##
## $boolean_flag
## [1] ""
cmd_list_to_flags
converts a list to a vector of commandline-style flags using the list names as flag names and the list values as the flag values (empty values return only the flag). This output can be directly fed to system2
or processx
.
## [1] "-input" "test" "-boolean_flag"
cmd_path_search()
allows package builders to search default locations for installed tools.
## [1] "//bin/ls"
cmdfun
attempts to solve the problem of wrapping external software in R. Calling external software is done with system2
or processx
.
For example, calling ls -l *.md
using system2
.
## [1] "-rw-r--r-- 1 snystrom its_employee_psx 1077 Aug 20 19:20 LICENSE.md"
## [2] "-rw-r--r-- 1 snystrom its_employee_psx 837 Aug 26 21:51 NEWS.md"
## [3] "-rw-r--r-- 1 snystrom its_employee_psx 7718 Aug 26 20:14 README.md"
## [4] "-rw-r--r-- 1 snystrom its_employee_psx 744 Aug 26 21:35 cran-comments.md"
However, when using multiple commandline flags each flag should be passed as a member of a character vector as follows:
When calling ls -l -i
## [1] "1163031755 -rw-r--r-- 1 snystrom its_employee_psx 1077 Aug 20 19:20 LICENSE.md"
## [2] "1163031757 -rw-r--r-- 1 snystrom its_employee_psx 837 Aug 26 21:51 NEWS.md"
## [3] "1163031758 -rw-r--r-- 1 snystrom its_employee_psx 7718 Aug 26 20:14 README.md"
## [4] "1163031762 -rw-r--r-- 1 snystrom its_employee_psx 744 Aug 26 21:35 cran-comments.md"
This becomes even more difficult if trying to support user input, as a significant amount of overhead is required to parse user inputs and optional flags into these vectors.
cmdfun
provides utilities for converting function arguments into lists which can easily convert to character vectors suitable for use with system2
or processx
.
library(cmdfun)
myFunction <- function(input, option1){
# Grabs named arguments as key/value pairs
cmd_args_named()
}
(argsList <- myFunction("myInput.txt", "foo"))
## $input
## [1] "myInput.txt"
##
## $option1
## [1] "foo"
## [1] "-input" "myInput.txt" "-option1" "foo"
ls
with cmdfunThese tools can be used to easily wrap ls
library(magrittr)
shell_ls <- function(dir = ".", ...){
# grab arguments passed to "..." in a list
flags <- cmd_args_dots() %>%
# prepare list for conversion to vector
cmd_list_interp() %>%
# Convert the list to a flag vector
cmd_list_to_flags()
# Run ls shell command
system2("ls", c(flags, dir), stdout = TRUE)
}
## [1] "LICENSE.md" "NEWS.md" "README.md" "cran-comments.md"
ls -l
can be mimicked by passing l = TRUE
to ‘…’.
## [1] "-rw-r--r-- 1 snystrom its_employee_psx 1077 Aug 20 19:20 LICENSE.md"
## [2] "-rw-r--r-- 1 snystrom its_employee_psx 837 Aug 26 21:51 NEWS.md"
## [3] "-rw-r--r-- 1 snystrom its_employee_psx 7718 Aug 26 20:14 README.md"
## [4] "-rw-r--r-- 1 snystrom its_employee_psx 744 Aug 26 21:35 cran-comments.md"
Commandline tools can have hundreds of arguments, many with uninformative, often single-letter, names. To prevent developers from having to write aliased function arguments for all, often conflicting flags, cmd_list_interp
can additionally use a lookup table to allow developers to provide informative function argument names for unintuitive flags.
For example, allowing long
to act as -l
in ls
.
shell_ls_alias <- function(dir = ".", ...){
# Named vector acts as lookup table
# name = function argument
# value = flag name
names_arg_to_flag <- c("long" = "l")
flags <- cmd_args_dots() %>%
# Use lookup table to manage renames
cmd_list_interp(names_arg_to_flag) %>%
cmd_list_to_flags()
system2("ls", c(flags, dir), stdout = TRUE)
}
## [1] "-rw-r--r-- 1 snystrom its_employee_psx 1077 Aug 20 19:20 LICENSE.md"
## [2] "-rw-r--r-- 1 snystrom its_employee_psx 837 Aug 26 21:51 NEWS.md"
## [3] "-rw-r--r-- 1 snystrom its_employee_psx 7718 Aug 26 20:14 README.md"
## [4] "-rw-r--r-- 1 snystrom its_employee_psx 744 Aug 26 21:35 cran-comments.md"
cut
with cmdfunHere is another example wrapping cut
which separates text on a delimiter (set with -d
) and returns selected fields (set with -f
) from the separation.
shell_cut <- function(text, ...){
names_arg_to_flag <- c("sep" = "d",
"fields" = "f")
flags <- cmd_args_dots() %>%
cmd_list_interp(names_arg_to_flag) %>%
cmd_list_to_flags()
system2("cut", flags, stdout = T, input = text)
}
## [1] "world"
## [1] "hello_hello"
Additionally, cmdfun
provides utilites for searching & checking valid tool installs, expecting system behavior, and helpful error handling to allow simple construction of external tool wrappers (see vignette for details).
See https://snystrom.github.io/cmdfun/articles/cmdfun.html for the most recent documentation and to learn about all cmdfun
features.
To file bug reports, please visit https://github.com/snystrom/cmdfun/issues while providing a reproducible example of your issue.