Note: slackr
is seeking a new maintainer! Please see this issue
slackr
provides a set of tools for making it easier to send messages,
data, alerts, etc. directly from R to Slack. You can use this package to
send well-formatted output from R to all teammates (or to specific
individuals) at the same time with little effort. You can send text, R
function output, images from the current graphics device and ggplots
,
R objects (as R data files), rendered LaTeX expressions, and uploaded
files.
# CRAN version
install.packages("slackr")
# Development version
devtools::install_github("mrkaye97/slackr")
Version 3.0.0+
removes all references to bot_user_oauth_token
(deprecated in v2.4.0
) in favor of token
. There have also been
significant changes to how slackr
and slackr_bot
handle errors. See
the changelog for
more details.
Version 2.4.0+
now allows users to choose between using a bot token
and a user token. See below for details and check the changelog
(NEWS.md
) for more changes.
There are three ways of interfacing with slackr
that provide
significantly different functionality:
-
Creating a single-channel bot
Using only a webhook to send messages to a channel
-
Creating a fully-functional multi-channel bot
Creating a bot user to send messages to multiple channels, including plots, tables, files, etc. as well as deleting messages, reading the channels in a workspace, etc.
-
Using a user token to send messages from a specific user’s account
Similar to the fully-scoped bot token, but connected to the account of a single user. This approach is not recommended in production settings – or any settings where a token needs to be shared – but it can be useful for one-off Slack messages as it lets users send data as themselves as opposed to through a bot.
In most cases, we recommend Option 1
above. This requires the fewest
permissions and is the simplest to set up, and will allow basic
messaging to a specific channel.
See the vignettes for setup instructions.
The vignettes contain setup instructions and example usage:
- Option 1 setup:
vignette('scoped-bot-setup', package = 'slackr')
- Option 2 setup:
vignette('webhook-setup', package = 'slackr')
- Usage:
vignette('using-slackr', package = 'slackr')
Important Note: The setup process for Option 2
and Option 3
are
roughly the same, with only slightly differing scopes.
The slackr_setup()
function will try to read setup values from a
~/.slackr
(you can change the default filepath by recording in the
SLACKR_CONFIG_FILE_PATH environment variable or supplying as an argument
to the config_file
parameter) configuration file, which may be easier
and more secure than passing them in manually (plus, will allow you to
have multiple slackr
configurations for multiple Slack.com teams).
The file is in Debian Control File (DCF) format since it really doesn’t
need to be JSON and R has a handy read.dcf()
function since that’s
what DESCRIPTION
files are coded in.
Here’s the basic format for the configuration file:
token: xox*-<your app's token>
channel: #general
username: slackr
incoming_webhook_url: https://hooks.slack.com/services/XXXXX/XXXXX/XXXXX
icon_emoji: 'boom'
As of slackr 2.3.0
, you can create a config file with
create_config_file()
instead of setting it up manually. See the docs
for details.
You can also change the default emoji icon (from the one you setup at
integration creation time) with icon_emoji
.
Many thanks to:
- Bob Rudis
- Jay Jacobs
- David Severski
- Quinn Weber
- Konrad Karczewski
- Ed Niles
- Rick Saporta
- Jonathan Sidi
- Matt Kaye
- Xinye Li
- Andrie de Vries
for their contributions to the package!
- Depending on your scopes,
slackr
could quietly fail (i.e. not throw an error, but also not post anything to your channel). If this happens, try explicitly adding the app you’re trying to haveslackr
post as to the channel you want in your Slack workspace with/invite @your_app_name
or make sure you havechat:write.public
enabled.