Summary: `width` and `height` are required for this style prop, for the sake of consistency (and Fabric). Callsites should set one of them to 0 instead of not specifying it.
Reviewed By: TheSavior, shergin
Differential Revision: D8371064
fbshipit-source-id: b0ffd6b6543ac5456a3708382966e7b3df241f7e
Summary:
Replaces the existing `YellowBox` with a modern one.
Here are the notable changes:
- Sort warnings by recency (with most recent on top).
- Group warnings by format string if present.
- Present stack traces similar to RedBox.
- Show status of loading source maps.
- Support inspecting each occurrence of a warning.
- Fixed a bunch of edge cases and race conditions.
Reviewed By: TheSavior
Differential Revision: D8345180
fbshipit-source-id: b9e10d526b262c3985bbea639ba2ea0e7cad5081
Summary:
Currently on iOS and Android focus/blur events trigger onPressIn/onPressOut. Based on discussions with people are several companies who use react-native we're proposing instead triggering new events onFocus/onBlur. Initial discussion on Slack with some from the core team on Slack seemed positive.
Couple reasons:
* The current API behavior overloads onPressIn/onPressOut. That means on platforms like react-native-web, if focus/blur support was added (as we're hoping for), even though onPressIn/onPressOut would be useful as the name describes, you wouldn't be able to distinguish between it and browser element focus/blur events.
* The names aren't as self-documenting/intuitive as onFocus/onBlur, especially for react-dom users.
There aren't any current tests around this, but I intend to add them if we solidify the API.
There's also an option question on the transition--do we deprecate the existing API with a warning? This PR just deprecates them, though it will on any TV platform when something becomes focused regardless of whether they use the API or not. This isn't ideal. It's not clear if there are alternatives or if just right away breaking the API for TV users is the correct solution, if we can get consensus between the few parties who are using it.
***
I'm interested to hear counter points or prior discussions.
Cc/ matthargett dlowder-salesforce rozele
Closes https://github.com/facebook/react-native/pull/18470
Differential Revision: D8368109
Pulled By: hramos
fbshipit-source-id: 22587b82e091645e748b6c2d721fdff06d54837f
Summary: When a third party library is vendored, both the original copyright header as well as React Native's header are required.
Reviewed By: fkgozali
Differential Revision: D8284116
fbshipit-source-id: 1748eb011c843a87e9ed421597571b66334edfd2
Summary:
This PR fixes a bug in `deepFreezeAndThrowOnMutationInDev` which did not take into account that objects passed to it may have been created with `Object.create(null)` and thus may not have a prototype. Such objects don't have the methods `hasOwnProperty`, `__defineGetter__`, or `__defineSetter__` on the instance.
I ran into an unrecoverable error in React Native when passing this type of object across the bridge because `deepFreezeAndThrowOnMutationInDev` attempts to call `object.hasOwnProperty(key)`, `object.__defineGetter__` and `object__defineSetter__` on objects passed to it. But my object instance does not have these prototype methods.
Changes:
* Defined `Object.prototype.hasOwnProperty` as a `const` (pattern used elsewhere in React Native)
* Modified calls to `object.hasOwnProperty(key)` to use `hasOwnProperty.call(object, key)` (Per ESLint rule [here](https://eslint.org/docs/rules/no-prototype-builtins))
* Modified calls to deprecated methods `object.__defineGetter__` and `object.__defineSetter__` to instead use `Object.defineProperty` to define get and set methods on the object. (Per guidance on [MDN](https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/Object/__defineGetter__))
* Added a new test to `deepFreezeAndThrowOnMutationInDev-test` to verify the fix.
I tried to create a reproducible example to post to Snack by passing prototype-less objects to a `Text` component, in various ways, but they appear to be converted to plain objects before crossing the bridge and therefore they do not throw an error.
However, I was able to create a new test to reproduce the issue. I added the following test to `deepFreezeAndThrowOnMutationInDev-test`:
```JavaScript
it('should not throw on object without prototype', () => {
__DEV__ = true;
var o = Object.create(null);
o.key = 'Value';
expect(() => deepFreezeAndThrowOnMutationInDev(o)).not.toThrow();
});
```
The changes in this PR include this new test.
ESLint test produced no change in Error count (3) or Warnings (671)
N/A
Other areas with _possibly_ the same issue:
c6b96c0df7/Libraries/vendor/core/mergeInto.js (L50)8dc3ba0444/Libraries/ReactNative/requireNativeComponent.js (L134)
[GENERAL] [BUGFIX] [Libraries/Utilities/deepFreezeAndThrowOnMutationInDev] -Fix for compatibility with objects without a prototype.
Closes https://github.com/facebook/react-native/pull/19598
Differential Revision: D8310845
Pulled By: TheSavior
fbshipit-source-id: 020c414a1062a637e97f9ee99bf8e5ba2d1fcf4f
Summary:
Change message in Animated.Interpolation to "inputRange must be monotonically non-decreasing" as it's allowed to give the same x's like in the test [example](4435f08771/Libraries/Animated/src/__tests__/Interpolation-test.js (L71))
Simply giving improper value of interpolation input
[GENERAL] [MINOR] [AnimatedInterpolation.js] - Change error message on interpolation improper range error
Closes https://github.com/facebook/react-native/pull/19571
Differential Revision: D8310791
Pulled By: TheSavior
fbshipit-source-id: 803ef55104ad2a36231c5f18c0c089bd14822bf3
Summary:
Bump Prettier to use version 1.13.4
All code changes are caused by running Prettier and should only affect files that have an `format` header.
All other changes caused by yarn.
Reviewed By: ryanmce
Differential Revision: D8251255
fbshipit-source-id: 0b4445c35f1269d72730f2000002a27c1bc35914
Summary:
This uses the reference for `libfishhook.a` from the Products, rather than
the reference from the Frameworks group.
This fixes the build for the new Xcode build system, on both Xcode 9 and
Xcode 10.
Fixes#19569
- Using Xcode 10:
- Open `RNTester/RNTester.xcodeproj`
- Build RNTester, noting build errors
- Using Xcode 9:
- Open `RNTester/RNTester.xcodeproj`
- Switch to using new build system in `File > Project Settings > Build System`, selecting `New Build System (Preview)`
- Build RNTester, noting build errors
none
[IOS] [BUGFIX] [RCTWebSocket] - Fix build for new Xcode build system
Closes https://github.com/facebook/react-native/pull/19579
Differential Revision: D8287487
Pulled By: hramos
fbshipit-source-id: 5bfc9decb09ebc763824df8474b5897099d39ad7
Summary:
This pr makes a little bit simpler to dismiss all warnings (1 click instead of 2)
Sometimes you don't want to use `YellowBox.ignoreWarnings` to remember.
RNTester sceenshot:
<img width="322" alt="screen shot 2018-05-30 at 09 51 25" src="https://user-images.githubusercontent.com/1488195/40701475-1142506a-63ef-11e8-8fc9-ea1696d9cb65.png">
Or RNTester -> Native Animation Example -> Force JS Stalls (in the end of the list) also cause warning to test.
[GENERAL][ENHANCEMENT][YellowBox] - Move `Dismiss All` yellow box button to another place which makes a little bit simpler to dismiss warnings.
Closes https://github.com/facebook/react-native/pull/19501
Differential Revision: D8243823
Pulled By: hramos
fbshipit-source-id: 31887469cddb4adfd7b889ae0c29a3bf41e87b7b
Summary:
As we migrate over to static typing solutions for props, we cannot rely on always having `propTypes` available at runtime.
This gets us started on that journey by removing the native prop validation that happens when we require native components.
bypass-lint
Reviewed By: TheSavior
Differential Revision: D7976854
fbshipit-source-id: f3ab579a7f0f8cfb716b0eb7fd4625f8168f3d96
Summary:
Jest will now exclude undefined props from snapshots (https://github.com/facebook/jest/pull/6162). Updating the snapshots should fix the current `test_javascript` failures.
Circle CI
[INTERNAL][MINOR][Snapshots] - Update snapshots
Closes https://github.com/facebook/react-native/pull/19414
Differential Revision: D8125193
Pulled By: hramos
fbshipit-source-id: db8dcfcd8afbf9d6256f83c6e922680a7872d776
Summary:
While investigating an issue about blobs (https://github.com/facebook/react-native/issues/18223), I noticed that the fetch polyfill (https://github.com/github/fetch) uses blobs as the response type by default if the module is available (https://github.com/github/fetch/blob/master/fetch.js#L454). This surfaced some issue with the blob implementation on iOS that has since been fixed.
However after further review of the fetch polyfill and the way Blobs work in RN, I noticed a major issue that causes blobs created by fetch to leak memory. This is because RN blobs are not deallocated automatically like in the browser (see comment https://github.com/facebook/react-native/blob/master/Libraries/Blob/Blob.js#L108) and the fetch polyfill does not deallocate them explicitly using the close method.
Ideally we should implement automatic blob cleanup when the instance is garbage collected but implementing that is probably somewhat complex as it requires integrating with JSC. For now I suggest disabling the default handling of requests as blobs in the fetch polyfill. This will mitigate the issue for people not using Blobs directly. I'm not sure how well documented the Blob module is but we should make it clear that they currently require explicit deallocation with the close method for people using them directly.
Run a simple http request using fetch and make sure it does not use the Blob module anymore.
[GENERAL] [BUGFIX] [fetch] - Do not use blobs to handle responses in the fetch polyfill, fixes potential memory leak.
Closes https://github.com/facebook/react-native/pull/19333
Differential Revision: D8125463
Pulled By: hramos
fbshipit-source-id: 8f4602190dfc2643606606886c698e8e9b1d91d1
Summary:
Spotted a few typos while going through the source code
No tests are required
[INTERNAL] [MINOR] [Webview] - Fixed typos
Closes https://github.com/facebook/react-native/pull/19462
Differential Revision: D8176774
Pulled By: hramos
fbshipit-source-id: f1a9024b210e1a935dcdccd7e27daedb71d794bc
Summary:
StrictMode compliance is important for Async Rendering and other Future Tech(tm).
Also clean up some lint.
== Test Plan ==
* No more StrictMode warnings for `FlatList` (`VirtualizedList` is another story....).
* props warnings still show up when appropriate.
Reviewed By: sophiebits
Differential Revision: D8026333
fbshipit-source-id: e6fa2f02d546b883df6f3cff8776c26c6ef91bc9
Summary: In the upcoming Jest version `genMockFunction` and `genMockFn` are deprecated, so we need to kill them.
Reviewed By: rafeca
Differential Revision: D8107155
fbshipit-source-id: 4f46ab58e6e34224eb95e9355385da44f005ea94
Summary:
Upgrade Babel from beta.40 to beta.47
There are two important breaking changes here.
- If you want an AST from the Babel `transform` functions you must pass on `ast:true` in the options.
- The `sourceType` is now the only source of truth on whether to parse with the Script or Module goal. It defaults to `script` and can also be `module` or `unambiguous`. In the `unambiguous` case it will first try to parse with the Module goal and only if it crashes it will try again with the Script goal.
Beyond that there were some fixes and some smaller changes that may affect you. See the Babel changelogs for details (https://github.com/babel/babel/tags).
Also updated the way we generate the babel helpers file.
Reviewed By: rubennorte
Differential Revision: D8075280
fbshipit-source-id: 2bb902690e8a4b19d9cada2b7b0c94812b3d4f0f
Summary:
This is to unbreak builds due to deprecated methods.
Cocoapods build seems unaffected.
Reviewed By: mmmulani
Differential Revision: D8060139
fbshipit-source-id: a4302d649dd75d29d293aeffdcc352bf09b0c504
Summary: Moving target deployment to iOS 9.0+ from now on, removing customization for iOS 8.
Reviewed By: shergin
Differential Revision: D8053439
fbshipit-source-id: 292c58f15c6e6caf8b28d15c1521812d6ed675c5
Summary:
Fixes a bug I accidentally introduced in the responder logic for `Text`.
I forgot that I was using arrow functions to preserve `context` while still relying on the creation of `arguments`. Oops.
Differential Revision: D8077595
fbshipit-source-id: 1f7dc11ea90ca4d6bb2129823ba09c79fb5a32b0
