Hi!

I assumed that `characterp' and the `character' type class was connected in Common Lisp. If they aren't I guess the current system makes sense.

I downloaded the ert changes and ran all my local tests and I haven't seen any problems.

Thanks!

    -- Anders

On Fri, Dec 4, 2015 at 7:17 PM, Stefan Monnier <monnier@iro.umontreal.ca> wrote:
> That's indeed what should be done if ERT needs this to be a plain
> character that can inserted in a string.  Common-Lisp's `character'
> includes not just characters but also "characters with modifiers" such
> as ?\M-\H-é, which can't appear in a string and are rejected by
> `characterp'.

I installed a patch which makes ERT use pcase over cl-typecase.
In most cases it doesn't make a big difference, but in a few spots, it
is cleaner because a subsequent cl-destructuring-bind can be merged into
it (and it got rid of those places where we used (member :foo)
as a type to just check equality, which is rather inefficient).


        Stefan