Vadim Belman

Good you reminded me about this case. Probably it should. I failed to find the discussion where I get the information about the grammar from to make sure if it...

@lizmat If memory serves me right, cores are cached in memory. So, requesting one doesn't affect the overall loading much comparing to the disk operations.

Apparently, the preferable fix would be to add `nqp::existslex*` (or `nqp::lex*exists`) family of ops. This would result in a little complication of symbol lookup code, produced by the compilers. But...

I have tripped over this problem while working on rakudo/rakudo#4495. One of the problems with pseudos which always annoyed me is that `.WHO:exists` is not always true even if `sym`...

`$*DISPATCHER` might not be the only symbol of the kind. Anyway, of course I can drop the idea of testing for roundtrip behavior of pseudos because it would be unreliable....

First of all, you can't do `$*FOO` if it's `VMnull`. The problem is: ```raku for DYNAMIC::.keys -> $sym { ok DYNAMIC::{$sym}:exists, "$sym exists"; # not ok on $*DISPATCHER } ```...

> maybe `DYNAMIC::.keys` should not produce `$*DISPATCHER` if it is `nqp::null` ? @lizmat I'm considering this. Not only `$*DISPATCHER`, but any other. After all, a purpose for the PR is...

> > maybe `DYNAMIC::.keys` should not produce `$*DISPATCHER` if it is `nqp::null` ? > > @lizmat I'm considering this. No, it's not an option. Things are totally different about lexicals....

> There are lots of ways you can break Rakudo and get completely bogus results by using nqp. I'm not talking about breaking, but rather about having a faster way...

My bad, I haven't done it in first place, but a search in _packages/Test-Helpers_ came up with `make-temp-*` family of routines in `Test::Util`. Moreover, they record all files/dirs created and...