voodooattack

This is something I have been working on as part of a larger project where we use it for programmatic test generation. This extended syntax allows the inclusion of type information within a document.

This means that after parsing, a user-defined deserializer can pick up the raw value and do anything with it (whatever it returns is included in the output of `parse`), and that a user-defined validator can allow or reject it with an error.

Documentation (mainly for deserializers/validators) is currently sparse since I'm somewhat busy with a deadline, the tests have many examples though. I plan to expand on it later.

Here is a glance at what it looks like: ``` { x: date "2001-01-01", // it can handle custom deserializers! y: 123n, // it can handle `BigInt` out of the box! z: /^I\sCAN\sPARSE\sREGEX$/im, // ...and also regular expressions! w: undefined, // ...and even `undefined`! } ```

There are more examples on GitHub!

I use CLion, I intentionally edited the Java VM settings to allow it to consume more memory.

It takes forever to analyse symbols for WebKit otherwise. WebKit is too massive to handle with the default memory limits.

That's exactly what I'm trying to change. JavaScript should be the right tool for that level of performance. It has JIT compilers, now it needs concurrency, and it'll perform with the best of them.

Try writing a raytracer using a single thread, now try again with multiple threads working in tandem.

Trust me, I tried.

This happens when there is no synchronization involved, the engine is printing line endings in a separate operation, so they come out weird like that. I'll add a mutex locking mechanism on console I/O at some point in the future.

Yeah, they're all executing on different cores.

We said no, the lawyer told him so, he asked for direct contact details and we haven't heard anything from him yet.