burntsushi / blog Goto Github PK

Line ~1245:
[I adapted the algorithm from the Hyperscan project to Rust](https://github.com/rust-lang-nursery/regex/blob/master/src/simd_accel/teddy128.rs), and included an extensive write up in the comments if you're interested.
Current: https://github.com/rust-lang/regex/blob/master/src/literal/teddy_ssse3/imp.rs
Or as found earlier in the post: https://github.com/rust-lang-nursery/regex/blob/3de8c44f5357d5b582a80b7282480e38e8b7d50d/src/simd_accel/teddy128.rs

If you'll try to follow

[`utf8-ranges`](https://burntsushi.net/rustdoc/utf8_ranges/)

you'll hit redirect to https://docs.rs/utf8_ranges/ (note the _), which is 404.

Doing s/_/-/ in the URL fixes this, and I almost created pull request with that very fix, but decided to just report this once I realized that I can also just replace both /rustdoc/ URLs in the text with direct links to docs.rs, and I don't really know what's better. (And if docs.rs is preferred, should links include versions from ripgrep's Cargo.toml at the time of writing, or just redirect to latest versions? Gaaah, I thought this was going to be an easy fix!).

Hi Andrew,

I just wanted to say thank you for the blog post. They were so good and helpful. I have been struggling to grasp rust's concept for quite some time and for the first time I felt like I understand them. Thank you for your time and effort.

I liked your guide to error-handling in Rust. One thing that stumped me for a bit, though, was the file_path_ext_explicit version of option-ex-string-find, which I'll paste here for convenience:

fn file_path_ext_explicit(file_path: &str) -> Option<&str> {
    match file_name(file_path) {
        None => None,
        Some(name) => match extension(name) {
            None => None,
            Some(ext) => Some(ext),
        }
    }
}

fn file_name(file_path: &str) -> Option<&str> {
  // implementation elided
  unimplemented!()
}

I understand that you were trying to illustrate the full case analysis without any cleanup, but it really bothered me that the inner match block was a complete no-op, and could have been replaced by Some(name) => extension(name). I wonder if comments might help here, perhaps something like:

    match file_name(file_path) {
        None => None,
        Some(name) => match extension(name) { // match to handle Option result
            None => None,
            Some(ext) => Some(ext) // Re-wrap ext to return an Option
        }
    }

...perhaps along with some note along the lines of "yes you could get rid of that match, but it's shown for pedagogical purposes here".

It's not a huge thing, but I did spend a bit of time checking if I'd missed something, so anything to indicate "no, it's not your imagination that this is weird" might be nice. :-)

Howdy, hit your blog via HN article on your ripgrep tool at http://blog.burntsushi.net/ripgrep/ and noticed that your canonical reference is to the non-trailing slash version, which then 301 redirects to the trailing slash version. This could/would hurt the way Google crawls/indexes your content. I thought you'd want to know :)

Hello, I am very impressed by the content you put on your blog but was disappointed that there is no feed that I could subscribe to. If you could provide a feed that my robots could monitor it would be greatly appreciated because I am sure to forget to do so manually.

Thanks, Kevin