Tidy evaluation

[hadley]

Work occurring in tidyeval branch

df <- data.frame(x = 0, y = 0)
ggplot(df, aes(x, y)) + geom_point()

z1 <- new_quosure(expr(z), env(z = 1))
z2 <- new_quosure(expr(z), env(z = 1:2))
aes(!!z1, !!z2)

ggplot(df, aes(x, y)) + geom_point()
ggplot(df, aes(!!z1, y)) + geom_point()
ggplot(df, aes(x, y)) + geom_point(aes(!!z1))

Need to figure out how to represent facetting spec using quosures

Small fixes :

• [x] make sure aes_() and aes_string() preserve existing interface
• [x] why did scales_add_defaults() need tryapply()? what breaks by removing it?
• [x] does using globalenv() instead of plot_env in find_scale cause problems? => tests assume find_scale can find local scales
• [x] test that subset still works

[hadley]

@lionel- did you look at either of the currently unchecked boxes?

[lionel-]

I don't think so, would you like me to?

[hadley]

Yeah, it’d be helpful to spend an hour or two on, in total.

[lionel-]

Sure thing, will do tomorrow morning.

[lionel-]

The tryapply() seems to have been there from the start of revision history: https://github.com/tidyverse/ggplot2/commit/6198457400c66d0774006c75353a1e7f1dd2806a#diff-70a17f07da991e42a800284db537be0aR106. You probably used it just to be on the safe side? It seems to make sense to just evaluate plainly though.

subset appears to have stopped working with the 2.0 rewrite: https://stackoverflow.com/questions/34588232/subset-parameter-in-layers-is-no-longer-working-with-ggplot2-2-0-0/34588455#34588455

[hadley]

Ok, I'll remove subset argument since it's clearly been broken for a while, and tryapply() since it no longer seems necessary.

[lock[bot]]

This old issue has been automatically locked. If you believe you have found a related problem, please file a new issue (with reprex) and link to this issue. https://reprex.tidyverse.org/