On 2026-09-21 21:52, Artyom Bologov wrote:

> Andrew,
>
>> Thoughts?
>
> I haven’t been tracking SRFI 269 development closely, but this
> particular change seems sensible and useful. STklos and Gauche use
> leading descriptions in their test libraries, so the idea is properly
> represented out there.

I did a small survey on fediverse, it seems that the description as the
last argument is prefered by people:
https://fosstodon.org/@abcdw/117314139715386132

Also, in the new draft, we introduced a testing macro, which allows to
annotate a group of assertions.

It works well both visually and semantically.

(testing "home is writable"
  (define status (spawn "touch" path))

  (is (= status 0))
  (is (file-exists? path)))

Now I don't need to annotate every assertion, because I have
"assertion/context", which clearly tells what we are doing here.

Descriptions for particular assertion still can be useful, but much more
rare, so we don't need them to be in "the right reading position"
anymore and can left them in the tail position.

--
Best regards,
Andrew Tropin