This is done before, but worth to put into issues as we may need to deal with it later.
Using pacman instead of common library() in code has advantage that it will install packages not available in user's machine. So user can run from github and update packages automatically.
However pacman is not recognized by shinyapps.io so it will fail in deployment.
In the backend shinyapps.io used packrat to manage package dependencies. I made some changes to packrat to make it recognize pacman.
RStudio don't plan to take the change because that means they need to support the feature officially. The ideal solution is for them to have some extension mechanisms.
Before that, I'm using my changed fork of packrat to deploy to shinyapps.io.
This is done before, but worth to put into issues as we may need to deal with it later.
pacmaninstead of commonlibrary()in code has advantage that it will install packages not available in user's machine. So user can run from github and update packages automatically.pacmanis not recognized byshinyapps.ioso it will fail in deployment.shinyapps.iousedpackratto manage package dependencies. I made some changes topackratto make it recognizepacman.packratto deploy toshinyapps.io.