Mirrors/dioxus
2
0
Fork 0
mirror of https://github.com/DioxusLabs/dioxus synced 2024-11-10 06:34:20 +00:00
Code Issues Projects Releases Packages Wiki Activity
dioxus/README.md

163 lines
9.5 KiB
Markdown
Raw Normal View History

Chore: update readme
2021-01-16 15:31:17 +00:00
<div align="center">
<h1>🌗🚀 Dioxus</h1>
<p>
Feat: update docs a bit
2021-03-02 06:47:27 +00:00
<strong>Frontend that scales.</strong>
Chore: update readme
2021-01-16 15:31:17 +00:00
</p>
</div>
Feat: - integrate subscription service into context. - Update documentation
2021-02-08 21:57:34 +00:00
Dioxus is a portable, performant, and ergonomic framework for building cross-platform user experiences in Rust.
Feat: more docs, dissolve vnode crate into dioxus-core
2021-01-16 06:30:48 +00:00
Feat: - integrate subscription service into context. - Update documentation
2021-02-08 21:57:34 +00:00
```rust
feat: move webview to wry
2021-07-08 16:01:31 +00:00
fn App(cx: Context<()>) -> VNode {
let mut count = use_state(cx, || 0);
Feat: move webview logic into library
2021-02-25 23:44:00 +00:00
docs: move around examples
2021-06-17 22:00:32 +00:00
cx.render(rsx! {
wip: more examples
2021-07-11 23:31:07 +00:00
h1 { "High-Five counter: {count}" }
feat: move webview to wry
2021-07-08 16:01:31 +00:00
button { onclick: move |_| count += 1, "Up high!" }
button { onclick: move |_| count -= 1, "Down low!" }
Feat: view -> render
2021-03-01 02:21:17 +00:00
})
Feat: - integrate subscription service into context. - Update documentation
2021-02-08 21:57:34 +00:00
};
```
docs: move around examples
2021-06-17 22:00:32 +00:00
Dioxus can be used to deliver webapps, desktop apps, static pages, liveview apps, eventually mobile apps (WIP), and more. At its core, Dioxus is entirely renderer agnostic and has great documentation for creating new renderers for any platform.
Feat: major overhaul to diffing, using a "diffing machine" now
2021-02-14 23:03:16 +00:00
Feat: massive changes to definition of components This change switches back to the original `ctx<props>` syntax for commponents. This lets lifetime elision to remove the need to match exactly which lifetime (props or ctx) gets carried to the output. As such, `Props` is currently required to be static. It *is* possible to loosen this restriction, and will be done in the future, though only through adding metadata about the props through the Props derive macro. Implementing the IS_STATIC trait is unsafe, so the derive macro will do it through some heuristics. For now, this unlocks sharing vnodes from parents to children, enabling pass-thru components, fragments, portals, etc.
2021-06-01 22:33:15 +00:00
If you know React, then you already know Dioxus.
Feat: update readme and examples
2021-03-01 05:16:48 +00:00
### **Things you'll love ❤️:**
Feat: massive changes to definition of components This change switches back to the original `ctx<props>` syntax for commponents. This lets lifetime elision to remove the need to match exactly which lifetime (props or ctx) gets carried to the output. As such, `Props` is currently required to be static. It *is* possible to loosen this restriction, and will be done in the future, though only through adding metadata about the props through the Props derive macro. Implementing the IS_STATIC trait is unsafe, so the derive macro will do it through some heuristics. For now, this unlocks sharing vnodes from parents to children, enabling pass-thru components, fragments, portals, etc.
2021-06-01 22:33:15 +00:00
wip: more on signals
2021-06-30 21:20:13 +00:00
- Ergonomic design
- Minimal boilerplate
Feat: massive changes to definition of components This change switches back to the original `ctx<props>` syntax for commponents. This lets lifetime elision to remove the need to match exactly which lifetime (props or ctx) gets carried to the output. As such, `Props` is currently required to be static. It *is* possible to loosen this restriction, and will be done in the future, though only through adding metadata about the props through the Props derive macro. Implementing the IS_STATIC trait is unsafe, so the derive macro will do it through some heuristics. For now, this unlocks sharing vnodes from parents to children, enabling pass-thru components, fragments, portals, etc.
2021-06-01 22:33:15 +00:00
- Simple build, test, and deploy
wip: more on signals
2021-06-30 21:20:13 +00:00
- Compile-time correct templating
- Support for fine-grained reactivity
- Support for html! and rsx! templates
Feat: update readme and examples
2021-03-01 05:16:48 +00:00
- SSR, WASM, desktop, and mobile support
wip: more on signals
2021-06-30 21:20:13 +00:00
- Support for asynchronous batched rendering
- Powerful and simple integrated state management
- Rust! (enums, static types, modules, efficiency)
wip: ....sigh..... so the diffing algorithm is robust but it's still not finished. We need to re-enable the fancier keyed diffing versions some point (soon!).
2021-07-12 07:58:46 +00:00
### Unique features:
wip: workshop the readme
2021-07-12 16:19:50 +00:00
- Incredible inline documentation. Supports hover and guides for all HTML elements, listeners, and events.
- Templates are "constified" at compile time. Nodes that don't change will won't be diffed.
- Custom bump-allocator backing for all components. Nearly 0 allocations for steady-state components.
wip: ....sigh..... so the diffing algorithm is robust but it's still not finished. We need to re-enable the fancier keyed diffing versions some point (soon!).
2021-07-12 07:58:46 +00:00
- Starting a new app takes zero templates or special tools - get a new app running in just seconds.
- Desktop apps running natively (no Electron!) in less than 10 lines of code.
- The most ergonomic and powerful state management of any Rust UI toolkit.
wip: workshop the readme
2021-07-12 16:19:50 +00:00
- And more! Read the full release post here.
wip: ....sigh..... so the diffing algorithm is robust but it's still not finished. We need to re-enable the fancier keyed diffing versions some point (soon!).
2021-07-12 07:58:46 +00:00
Feat: update readme and examples
2021-03-01 05:16:48 +00:00
## Get Started with...
Feat: massive changes to definition of components This change switches back to the original `ctx<props>` syntax for commponents. This lets lifetime elision to remove the need to match exactly which lifetime (props or ctx) gets carried to the output. As such, `Props` is currently required to be static. It *is* possible to loosen this restriction, and will be done in the future, though only through adding metadata about the props through the Props derive macro. Implementing the IS_STATIC trait is unsafe, so the derive macro will do it through some heuristics. For now, this unlocks sharing vnodes from parents to children, enabling pass-thru components, fragments, portals, etc.
2021-06-01 22:33:15 +00:00
wip: more
2021-02-08 22:09:21 +00:00
<table style="width:100%" align="center">
<tr >
wip: todomvc
2021-03-29 16:31:47 +00:00
<th><a href="http://github.com/jkelleyrtp/dioxus">Web</a></th>
Chore: fix table again
2021-02-08 22:08:26 +00:00
<th><a href="http://github.com/jkelleyrtp/dioxus">Desktop</a></th>
<th><a href="http://github.com/jkelleyrtp/dioxus">Mobile</a></th>
wip: more work on updating syntad
2021-06-15 14:02:46 +00:00
<th><a href="http://github.com/jkelleyrtp/dioxus">State</a></th>
Chore: fix table again
2021-02-08 22:08:26 +00:00
<th><a href="http://github.com/jkelleyrtp/dioxus">Docs</a></th>
wip: more
2021-02-08 22:09:21 +00:00
<th><a href="http://github.com/jkelleyrtp/dioxus">Tools</a></th>
Feat: add table to docs
2021-02-08 22:05:58 +00:00
<tr>
</table>
Chore: fix table
2021-02-08 22:07:27 +00:00
Feat: shorten readme
2021-02-16 06:41:41 +00:00
## Explore
Feat: massive changes to definition of components This change switches back to the original `ctx<props>` syntax for commponents. This lets lifetime elision to remove the need to match exactly which lifetime (props or ctx) gets carried to the output. As such, `Props` is currently required to be static. It *is* possible to loosen this restriction, and will be done in the future, though only through adding metadata about the props through the Props derive macro. Implementing the IS_STATIC trait is unsafe, so the derive macro will do it through some heuristics. For now, this unlocks sharing vnodes from parents to children, enabling pass-thru components, fragments, portals, etc.
2021-06-01 22:33:15 +00:00
concept: signals guide
2021-06-30 20:47:30 +00:00
- [**Fine-grained reactivity**: Skip the diff overhead with signals ](docs/guides/00-index.md)
Feat: update readme and examples
2021-03-01 05:16:48 +00:00
- [**HTML Templates**: Drop in existing HTML5 templates with html! macro](docs/guides/00-index.md)
- [**RSX Templates**: Clean component design with rsx! macro](docs/guides/00-index.md)
Feat: wire up explore in readme
2021-02-16 06:43:50 +00:00
- [**Running the examples**: Explore the vast collection of samples, tutorials, and demos](docs/guides/00-index.md)
- [**Building applications**: Use the Dioxus CLI to build and bundle apps for various platforms](docs/guides/01-ssr.md)
- [**Liveview**: Build custom liveview components that simplify datafetching on all platforms](docs/guides/01-ssr.md)
- [**State management**: Easily add powerful state management that comes integrated with Dioxus Core](docs/guides/01-ssr.md)
- [**Concurrency**: Drop in async where it fits and suspend components until new data is ready](docs/guides/01-ssr.md)
Feat: update readme and examples
2021-03-01 05:16:48 +00:00
- [**1st party hooks**: Cross-platform router hook](docs/guides/01-ssr.md)
Feat: wire up explore in readme
2021-02-16 06:43:50 +00:00
- [**Community hooks**: 3D renderers](docs/guides/01-ssr.md)
Feat: updates to docs, extension
2021-01-29 16:57:52 +00:00
Feat: buff the readme and docs
2021-05-20 00:57:19 +00:00
## Blog Posts
Feat: massive changes to definition of components This change switches back to the original `ctx<props>` syntax for commponents. This lets lifetime elision to remove the need to match exactly which lifetime (props or ctx) gets carried to the output. As such, `Props` is currently required to be static. It *is* possible to loosen this restriction, and will be done in the future, though only through adding metadata about the props through the Props derive macro. Implementing the IS_STATIC trait is unsafe, so the derive macro will do it through some heuristics. For now, this unlocks sharing vnodes from parents to children, enabling pass-thru components, fragments, portals, etc.
2021-06-01 22:33:15 +00:00
Feat: buff the readme and docs
2021-05-20 00:57:19 +00:00
- [Why we need a stronger typed web]()
- [Isomorphic webapps in 10 minutes]()
- [Rust is high level too]()
- [Eliminating crashes with Rust webapps]()
- [Tailwind for Dioxus]()
- [The monoglot startup]()
Feat: massive changes to definition of components This change switches back to the original `ctx<props>` syntax for commponents. This lets lifetime elision to remove the need to match exactly which lifetime (props or ctx) gets carried to the output. As such, `Props` is currently required to be static. It *is* possible to loosen this restriction, and will be done in the future, though only through adding metadata about the props through the Props derive macro. Implementing the IS_STATIC trait is unsafe, so the derive macro will do it through some heuristics. For now, this unlocks sharing vnodes from parents to children, enabling pass-thru components, fragments, portals, etc.
2021-06-01 22:33:15 +00:00
docs: move into a fromjs tutorial
2021-06-03 16:02:46 +00:00
## Why?
Feat: massive changes to definition of components This change switches back to the original `ctx<props>` syntax for commponents. This lets lifetime elision to remove the need to match exactly which lifetime (props or ctx) gets carried to the output. As such, `Props` is currently required to be static. It *is* possible to loosen this restriction, and will be done in the future, though only through adding metadata about the props through the Props derive macro. Implementing the IS_STATIC trait is unsafe, so the derive macro will do it through some heuristics. For now, this unlocks sharing vnodes from parents to children, enabling pass-thru components, fragments, portals, etc.
2021-06-01 22:33:15 +00:00
TypeScript is a great addition to JavaScript, but comes with a lot of tweaking flags, a slight performance hit, and an uneven ecosystem where some of the most important packages are not properly typed. TypeScript provides a lot of great benefits to JS projects, but comes with its own "tax" that can slow down dev teams. Rust can be seen as a step up from TypeScript, supporting:
- static types for _all_ libraries
- advanced pattern matching
- immutability by default
- clean, composable iterators
- a good module system
- integrated documentation
- inline built-in unit/integration testing
- best-in-class error handling
wip
2021-07-01 18:14:59 +00:00
- simple and fast build system (compared to webpack!)
Wip: pre-diffmachine merge fork
2021-06-03 17:57:41 +00:00
- powerful standard library (no need for lodash or underscore)
Feat: massive changes to definition of components This change switches back to the original `ctx<props>` syntax for commponents. This lets lifetime elision to remove the need to match exactly which lifetime (props or ctx) gets carried to the output. As such, `Props` is currently required to be static. It *is* possible to loosen this restriction, and will be done in the future, though only through adding metadata about the props through the Props derive macro. Implementing the IS_STATIC trait is unsafe, so the derive macro will do it through some heuristics. For now, this unlocks sharing vnodes from parents to children, enabling pass-thru components, fragments, portals, etc.
2021-06-01 22:33:15 +00:00
- include_str! for integrating html/css/svg templates directly
docs: move into a fromjs tutorial
2021-06-03 16:02:46 +00:00
- various macros (`html!`, `rsx!`) for fast template iteration
Feat: massive changes to definition of components This change switches back to the original `ctx<props>` syntax for commponents. This lets lifetime elision to remove the need to match exactly which lifetime (props or ctx) gets carried to the output. As such, `Props` is currently required to be static. It *is* possible to loosen this restriction, and will be done in the future, though only through adding metadata about the props through the Props derive macro. Implementing the IS_STATIC trait is unsafe, so the derive macro will do it through some heuristics. For now, this unlocks sharing vnodes from parents to children, enabling pass-thru components, fragments, portals, etc.
2021-06-01 22:33:15 +00:00
And much more. Dioxus makes Rust apps just as fast to write as React apps, but affords more robustness, giving your frontend team greater confidence in making big changes in shorter time. Dioxus also works on the server, on the web, on mobile, on desktop - and it runs completely natively so performance is never an issue.
Docs: more examples and docs
2021-06-24 15:09:38 +00:00
# Parity with React
docs: more work
2021-06-24 15:18:58 +00:00
Dioxus is heavily inspired by React, but we want your transition to feel like an upgrade. Dioxus is _most_ of the way there, but missing a few key features. This parity table does not necessarily include important ecosystem crates like code blocks, markdown, resizing hooks, etc.
Docs: more examples and docs
2021-06-24 15:09:38 +00:00
docs: split table up
2021-06-24 15:17:59 +00:00
### Phase 1: The Basics
wip: more examples
2021-07-11 23:31:07 +00:00
| Feature | Dioxus | React | Notes for Dioxus |
| ----------------------- | ------ | ----- | ----------------------------------------------------------- |
| Conditional Rendering | ✅ | ✅ | if/then to hide/show component |
| Map, Iterator | ✅ | ✅ | map/filter/reduce to produce rsx! |
| Keyed Components | ✅ | ✅ | advanced diffing with keys |
| Web | ✅ | ✅ | renderer for web browser |
| Desktop (webview) | ✅ | ✅ | renderer for desktop |
| Shared State (Context) | ✅ | ✅ | share state through the tree |
| Hooks | ✅ | ✅ | memory cells in components |
| SSR | ✅ | ✅ | render directly to string |
| Component Children | ✅ | ✅ | cx.children() as a list of nodes |
| Headless components | ✅ | ✅ | components that don't return real elements |
| Fragments | ✅ | ✅ | multiple elements without a real root |
| Manual Props | ✅ | ✅ | Manually pass in props with spread syntax |
| Controlled Inputs | ✅ | ✅ | stateful wrappers around inputs |
| CSS/Inline Styles | ✅ | ✅ | syntax for inline styles/attribute groups |
| Custom elements | ✅ | ✅ | Define new element primitives |
| Suspense | 🛠 | ✅ | schedule future render from future/promise |
| Cooperative Scheduling | 🛠 | ✅ | Prioritize important events over non-important events |
| Fine-grained reactivity | 🛠 | ❓ | Skip diffing for fine-grain updates |
| Compile-time correct | ✅ | ❓ | Throw errors on invalid template layouts |
| Runs natively | ✅ | ❓ | runs as a portable binary w/o a runtime (Node) |
| 1st class global state | ✅ | ❓ | redux/recoil/mobx on top of context |
| Subtree Memoization | ✅ | ❓ | skip diffing static element subtrees |
| Heuristic Engine | 🛠 | ❓ | track component memory usage to minimize future allocations |
| NodeRef | 🛠 | ✅ | gain direct access to nodes [1] |
docs: more work on docs
2021-06-25 13:31:13 +00:00
docs: more readme
2021-06-25 13:35:01 +00:00
- [1] Currently blocked until we figure out a cross-platform way of exposing an imperative Node API.
docs: split table up
2021-06-24 15:17:59 +00:00
### Phase 2: Advanced Toolkits
docs: move suspense into a "figured out task"
2021-06-25 17:17:18 +00:00
| Feature | Dioxus | React | Notes for Dioxus |
| --------------------- | ------ | ----- | ---------------------------------- |
feat: add edits back! and more webview support! This commit adds a new type - the DomEdit - for serializing the changes made by the diffing machine. The architecture of how DomEdits fit into the cooperative scheduling is still TBD but it will allow us to build change lists without applying them immediately. This is more performant and allows us to only render parts of the page at a time. This commit also adds more infrastructure around webview. Dioxus can now run on the web, generate static pages, run in the desktop, and run on mobile, with a large part of thanks to webview.
2021-07-05 22:37:15 +00:00
| 1st class router | 👀 | ✅ | Hook built on top of history |
| Assets | 👀 | ✅ | include css/svg/img url statically |
| Integrated classnames | 🛠 | ❓ | built-in `classnames` |
| Transition | 👀 | 🛠 | High-level control over suspense |
| Animation | 👀 | ✅ | Spring-style animations |
| Mobile | 👀 | ✅ | Render with cacao |
| Desktop (native) | 👀 | ✅ | Render with native desktop |
| 3D Renderer | 👀 | ✅ | react-three-fiber |
docs: split table up
2021-06-24 15:17:59 +00:00
### Phase 3: Additional Complexity
| Feature | Dioxus | React | Notes for Dioxus |
| -------------------- | ------ | ----- | ------------------------------------ |
feat: add edits back! and more webview support! This commit adds a new type - the DomEdit - for serializing the changes made by the diffing machine. The architecture of how DomEdits fit into the cooperative scheduling is still TBD but it will allow us to build change lists without applying them immediately. This is more performant and allows us to only render parts of the page at a time. This commit also adds more infrastructure around webview. Dioxus can now run on the web, generate static pages, run in the desktop, and run on mobile, with a large part of thanks to webview.
2021-07-05 22:37:15 +00:00
| Portal | ❓ | ✅ | cast elements through tree |
| Error/Panic boundary | ❓ | ✅ | catch panics and display custom BSOD |
| Code-splitting | 👀 | ✅ | Make bundle smaller/lazy |
| LiveView | 👀 | ❓ | Example for SSR + WASM apps |
docs: add a table to the readme
2021-06-24 15:15:25 +00:00
docs: fix formatting
2021-06-24 15:15:55 +00:00
- ✅ = implemented and working
- 🛠 = actively being worked on
docs: more readme
2021-06-25 13:33:59 +00:00
- 👀 = not yet implemented or being worked on
docs: fix formatting
2021-06-24 15:15:55 +00:00
- ❓ = not sure if will or can implement
Reference in a new issue Copy permalink