min-sized-rust VS regex

This is a good guide on building small Rust binaries: https://github.com/johnthagen/min-sized-rust

This talks about going to extreme lengths on making the smallest Rust binary possible, 400 bytes when it was written, https://darkcoding.net/software/a-very-small-rust-binary-ind...

The thing is, you lose a lot of nice features when you do this, like panic unwinding, debug symbols, stdlib… for kernel and some embedded development it’s definitely important, but for most use cases, does it matter?

Oh it was 137, haha. I will link you to this older comment of mine: https://news.ycombinator.com/item?id=29408906

See also https://github.com/johnthagen/min-sized-rust

Are you sure? If so then this is awesome news, but I'm a bit confused; the commit in that min-sized-rust repo adding `build-std` to the README was merged in August 2021: https://github.com/johnthagen/min-sized-rust/pull/30

Are you saying that at that point the feature still hadn't "landed in Rust nightly" until recently? If so then what's the difference between a feature just being available in Rust nightly, vs having "landed"?

Rust binaries are by default nowhere close to 500MB. If they are not small enough for you, you can try https://github.com/johnthagen/min-sized-rust. By avoiding the formatting machinery and using `panic_immediate_abort` you can get about the size of C binaries.

A useful reference: https://github.com/johnthagen/min-sized-rust

That must be without stripping. Also there are ways to reduce binary size. See e.g. [min-sized-rust](https://github.com/johnthagen/min-sized-rust). I've gotten stripped binaries of small cli utils less than 400KiB without doing anything special, less than 150 KiB by customizing profile settings and compressing with upx, and less than 30 KiB by replacing the std with the libc as the link shows. Haven't tried with fltk though...

This is not quite what you're asking, but it does also address the underlying concern: https://github.com/johnthagen/min-sized-rust

The homepage has a benchmark that compares Zed's "insertion latency" to other editors, and this is the description:

> Open input.rs at the end of line 21 in rust-lang/regex. Type z 10 times, measure how long it takes for each z to display since hitting the z key.

Could someone clarify what that means? My interpretation of that was to go to https://github.com/rust-lang/regex/blob/master/regex-cli/arg... and start typing 'z' at the end of line 21, but that doesn't seem to make any sense. I guess that repo got refactored and those instructions are out of date?

We also used the avenue to sluggify the question title. We used regex to fish out and replace all occurrences of punctuation and symbol characters with an empty string and using the itertools crate, we joined the words back together into a single string, where each word is separated by a hyphen ("-").

Command-Line Rust taught me how to use crates like clap, assert_cmd, and regex. I felt lost before because I didn't know about Rust's ecosystem--which is arguably as important as the language itself. Also, looking up and comparing libraries is a tiring task! blessed.rs is nice but Command-Line Rust really saved me from analysis paralysis.

burntsushi actually regrets making regex replace return a Cow: https://github.com/rust-lang/regex/issues/676#issuecomment-6.... I’m glad it does, and wish it took an impl Into> there, for the reasons discussed in the issue, but burntsushi has a lot more experience of the practical outcomes of this. Just something more to think about.

I'm not familiar with the AoC problem. You might be able to. But RegexSet doesn't give you match offsets.

You can drop down to regex-automata, which does let you do multi-regex search and it will tell you which patterns match[1]. The docs have an example of a simple lexer[2]. But... that will only give you non-overlapping matches.

You can drop down to an even lower level of abstraction and get multi-pattern overlapping matches[3], but it's awkward. The comment there explains that I had initially tried to provide a higher level API for it, but was unsure of what the semantics should be. Getting the starting position in particular is a bit of a wrinkle.

[1]: https://docs.rs/regex-automata/latest/regex_automata/meta/in...

[2]: https://docs.rs/regex-automata/latest/regex_automata/meta/st...

[3]: https://github.com/rust-lang/regex/blob/837fd85e79fac2a4ea64...

It’s not quite that simple, but folks are working on it.

https://github.com/rust-lang/regex/issues/425#issuecomment-1...

https://github.com/helix-editor/helix/pull/211#issuecomment-...

Dependencies used: - regex - unicode_reader - rust decimal - tokio

https://www.cs.princeton.edu/courses/archive/fall19/cos226/l... and https://kean.blog/post/lets-build-regex are excellent introductions to implementing a (very) simplified regex engine: construct a nondetermistic finite state automaton for the regex, then perform a graph search on the resulting digraph; if the vertex corresponding to your end state is reachable, you have a match.

I think this exercise is valuable for anyone writing regexes to not only understand that there's less magic than one might think, but also to visualize a bunch of balls bouncing along an NFA - that bug you inevitably hit in production due to catastrophic backtracking now takes on a physical meaning!

Separately re: the OP, https://github.com/rust-lang/regex/issues/822 (and specifically BurntSushi's comment at the very end of the issue) adds really useful context to the paragraph in the OP about niche APIs: https://blog.burntsushi.net/regex-internals/#problem-request... - searching with multiple regexes simultaneously against a text is both incredibly complex and incredibly useful, and I can't wait to see what the community comes up with for this pattern!