rapidyaml VS text-unicode

Interesting, I hadn't come across extism before. How hard would it be to package https://github.com/biojppm/rapidyaml in this way? (And do you have a extism for dummies guide?)

The inability, or brittleness, to embed NUL bytes into the string, for once. Zeroed bytes can be valid as an internal bytes of a longer encoded character. And the inefficiency of tempting every caller to rederive the string length on every use, leading to such bugs as quadratic parsing behavior with sscanf. The extra register for an explicit length is a very minute price to pay compared to that.

Although you still have the other problem of sscanf, where read times can become quadratic.

You are really simplifying that issue. A lot of it has to be how the implementation of sscanf works and it's super easy to abuse by complete accident when a lot of data needs to be read and parsed to correct values. See more here: https://github.com/biojppm/rapidyaml/issues/40

Leading to eye-rolling problems like these: https://github.com/biojppm/rapidyaml/issues/40

We had to handroll our own OT implementation inhouse (based on https://github.com/ottypes/text-unicode) since we had built the system for it already. I suspect we would've used or forked CodeMirror's collab package if we were starting today.

Also huge fan of yjs, but the implementation is not compatible with codemirror 6, only version 5 (AFAIK)

Well said.

I've had a lot of conversations with javascript engineers over the years who've argued to me that well tuned JS will be nearly as fast as the equivalent C code. I've written plenty of little toy benchmarks over the years, and in my experience they're partly right. Well written JS code in V8 can certainly run fast - sometimes around half the speed of C code. But a massive performance gap opens up when you use nontrivial data structures. Nested fields, non-uniform arrays, trees, and so on will all cripple javascript's performance when compared to C's equivalent of simply embedding nested structs. If you couple clean C data structures with allocation-free hot paths from arenas, the performance of C will easily hit 10-20x the performance of the equivalent JS code.

From memory my plain text based operational transform code does ~800k transforms / second in javascript. The equivalent C code does 20M/second. The C implementation is about twice as many lines of code as the JS version though.

(The code in question: https://github.com/ottypes/text-unicode/blob/master/lib/type... / https://github.com/ottypes/libot )