prechelt-phone-number-encoding VS prechelt-phone-number-encoding

You're looking at the non-optimized version. If you read the blog post you would've seen your suggestions had already been implemented.

> it’s using Unicode-aware string stuff

Rust uses UTF-8 internally for Strings, so it's very efficient to parse a file into a String, then using slices to go through it... this is probably the best you can get as parsing ASCII input as UTF-8 is very efficient (the 0-bit is always zero in ASCII, the unicode decoder only needs to check that's the case for every byte, so it's not some kind of complicated computation it's doing to decode)...

If you use bytes for everything, you will make the whole code much harder to follow and it still won't run faster.

Check for yourself: https://github.com/renatoathaydes/prechelt-phone-number-enco...

This is a pretty introductory CL article, mostly a commentary on Norvig's solution to the problem. Still, I learned about the #. readmacro from it. The conclusion: "[The Lisp implementation] was the fastest implementation for all input sizes except the largest one, where it performed just slightly worse than my best Java implementation." GH repo at https://github.com/renatoathaydes/prechelt-phone-number-enco.... Sounds like he was mostly measuring the performance of the SBCL bignum implementation.

Here are the commits I've made so far. If anyone wants to help write the most efficient possible Lisp implementation, please send suggestions here!

OMG you're right... I'm the author, and the reason it was allocating in my original code was that I was calling the operators on a reference to n. See https://github.com/renatoathaydes/prechelt-phone-number-encoding/commit/6683dc10cc4fb380abead632b87d94e8937f8377

This commit show how to improve that: https://github.com/renatoathaydes/prechelt-phone-number-encoding/commit/561a7307b5574bd6fd7b8cc638abf6f29884b6ca

See https://github.com/renatoathaydes/prechelt-phone-number-encoding/issues/6

I get the need to want to obsessively optimize the code. There's nothing more fun than to optimize something simple, artificial, and narrowly-defined. But y'all need to take a deep breath, step back, and realize that one blog post isn't going to suddenly define the language (nor should it personally define you).

source: https://github.com/renatoathaydes/prechelt-phone-number-enco...

In this case the linked list is just tracking "breadcrumbs" which mirror the path back up the call chain, so you don't even need reference counting; a simple borrowed reference will do. I put together a version in Rust based on stack-allocated linked lists[0], and the results were promising: on 10 million inputs the Rust version completes in about 40 seconds on my machine, while the Java port (Main2.java in the code; "Java1" from the article IIUC) takes 82 seconds. This Rust version doesn't allocate anything from the heap, apart from loading the dictionary at startup; all the bookkeeping, including the cons cells, is kept on the stack.

[0] https://github.com/nybble41/prechelt-phone-number-encoding