Oscar N.

Great work on this @JohnAD :) I tried out your fork and wrote a couple of simple plugins. Some feedback (note: I haven't read all the comments in the PR,...

@JohnAD > The plugins are not supposed to be global (to all the routers.) In fact, I use different plugins and plugin settings in different routers on my websites. But...

`input.json5` contains `"\uDEAD"`. After running `json5 input.json5 -o output.json`, `output.json` contains the following bytes (in hex): `22 EF BF BD 22`, which is a string only containing the replacement character.

Looks like a bug or limitation with `strictFuncs` to me. This is the stdlib code in `newXmlTree` the compiler complains about: ```nim for i in 0..children.len-1: result.s[i] = children[i] ```...

@andreaferretti here is an annoying case with `{.discardable.}`: ```nim proc foo(): int {.discardable.} = discard # Does bar return void or int? # Currently it will return void, but with...

> makes clear distinction for callers bw runtime args and compile time types, eg: This isn't true in general for Nim, since `static[T]` parameters look like normal parameters but are...

There already is an `Endianness` enum + `cpuEndian` const in system that can be used, no need to introduce another enum.

I'm not sure I fully understand this discussion, but it's my understanding that `\xHH` is not a byte escape sequence, it's a unicode escape sequence for the range `U+0000-U+00FF`. From...

Very late answer, but no. This is not trivial to implement even on unix systems

I usually avoid reference types in Nim and I'm not sure samson should support it. Typically there shouldn't be a reason to use a ref type for serialization/deserialization. I think...