rlang

rlang copied to clipboard

Should dynamic dots support checking whether all elements are named?

1

`rlang` offers a family of functions that operate on list expressions (this includes `enxprs`, `enquos`, but also `dots_list`). Most of these functions offer optional support for enforcing list element names...

tzakharko

Should name-unquoting work everywhere in the AST?

3

```r expr(c(!! "foo" := 1)) ```

lionel-

lionel-

More robust data mask rechaining

1

Currently the lexical scope of environments that escape data masking contexts is rather undefined. These environments consistently include the masked data but beyond that it depends which quosure was first...

lionel-

Clarification on .homonyms?

1

Hi I have trouble understanding the use of `.homonyms` in `enquos()`. I thought it would keep/filter arguments of the same name, which I understand to be inputs in `...`. Now...

MatthieuStigler

`as_label()` and `expr_text()` should use `expr_deparse()`

1

Instead of deparse

hadley

`eval_tidy()` with `...` in two-sided formula

2

```r rlang::eval_tidy(... ~ 1) #> Error in rlang::eval_tidy(... ~ 1) : '...' used in an incorrect context rlang::eval_tidy(1 ~ ...) Error in rlang::eval_tidy(1 ~ ...) : '...' used in an...

lionel-

We could replace formulas by `rlang::q(expr, )` calls (or `rlang::q(expr)` if we store the env in attributes like with formulas, but I like @brodieG's idea of leaving it apparent in...

lionel-

list2() doesn't evaluate arguments in way ellipsis::check_dots_used() checks

2

``` r f Error: 1 components of `...` were not used. #> #> We detected these problematic arguments: #> * `a` #> #> Did you misspecify an argument? ```

yutannihilation

See https://github.com/tidyverse/purrr/issues/749.

lionel-