Ian Lyttle

Upon further reflection, I see `has_package()` would not be the right name. What I am proposing is more like `assert_package()`. (`has_package()` would be a vectorized wrapper to `requireNamespace()`) There are...

I have developed this function, now called `assert_packages()`, in one of my packages: https://github.com/vegawidget/vegawidget/blob/cran-release-0.1.0/R/utils-package.R I have found it useful, so I'd be happy to make a PR here if you...

I think the only motivation was to keep parity with `here()` - if you are used to `...` working there, you might expect for it to work with `i_am()`. Otherwise,...

Here's how I approached this problem, following the example from this related [Stack Overflow question](https://stackoverflow.com/questions/40407909/midpoint-of-discrete-diverging-scale-in-ggplot2): There's an internal function in ggplot2, [`mid_rescaler()`](https://github.com/tidyverse/ggplot2/blob/6207d2f8d6ad49a9eda670c874448d6fe454fad1/R/scale-gradient.r#L111) that I find to be very useful -...

I agree that it is an upstream wrangling problem, and sometimes (I think) the best solution is more than one dataset. The use case I have in mind is where...

I looked a little more at the problem; I suspect this is a Vega "thing", rather than a Vega-Lite "thing". I'm not so conversant with Vega, but I spent some...

Looks like it... ## `warnPartialMatchArgs` ``` r # expected getOption("warnPartialMatchArgs") #> NULL tmp to 'select' # *not* expected options(warnPartialMatchArgs = NULL) tmp Warning in subset.data.frame(mtcars, sel = mpg): partial argument...

@HenrikBengtsson - thanks! I agree that `NULL` should be deterministic rather than sticky. I have requested an account on Bugzilla, we'll see what happens!

Just to link this with the Bugzilla report: https://bugs.r-project.org/show_bug.cgi?id=18372

Could something like `write_args`, a list with arguments to pass to the writing-function, work here? (I'm certain there's a better name) ```r object_write