Nenad Rakocevic

I have changed the `trim` default behavior on `binary!` to remove NUL bytes (commit: https://github.com/red/red/commit/e707df880de6c9c3652886262eeb200e01f3a7b2) > The other change I propose is that /with should NOT imply /all. That one...

It works fine with the latest version.

I have tested on W10, we don't officially support Win7 anymore since a while. Though, if it's a Win7-specific issue, I guess it should not be a serious one.

At this point, it's most likely not an issue with Red, but your local environment. If you have any third-party anti-virus, disable it. Also disable the Windows Defender AV entirely...

> I would expect GUI console built with -d also to crash with a message though... strange if it's silent It also needs to target the shell instead of the...

> Note: I'm working on HOF-related foreach, so fixing this is not necessary. I don't see how it's related? This needs to be fixed, even if it's an edge case.

I can't find any good fix as the current stack management implementation has grown too messy, with not well-defined purpose/usage of internal stack frame flags. I would defer that issue...

Just to be clear, this is not a compiler (vs interpreter) limitation strictly speaking, but an implementation problem. The future unified Red stack will solve that issue.

For a simple function call, the arguments cache array is stored in the `function!` value itself, while for a function call with refinements, it's a specific occurence that is then...

@greggirwin It's not dependent on locals specifically, it's about using a function A spec to fetch the arguments and set locals for a function B. This only happens in interpreter...