depends vs imports on ggplot/shiny about shinystan

[jgabry]

I agree with all of this, and I'm not sure what the right answer is for ggplot2. As you say, shinystan really does depend on ggplot2 and my hunch is that any problems with depending on ggplot2 and not using ggplot2::fun would be limited to cases like the one in #82 and wouldn't occur in regular use of shinystan outside other packages.

However, it's cool that someone's building a package that uses shinystan and it'd be nice to support such endeavors. The easiest/quickest solution would be to just say that any package that is going to call launch_shinystan should also depend on ggplot2, as I don't think ggplot2 would interfere with much.

But I'm not opposed to going the ggplot2::fun route. Let me take a look at a few things quickly and then I'll post a follow-up.

[dpastoor]

sure for now I'm tracking in #84 it looks like it shouldn't be that hard to nail down most things by just recursively searching for specific components in ggplot

[jgabry]

It should also be almost entirely within the files under helper_functions, with the exception of the ppcheck ones, which aren't in the same directory for a reason that isn't relevant anymore.