Mark Thom

Thanks, it should make a good test case for the GC.

Everything you've observed is true. The idea for meta-predicates like `maplist/N` is to keep their `call/N` calls compiled as they are but to wrap their goal arguments in the '$call'...

You're right to point out that '$call/N' isn't necessary. Two points in favor of the approach I outlined above, though: 1. Compiling a `maplist_for_p` for every predicate `p` used in...

> This issue happens because the number of registers is limited: (etc) True, but this could be better handled. I think, if the code generator allocates register 1024, an error...

This issue was finally fixed by a recent commit.

I'm not able to reproduce the `subsumes_terms/2` bug.

> Are you trying to reproduce those bugs just by typing the test goals at the top-level or by running the Logtalk tests? Typing them at the top-level. I manually...

> Partial evaluation telle quelle is quite a rabbit hole to fall into. What might be a way out is to take the meta_predicate declarations as (one) precondition for such...

It's just the result of considering the meta-predicate declaration.

The outline Jan Weilemaker gives in the second post of this page https://discourse.swi-prolog.org/t/how-is-goal-expansion-for-closures-defined/3365 is exactly the idea I have in mind. But it appears (on StackOverflow for example) that people...