On 12/13/25 22:33, Arthur A. Gleckler wrote:

> Thank you both.  Fixed and pushed.
>
> It amazes me that we still find bugs in widely used SRFIs so many years later — nineteen years later, in this case.

It would only be an issue if you *don'* use srfi-9 (record types), which at least
Kawa, Guile, and SISC do. If other Scheme implementation use SRFI-64 they have probably do something similar.
Also, from what I can see, you only run into that code-path
if you use test-apply, which "normal" tests would not. In the Kawa testsuite
it is only used in srfi-64-test.scm, the testsuite for srfi-64 itself.
--
	--Per Bothner
xxxxxx@bothner.com   http://per.bothner.com/