Docs for discrete_scale inaccurate

[alistaire47]

In the documentation for ?discrete_scale, the labels parameter is documented as

NULL for no labels, waiver() for default labels (labels the same as breaks), a character vector the same length as breaks, or a named character vector whose names are used to match replacement the labels for matching breaks.

However, in ?scale_x_discrete, the final example shows a very important alternative: passing a function as a formatter:

# Use abbreviate as a formatter to reduce long names
ggplot(mpg, aes(reorder(manufacturer, displ), cty)) +
  geom_point() +
  scale_x_discrete(labels = abbreviate)

and thus should probably be documented more like continuous_scale's labels parameter:

One of:

• NULL for no labels
• waiver() for the default labels computed by the transformation object
• A character vector giving labels (must be same length as breaks)
• A function that takes the breaks as input and returns labels as output

I'm happy to assemble a quick PR if you like, but will everything get redone in the course of dealing with #1750 anyway?

[hadley]

A PR would be handy. It'll be needed for #1750 too.

[hadley]

Oooh, if you were interested in a few more docs tweaks, I think an partial fix for #1750 would be to use the new @inheritDots from the latest version of roxygen2.