languageserversetup

Setup and initialize R language server seamlessly and automatically with editors such as VS Code across platforms.

Features

Installs the languageserver package and all of its dependencies into a separate, independent library
Adds code to .Rprofile to make startup of the language server seamless when appropriate
Checks whether an R process is triggered as an instance of the languageserver and if so uses the independent library to prevent conflicts with other packages that the user uses
The languageserver installation does not affect the packages users currently have installed

Installation

You can install languageserversetup from CRAN. It has no dependencies, so no other packages are installed:

install.packages("languageserversetup")

You can also install the latest development version from the master branch on GitHub using the remotes package:

remotes::install_github("jozefhajnala/languageserversetup")

Usage

Install the languageserver package and all it's dependencies into a separate independent library (Will ask for confirmation before taking action)

languageserversetup::languageserver_install()

Add code to .Rprofile to automatically instantiate languageserver if the process is an instance of the languageserver, otherwise, the R session will run as usual with library paths unaffected

languageserversetup::languageserver_add_to_rprofile()

Enjoy the cool functionality with VS Code or other editors

In action with VS Code

Install languageserversetup and use `languageserver_install()`

Initialize the functionality with `languageserver_add_to_rprofile()`

All done, enjoy the awesomeness!

Platform support

Currently, the functionality is tested on 64bit versions of MS Windows 7, 10, Ubuntu 18.10 and MacOS El Capitan. Automated deployments and tests run via GitHub actions on macos-elcapitan-release, windows-x86_64-devel, ubuntu-gcc-release and fedora-clang-devel.

All PR and issues related to platform support are most welcome!

Removing the functionality

To remove the functionality, run (Will ask for confirmation before taking action):

languageserversetup::languageserver_remove_from_rprofile()

If the above does not succeed, remove the languageserersetup related code from your .Rprofile. Optionally, you can also delete the library where languageserver was installed itself.

Customizing the behavior

Location of the languageserver library

The default location for the library is path.expand(file.path("~", "languageserver-library")).
To customize the location of the library, set options(langserver_library = "/desired/path/to/library") before running languageserver_install().
To tweak the behavior manually you can always change the .Rprofile code chunk created by languageserver_add_to_rprofile() to your liking manually. Since the installation is stand-alone, if you want, you can also manually move the entire library to a new location by simply copying the library directory.

Used `.Rprofile`

By default, an .Rprofile file located in the user's home directory is used. If it does not exist, it is created. If it does exist, code is appended to the end of the file (after user confirmation).

Development or CRAN version of languageserver?

The languageserver_install() function has a fromGitHub argument. Set it to FALSE to install the CRAN version of the languageserver package. Otherwise, the latest development version is installed from the master branch of the languageserver GitHub repository.

Options

The default package behavior can be adjusted by the following options, ideally placed in the .Rprofile file before the call to library(languageserversetup)

for troubleshooting purposes, use options(langserver_quiet = FALSE) to enable diagnostic messages
use options(langserver_quiet_serverproc = TRUE) to disable diagnostic messages for the language server identified process
to define a custom library location for the languageserver package and its dependencies, use options(langserver_library = "desired/path/to/library")

How does it work, extra options and arguments

Please refer to the help files:

?languageserver_install
?languageserver_startup
?languageserver_add_to_rprofile
?languageserver_remove_from_rprofile

Acknowledgments

This package would have no point if there was no implementation of the language server protocol for R
Thanks to rhub for making cross-platform checking and testing of R packages as easy as it gets
Thanks to the Rocker project for providing useful Docker images for R work

License

AGPL-3