React Native vs Flutter. What to Choose for Your Next Project

we plan to migrate to Angular 1, and we'll finish out the decade on React

The alternative is uncontrolled components, where form data is handled by the DOM itself.

PopupState HelperTo make managing the popover state easier, we can use the material-ui-popup-state library to manage the state.

React abstracts the DOM with functionally pure declarative rendering and provides escape hatches back to mutable imperative DOM land. This is a profound philosophical difference that Rich gave a talk about.

It's true that Svelte does not allow you to map over children like React, but its slot API and <svelte:component> provide similarly powerful composition. You can pass component constructors as props and instantiate them with <svelte:component>, and use slots and their let bindings for higher order composition. It sounds like you're thinking in virtual DOM idioms instead of Svelte's.

My frustration is mainly from Svelte's choices that are very un-JavaScript-like. It doesn't have to be "like React/Vue". React is React because it doesn't restrict what you can do with JavaScript for the most part. It's just common FP practice to fold/map.

For use$ since svelte is never going to support actions for components, i designed something that reminds React hooks that will in some ways replace this feature.

Another difference is that context in Svelte does not insert anything into the visual component tree. There is no <Context.Provider> element like in React

In Svelte, all reactive statements are memoized. Instead of const var = useMemo(() => expression, dependencies), you can use $: var = expression. Notice with Svelte, you don't need to declare the dependencies. The compiler infers them for you.

Although Capacitor is developed by Ionic, you can use it in combination with any framework and UI library you want. In fact, Capacitor itself promotes using it with whatever framework you want.

Frontend frameworks are a positive sum game! Svelte has no monopoly on the compiler paradigm either. Just like I think React is worth learning for the mental model it imparts, where UI is a (pure) function of state, I think the frontend framework-as-compiler paradigm is worth understanding. We're going to see a lot more of it because the tradeoffs are fantastic, to where it'll be a boring talking point before we know it.

I use a mutator and use it's changeValue function to 'change' the value of the relevant field (I supply the same value). This in turn notifies all relevant parties of the change to the form's state, and a validation is triggered.

Some user experience issue with the proposed solution: This won't show up the inline field error message if you just hit on submit and other fields already contained errors since the code on the onSubmit won't be hit

This is for a time picker. If you're picking times for today, you may pick a time that is 15 minutes from now. It's valid now because it's currently in the future. If you don't touch the form for the next 20 minutes then click submit, the submission should be prevented because your selected time is now 5 minutes in the past.

Fonk is framework extension, and can be easily plugged into many libraries / frameworks, in this documentation you will find integrations with:

React Final Forms is a great library, an enhanced version of Redux Form

Form validation can get complex (synchronous validations, asynchronous validations, record validations, field validations, internationalization, schemas definitions...). To cope with these challenges we will leverage this into Fonk and Fonk Final Form adaptor for a React Final Form seamless integration.

Sometimes you might need to use an element other than a native <button>. useButton supports this via the elementType prop. When used with an element other than a native button, useButton automatically applies the necessary ARIA roles and attributes to ensure that the element is exposed to assistive technology as a button.

const field = state.fields[name] if (field) { field.data = { ...field.data, ...data } }

suite of over 30 separate packages of React visualization primitives that fall into several categories (Figure 2). It is un-opinionated on state management, animation, and styling so it can integrate with any React codebase, and its emphasis on modularity (similar to D3) lets you keep your bundle sizes down by only using the packages you need to create your reusable chart library or a custom one-off chart.

In our research (Figure 1), we found that existing React visualization libraries are often high-level abstractions and optimized for ease of use (i.e., fewer lines of code) at the expense of expressivity. None offer the expressivity of D3 primitives and many don’t allow for the optimization we want in production because computation, animations, state management, styles, and rendering are all encapsulated.

because D3 and React both want to own DOM manipulation, we’ve found that it’s best to only use D3 for the math and React for the DOM because two mental models for updating the DOM opens the door for bugs to sneak in. However, using D3 solely for math means a significant amount of its (DOM-based) functionality is not available for use: selection.join, zoom, drag, brush, and transitions. Additionally, as mentioned above, D3 has its own learning curve and we would like developers to feel like they are writing native React code with standard APIs and familiar patterns.

Leveraging React and its ecosystem would provide learnability and performance, and a low-level, modular API would deliver expressivity.

You should not create a new debounce function on every render with: return new Promise(resolve => { debounce(() => resolve(this.getIsNameUnique(name)), 2000); }); Instead you should just wrap your whole function isNameUnique with the debounce (see my sandbox). By creating a new debounce function on every hit, it cannot 'remember' that is was called or that is will be called again. This will prevent the debouncing.

FormValidation can be used with popular JavaScript frameworks such as React, Preact, Vue, Svelte, etc.

mixing the turing complete of javascript with the markup of HTML eliminates the readability of JSX so that it is actually harder to parse than a solution like hyperscript

I'm personally open to any other solutions, especially in how we might be able to untie from React, but, for now, inline-XML markup (JSX) is OK with me.

I've recently started playing with segmentio/deku, an alternative to React, and I'm also using Babel to transpile my js and jsx code.

Hyperscript syntax for React.js markup

In React 0.12 time frame we did a bunch of small changes to how key, ref and defaultProps works. Particularly, they get resolved early on in the React.createElement(...) call. This made sense when everything was classes, but since then, we've introduced function components. Hooks have also make function components more prevalent. It might be time to reevaluate some of those designs to simplify things (at least for function components).

In the next major, we'll start copying the ref onto both the props and the element.ref. React will now use the props.ref as the source of truth for forwardRef and classes and it will still create a shallow copy of props that excludes the ref in these cases. At the same time, we'll add a getter for element.ref in DEV that warns if you access it. The upgrade path is now to just access it off props if you need it from the element.

This proposal simplifies how React.createElement works and ultimately lets us remove the need for forwardRef.

For event listeners we support the standard jsx naming convention onEventname (this is converted to on:eventname in svelte) as well.

Facebook’s React has an optional language extension that enables you to embed HTML inside JavaScript. This extension can make your code more concise, but it also breaks compatibility with the rest of the JavaScript ecosystem. ECMAScript 6 will have template strings [1], which enable you to implement JSX (or something close to it) inside the language.

hyperx(React.createElement)

Since “virtual DOM” is more of a pattern than a specific technology, people sometimes say it to mean different things. In React world, the term “virtual DOM” is usually associated with React elements since they are the objects representing the user interface

Deku is a library for rendering interfaces using pure functions and virtual DOM.

Then at some moment I just stumbled upon limitations and inexpressiveness of templates and started to use JSX everywhere — and because JSX was not a typical thing for Vue I switched to React over time. I don’t want to make a step back.

There is a killer feature of vDOM that Svelte has nothing to replace with. It is the ability to treat component hierarchy as an object.

Typically, unified compilers return string. This compiler returns a ReactElement.

To fix our Svelte version you might think we could use beforeUpdate or afterUpdate, but these lifecycle functions are related to the DOM being updated, not to prop updates. We only want to rerun our fetching when the album prop is changed.

Context can only store a single value, not an indefinite set of values each with its own consumers.

We want to improve this while keeping both the API and the semantics and behavior as Reactish as possible.

Recoil defines a directed graph orthogonal to but also intrinsic and attached to your React tree.

Component state can only be shared by pushing it up to the common ancestor, but this might include a huge tree that then needs to re-render.

For reasons of compatibility and simplicity, it's best to use React's built-in state management capabilities rather than external global state.

I think Svelte has some things to learn from React hooks.

While react hooks were one of the catalysts for v3 we don't agree with with the APIs or the model and won't be emulating it.

Hooks embrace JavaScript closures and avoid introducing React-specific APIs where JavaScript already provides a solution.

You want to write maintainable tests that give you high confidence that your components are working for your users. As a part of this goal, you want your tests to avoid including implementation details so refactors of your components (changes to implementation but not functionality) don't break your tests and slow you and your team down.

we are using RTL's findBy search variant to wait for element(s) which appear eventually.

For any element that isn't there yet but will be there eventually, use findBy over getBy or queryBy. If you assert for a missing element, use queryBy. Otherwise default to getBy

The neat thing about getByRole: it shows all the selectable roles if you provide a role that isn't available in the rendered component's HTML:

// recommended

Whereas the describe-block is the test suite, the test-block (which also can be named it instead of test) is the test case. A test suite can have multiple test cases and a test case doesn't have to be in a test suite. What you put into the test cases are called assertions (e.g. expect in Jest) which either turn out to be successful (green) or erroneous (red). Here we have two assertions which should turn out successful:

In general it is recommended you handle forms in this "controlled" manner. In some cases it might make sense to manage the form state outside of Solid via refs. These "uncontrolled" forms can also work. Just be conscious of the difference as mixing approaches can lead to unexpected results.

So while Solid's JSX and might resemble React it by no means works like React and there should be no illusions that a JSX library will just work with Solid. Afterall, there are no JSX libraries, as they all work without JSX, only HyperScript or React ones.

Remember even though the syntax is almost identical, there are significant differences between how Solid's JSX works and a library like React.

Solid supports most React features like Fragments, Portals, Context, Suspense, Error Boundaries, Lazy Components, Async Rendering, Implicit Event Delegation, SSR

I used them a bit and realized almost immediately they felt like a leaky abstraction.

There's one downside to Reacts reactivity model - the hooks (useState and useEffect) have to always be called in the same order and you can't put them inside an if block.

It's possible to run a function whenever some reactive state changes using the useEffect hook. In the example we log the length of the todoList whenever it changes. The first argument to useEffect is the function we want to run, and the second is a list of reactive values to track - whenever one of these values changes the effect will run again.

MobX - for me personally MobX is a far better way to manage state than React Hooks. It doesn't care about the UI layer so it can be used outside the React ecosystem, and it's simple to mutate data.

derived values are simple to declare

Solid is a declarative JavaScript library for creating user interfaces. It's kinda like if React and Svelte had a baby.

This library exports a single React Hook, useMethods, which has all the power of useReducer but none of the ceremony that comes with actions and dispatchers.

Keep in mind that the values in meta are dependent on you having subscribed to them with the subscription prop.

useField() returns FieldRenderProps. It will manage the rerendering of any component you use it in, i.e. the component will only rerender if the field state subscribed to via useField() changes.

for example, reactive declarations essentially do the work of React's useMemo, useCallback and useEffect without the boilerplate (or indeed the garbage collection overhead of creating inline functions and arrays on each state change).

urql is a GraphQL client that exposes a set of helpers for several frameworks.

One package to get a working GraphQL client in React, Preact, and Svelte

This is the same as useEffect in React, incidentally — the function must be synchronous in order to avoid race conditions.

I think Svelte's approach where it replaces component instances with the component markup is vastly superior to Angular and the other frameworks. It gives the developer more control over what the DOM structure looks like at runtime—which means better performance and fewer CSS headaches, and also allows the developer to create very powerful recursive components.

Since one of Svelte's main principles is reacting only to data changes (as opposed to re-render the whole app with the full state all the time)

Re Object.keys(undefined), I think I'm ok with that failing. AFAIK it would also fail in React

The more I think about this, the more I think that maybe React already has the right solution to this particular issue, and we're tying ourselves in knots trying to avoid unnecessary re-rendering. Basically, this JSX... <Foo {...a} b={1} {...c} d={2}/> ...translates to this JS: React.createElement(Foo, _extends({}, a, { b: 1 }, c, { d: 2 })); If we did the same thing (i.e. bail out of the optimisation allowed by knowing the attribute names ahead of time), our lives would get a lot simpler, and the performance characteristics would be pretty similar in all but somewhat contrived scenarios, I think. (It'll still be faster than React, anyway!)

tip no 8. API should be explicit!

This is similar to the :focus-visible pseudo class in CSS.

In general, I'm unpersuaded by these arguments (learning curve is determined not just by unfamiliar syntax, but by unfamiliar semantics and APIs as well, and the frameworks in question excel at adding complexity in those areas).

Like with React, you can pass in callback props like onSave and onDelete, which is the main way you send data out of a component to a parent.

$: declares a relationship, where remaining should always be 10 - count. This might seem weird, but it’s conceptually the same as declaring a variable that you know will be re-calculated in every React render loop — except that remaining will get recalculated only when count changes.

My solution idea is just to let the compiler knows that rest is of type T, using a custom type guard function, but you could use other approaches, like a type casting: <Child {...((rest as unknown) as T)} />

With useEffect, you can handle lifecycle events directly inside function components. Namely, three of them: componentDidMount, componentDidUpdate, and componentWillUnmount. All with one function!