I mean serde is -- in my understanding -- the most useful crate out there. It does exactly one thing and that very well.

Here are some I found and used in my own code:

• itertools
• regex
• anyhow and thiserror (error handling)
• indoc (indented/formatted multi line string literals)
• strum (various derive macros for enums)
• petgraph (for working with general graphs)
• winnow is a great (and fast) parser combinator library.
• bpaf, clap and xflags are three different command line argument parser libraries. Which one to use depends on the needs of the project and if you need to match the behaviour of an existing non-rust program (as I needed to in one case)

If you are looking for something specific or a category of crate you may want to checkout lib.rs, a great alternative frontend for crates.

This is the first time I'm hearing about blessed.rs, thank you. Havent used lib.rs since they closed the source. Thanks for sharing.

I wrote a string case conversion library called "convert-case". Most people use "heck". They use different patterns, and different naming conventions. I prefer mine (more features), but it also came out long after heck became standard.

I like isolang for handling ISO-639 language codes.

(Disclaimer: I contributed to it)

In the async world, I love the smol stack where each functionality is split into an independent subcrate.

Here is an originally random list (using cargo tree --prefix=depth) with some very loose logical grouping. Wide-scoped and well-known crates removed (some remaining are probably still known by most).

mime data-encoding percent-encoding textwrap unescape unicode-width scraper
arrayvec bimap bstr enum-iterator os_str_bytes pretty_assertions paste
clap_complete console indicatif shlex
lz4_flex mpeg2ts roxmltree speedy
aes base64 hex cbc sha1 sha2 rsa
reverse_geocoder trust-dns-resolver
signal-hook signal-hook-tokio
blocking
fs2
semver
snmalloc-rs

I love chumsky and iced. They are still under heavy feature development, but their design feels great.