Closed
Description
I will be glad if you consider to support one of the metaprogramming scenario.
Unfortunately I find out quite dangerous side effect of styler styling.
Example where we will expect the styler to NOT add brackets in %>% pipeline:
styler::style_text('substitute(DATA %>% FUN_EXPR, env = list(FUN_EXPR = call("FUN")))')
# substitute(DATA %>% FUN_EXPR(), env = list(FUN_EXPR = call("FUN")))
# whereas should be neutral
# substitute(DATA %>% FUN_EXPR, env = list(FUN_EXPR = call("FUN")))
The usage of bquote might be a solution here nevertheless substitute is way more efficient:
> microbenchmark::microbenchmark(bquote = {bquote(x <- .(letters))}, substitute = {substitute(x <- ll, list(ll = letters))})
#Unit: nanoseconds
# expr min lq mean median uq max neval cld
# bquote 4387 4510 4809.71 4592 4674.0 24313 100 b
# substitute 164 246 281.67 246 266.5 2378 100 a