Summary:
With a multiline TextInput, the text is initially rendered correctly but once you try to edit it, the cursor slides off the top of the view and out of its bounds. Also if the TextInput is scrollable, you can scroll the text out of the bounds of the view, which looks buggy unless you clip the overflow. This occurs because the top content inset is applied to the RCTTextView instead of the UITextView's text container.
This diff fixes both bugs by applying the vertical insets to the UITextView's textContainerInset instead of the RCTTextView's frame (which is a wrapper around a real UITextView).
The left inset is still applied to the frame because there is a bug with the text rendering when the left textContainerInset is negative: the initial text doesn't show up until you focus the text view. The bug doesn't occur when setting the right textContainerInset, so I apply this workaround to only the left inset.
Closes https://github.com/facebook/react-native/pull/2297
Github Author: James Ide <ide@jameside.com>
Summary:
Our events all follow a common pattern, so there's no good reason why the configuration should be so verbose. This diff eliminates that redundancy, and gives us the freedom to simplify the underlying mechanism in future without further churning the call sites.
Summary:
Addresses #1534
When an image has a known width, but a height of 0 (which can happen if `flex: 1` is set on the `Image` element), `RCTDownloadManager` attempts to scale it to an invalid size, which results in a `NULL` `CGContextRef` and some scary warnings from UIKit:
```
<Error>: CGContextTranslateCTM: invalid context 0x0. This is a serious error. This application, or a library it uses, is using an invalid context and is thereby contributing to an overall degradation of system stability and reliability. This notice is a courtesy: please fix this problem. It will become a fatal error in an upcoming update.
```
This adds a check for zero width or height to the `RCTClipRect` function. If either dimension is zero, it is calculated based on the aspect ratio of the source image. This ensures that we don't try to create an invalid `CGContextRef`, and that images with an unknown dimension are still scaled, blended, etc.
Closes https://github.com/facebook/react-native/pull/2278
Github Author: Yusef Napora <yusef@napora.org>
Summary:
In the latest 0.9.0-rc of React Native, the default image won't load due to a typo and a missing condition in `setImage`. This PR contains fixes for both of them.
Closes https://github.com/facebook/react-native/pull/2269
Github Author: Tom Hastjarjanto <tom@intellicode.nl>
Summary:
Moved the view creation & property binding logic out of RCTUIManager into a separate RCTComponentData class - this follows the pattern used with the bridge.
I've also updated the property binding to use pre-allocated blocks for setting the values, which is more efficient than the previous system that re-contructed the selectors each time it was called. This should improve view update performance significantly.
Summary:
When mutation of a stack happens, we'd like to compute the diff of the stacks (before and after) so that
we can know which routes are removed in the new stack.
This diff adds a new method `substract` which does what we need.
Summary:
- Enables async/await in .babelrc and transformer.js
- Adds regenerator to package.json. Users still need to explicitly require the regenerator runtime -- this is so that you only pay for what you use.
- Update AsyncStorage examples in UIExplorer to use async/await
- Update promise tests in UIExplorer to use async/await in addition to the promise API
Closes https://github.com/facebook/react-native/pull/1765
Github Author: James Ide <ide@jameside.com>
Summary:
We already reimplement the spring computation, we were only using rebound for the tension/friction conversion. Turns out that it is quite small so we can just embed it. I'm doing this as I'm preparing for doing a web version of the feature and I'm trying to minimize the number of dependencies.
https://github.com/facebook/rebound-js/blob/master/rebound.js#L932
Summary:
Dynamic Text Sizes for Text component.
Text gains new prop - allowFontScaling (false by default).
There is also AccessibilityManager module that allows you to tune multipliers per each content size category.
Summary:
The bridge implementation on React Android does not currently support boxed numeric/boolean types (the equivalent of NSNumber arguments on iOS), nor does Java support Objective-C's nil messaging system that transparently casts nil to zero, false, etc for primitive types.
To avoid platform incompatibilities, we now treat all primitive arguments as non-nullable rather than silently converting NSNull -> nil -> 0/false.
We also now enforce that NSNumber * objects must be explicitly marked as `nonnull` (this restriction may be lifted in future if/when Android supports boxed numbers).
Other object types are still assumed to be nullable unless specifically annotated with `nonnull`.
Summary:
Disabling the scene this way would make the scene height go to zero and mess up the scroll position. By setting the bottom to the same distance, the view does not get resized and the scroll position is preserved through a scene disable cycle.
Summary:
# Summary
Add a method `keyOf` to NavigationRouteStack.
The method `keyOf` returns a key that is associated with the route.
The a route is added to a stack, the stack creats an unique key for it and
will keep the key for the route until the route is rmeoved from the stack.
The stack also passes the keys to its derived stack (the new stack created by the
mutation API such as `push`, `pop`...etc).
The key for the route persists until the initial stack and its derived stack no longer
contains this route.
# Why Do We Need This?
Navigator has needs to use an unique key to manage the scenes rendered.
The problem is that `route` itself isn't a very reliable thing to be used as the key.
Consider this example:
```
// `scene_1` animates into the viewport.
navigator.push('scene_1');
setTimeout(() => {
// `scene_1` animates off the viewport.
navigator.pop();
}, 100);
setTimeout(() => {
// Should we bring in a new scene or bring back the one that was previously popped?
navigator.push('scene_1');
}, 200);
```
Because we currently use `route` itself as a key for the scene, we'd have to block a route
until its scene is completely off the components tree even the route itself is no longer
in the stack otherwise we'd see strange animation of jumping scenes.
# What's Next
We're hoping that we can build pure reactive view for NavigationRouteStack easily.
The naive implementation of NavigationRouteStackView may look like this:
```
class NavigationRouteStackView {
constructor() {
this.state = {
staleScenes: {},
};
}
componentWillReceiveProps(nextProps) {
if (nextProps.stack !== this.props.stack) {
var stale;
var staleScenes = {...this.state.staleScenes};
this.props.stack.forEach((route, index, key) => {
if (nextProps.stack.keyOf(route) !== key) {
stale = true;
staleScenes[key] = {route, index, key, stale};
}
});
if (stale) {
this.setState({
staleScenes,
});
}
}
}
render() {
var scenes = [];
this.props.stack.forEach((route, index, key) => {
scenes.push({route, index, key});
});
Object.keys(this.state.staleScenes).forEach(key => {
scenes.push(this.state.staleScenes[key]);
});
scenes.sort(stableSortByIndex);
return <View>{scenes.map(renderScene)}</View>;
}
}
```
Summary:
If user taps the back button quickly, the app crashes becuase "pop"
internally only checks `this.state.presentedIndex` which does not
always update when transtion happens.
This diff addresses this issue.