This article will walk you through using the promoutils functions to schedule messages for Linkedin. First make sure you are authorized to work with the LinkedIn API:
keys_check()
#> Key status
#> ✔ slack: TRUE
#> ✔ matomo: TRUE
#> ✔ linkedin: TRUE
#> ✔ linkedin_org: TRUE
#> ✔ github: TRUE
#> → Good to go!If not, see the Setting Up LinkedIn article.
Accessing the API
Once you have set up your LinkedIn access, you can use the LinkedIn API to access the rOpenSci LinkedIn space using httr2 directly or via the promoutils helper functions.
Most of the time we’ll be using the ro_urn object to
define the ‘author’ (i.e. rOpenSci) of our requests.
For example, to get a list of recent posts by rOpenSci we can use
p <- li_posts_read(ro_urn)We can print the most recent message with
p$elements[[1]]$commentary |> cat()
#> 👋 Meet Philippe Massicotte! 🇨🇦
#>
#> Research Assistant and maintainer of \{rnaturalearth\}, \{rnaturalearthdata\}, and \{gitignore\} for rOpenSci. The first two packages facilitate mapping by providing world vector maps and data. \{gitignore\} helps you to create useful .gitignore files for your project.
#>
#> Find Philippe at:
#> 🐘 \@philmassicotte\@fosstodon.org
#> 💻 github.com/PMassicotte
#> 🔗 https://lnkd.in/dw7nB-7y
#>
#> {hashtag|\#|MaintainerMonth} {hashtag|\#|rOpenSci} {hashtag|\#|RStats}To write a post you can use the li_posts_write()
function. Let’s do a dry run (which won’t actually post).
id <- li_posts_write(
author = ro_urn, # Post on behalf of rOpenSci
body = "Testing out the LinkedIn API via R and httr2!",
dry_run = TRUE
)
#> POST /rest/posts HTTP/1.1
#> accept: */*
#> accept-encoding: deflate, gzip, br, zstd
#> content-length: 291
#> content-type: application/json
#> host: api.linkedin.com
#> linkedin-version: 202510
#> user-agent: httr2/1.2.2 r-curl/7.1.0 libcurl/8.5.0
#> x-restli-protocol-version: 2.0.0
#>
#> {
#> "author": "urn:li:organization:77132573",
#> "commentary": "Testing out the LinkedIn API via R and httr2!",
#> "visibility": "PUBLIC",
#> "distribution": {
#> "feedDistribution": "MAIN_FEED",
#> "targetEntities": [
#>
#> ],
#> "thirdPartyDistributionChannels": [
#>
#> ]
#> },
#> "lifecycleState": "PUBLISHED",
#> "isReshareDisabledByAuthor": false
#> }It’s a good idea to capture the returned id (URN) in
case you want to remove the post later (but you can always get this with
li_posts_read() too).
If you want to post from your personal account, you’ll need to use
your personal ID (URN). You can fetch this with
li_urn_me().
li_urn_me()
#> urn:li:person:Bxn4HyByQ5To schedule messages we recommend using a GitHub action on a cron job (or similar).