I'd like to self-nominate dlhn.
dlhn is a new serialization format, implemented using serde.
We have just released a version that fixes a critical bug.
I would like to make this known to those who are using the pre-fixed version.
I'd like to self-nominate dlhn.
HyperQueue, a runtime for ergonomic execution of programs on a distributed cluster, with built-in support for HPC systems.
I'd like to self-nominate
emojis, which is a little crate to lookup emojis by Unicode value or gemoji shortcode. You can also iterate over all emojis, iterate over emoji skin tones, and emojis within a group.
I'd like to nominate Screen 13, an easy-to-use 3D rendering library. It provides a render graph for Vulkan compute, graphic, and ray-trace shader pipelines. Includes documentation and examples!
I’d like to nominate humantime — Rust date/time library // Lib.rs, a zero-dependency crate to convert between SystemTime and UTC date/time without making any syscalls. If you only need UTC, this is a much smaller and simpler alternative than either
I'd like to self-nominate edres - named because it's
serde backwards. It takes markup files (json/toml/yaml) and generates compatible Rust structs and enums based on them.
Essentially it lets the config files be the source of truth for the data structure, while still being compile-time type-safe when you deserialize and use them. So when the schema changes, you don't have to update your code at all unless you've made a breaking change - which will fail compilation instead of breaking at runtime.
I recently overhauled the documentation, testing, and configuration - so I think it's in a nice state to show off~
I'd like to nominate soa-derive, for making the Array of Structs to Struct of Arrays transformation far more accessible and less mentally taxing, with (often significant) performance and storage improvements!
It has a lovely benchmark, and a pretty nice API.
I'd like to nominate type_description - a crate for machine-readable type descriptions that can be used for self-documenting configuration types.
I'd like to nominate enum_delegate - a crate that, in essence, helps you implement a trait on many types across crate boundaries, while avoiding the runtime cost of dynamic dispatch. This is especially useful for digital/audio signal processing with graphs, where nodes with different processes need to be executed under tight timing constraints.
Remember that some of us are newbies, and any pointers to crates we should prioritize learning about are helpful, even if they're obvious to experienced users.
The problem with an open library is that there may be multiple solutions for a given problem. I haven't looked yet -- do we have a mechanism for "yes, this is the current best of breed in this area?"
Hurl allows you to test and run HTTP requests with a plain text format and libcurl. Hurl can be used for fetching data, testing HTTP sessions and testing XML / JSON APIs. It is available on Linux, Mac and Windows. We use it intensively to test our Web site and APIs in a CI/CD pipeline.
I'd like to nominate
It allows you to straightforwardly write trait bounds for a generic function that accepts an
async fn, without needing to introduce a second type parameter for the future. And that's all it does, defining
AsyncFn* and blanket trait impls — no macros you need to use, or any other complication.
I'd like to self-nominate lngcnv which is a command-line linguistic tool. After 6 months of work, today, I released v1.6.0. The source code has been almost completely rewritten but the program is fully backwards compatible with v1.5.1 and earlier versions released earlier this year.
lngcnv (1) displays the exact phonetic (not just phonemic) pronunciation of a phrase in the International Phonetic Alphabet (IPA), (2) translates between different (e.g., dialectal) varieties of a language, and (3) converts between different (e.g., regional) norms of orthography.
A word or some piece of text can be provided either (a) as an argument directly from the command line or (b) in a file, such as a hundreds of pages long book. [(c) a REPL mode is planned for v1.7.0 ]
Six modern and ancient languages are currently supported, often including several dialects:
Polish (three dialects)
Spanish (five dialects)
Tikuna (five dialects)
This includes two languages spoken by Native Americans, which I'm studying, one from the Amazon basin and one from the Andes.
I partially self-nominate graph: A library that provides a collection of high-performant graph algorithms and an API for building your own algorithms.
It is build to support parallel algorithms on large-scale graphs with billions of nodes and edges. In addition to the Rust library, it also offers a Python extension and an Arrow Flight based RPC interface.
We just released a new version that can be used on stable, thanks to the stabilization of GATs.
The library is written by a friend of mine and myself.
I listen to freeCodeCamp's Code Radio every time I want to focus when coding. It's a network music radio that plays music designed for coding, 24/7.
When I started learning Rust, I thought the fastest way to learn it is to port some programs I wrote before to Rust. I wrote a Code Radio GUI client before, so I ported it to Rust as a cool console program.
Now when I want some music while coding, I just run
code-radio in my IDE's terminal.
Nice work, @JasonWei512 ! This is now one of my favourite crates!
I'd like to self-nominate file-format: a crate for determining binary-based file formats.
This crate mainly uses the magic number to detect the file formats.
I'd like to self-nominate solomondb which is an embedded Gremlin-compatible graph database that supports multiple storage engine, currently RocksDB and Redb. The current version of SolomonDB allows you to work wit RocksDB or Redb using Gremlin query language. Hence, applying graph data structure to manage key-value pairs in those embedded storage layer.
SolomonDB embedded storage is easy to set up. You only need to identify the path where your database will be located and it's good to go.
use solomondb::Datastore; let datastore = Datastore::new(path); let db = Database::new(datastore.borrow());
SolomonDB supports GQL (Gremlin Query Language) and it does not require Gremlin Server or Apache TinkerPop to operate. You can query data on top of embedded storages. Below are examples on how to create vertices, properties and traverse to retrieve data.
// Create two new vertices with properties let t1 = db .traverse() .v(1) .add_v("person") .property("github", "tin-snowflake") .property("name", "Tin Chung") .property("age", 21) .add_v("coder") .property("github", "chungquantin") .property("age", 30); // Traverse vertices which have property "github" and label "person" let t2 = t1.clone().has_key("github").has_label("person").exec().to_list().await.unwrap(); // Traverse vertices which have property "github" let t3 = t1.clone().has_key("github").exec().to_list().await.unwrap(); // Traverse vertices which does not have property "name" let t4 = t1.clone().has_not("name").exec().next().await.unwrap();
The road map of SolomonDB will include:
- Support more storage engines, help to graph modelize the embedded storage
- Add protobuf and enable single-node mode.
The mission of SolomonDB is to make graph traversal become easier, aim at plug-and-play database that can run directly in client app.
I'd like to self-nominate dcp. dcp is a container utility tool that enables you to easily extract data from a container filesystem. It has extensive documentation and has an MIT license. It is available to download via pre-built binaries and crates.io.
I'd like to nominate RustTyC, which I discovered a couple years ago (but only sent a PR to recently). It's a library for general, lattice-based type checking and inference, and it makes writing (correct) type checkers much easier.