status
stringclasses 1
value | repo_name
stringlengths 9
24
| repo_url
stringlengths 28
43
| issue_id
int64 1
104k
| updated_files
stringlengths 8
1.76k
| title
stringlengths 4
369
| body
stringlengths 0
254k
⌀ | issue_url
stringlengths 37
56
| pull_url
stringlengths 37
54
| before_fix_sha
stringlengths 40
40
| after_fix_sha
stringlengths 40
40
| report_datetime
timestamp[ns, tz=UTC] | language
stringclasses 5
values | commit_datetime
timestamp[us, tz=UTC] |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
closed
|
facebook/react
|
https://github.com/facebook/react
| 10,906 |
["src/renderers/dom/shared/eventPlugins/__tests__/EnterLeaveEventPlugin-test.js", "src/renderers/shared/shared/ReactTreeTraversal.js"]
|
In React 16, onMouseEnter is triggered an extra time when entering a new child
|
版本（versions）：
[email protected]
[email protected]
描述（description）：
给父级别绑定 onMouseEnter 事件。在首次渲染后，鼠标移入新渲染进去的子元素会触发父元素的 onMouseEnter 事件。
示例（demo）：
```js
import React, {Component} from 'react';
class Demo extends Component {
constructor(prop) {
super(prop);
this.enter = this.enter.bind(this)
this.state = {
flag: false
}
}
enter(event) {
console.log(event.nativeEvent);
console.log("enter")
}
click() {
this.setState({flag:!this.state.flag})
}
render() {
let old = {
height: "100px",
width: "100px",
border: "1px solid #6dbbff"
};
let aa = {
height: "100px",
width: "100px",
border: "1px solid red"
};
let content = {
border: "1px solid black",
display: "flex",
padding:"10px"
}
return(
<div>
<div style={content} onMouseEnter={this.enter}>
{this.state.a}
<div style={old}>old</div>
<div style={aa}>old</div>
{this.state.flag && <div style={aa}>new</div>}
</div>
<button onClick={this.click.bind(this)}>点我（click me）</button>
</div>
)
}
}
```
|
https://github.com/facebook/react/issues/10906
|
https://github.com/facebook/react/pull/11164
|
08cbc257bdd75862c88ab1cef6ca2ac410cee8df
|
80596c2c92dce264f8572038730ce7aef647c5fa
| 2017-09-28T07:06:35Z |
javascript
| 2017-10-09T19:30:03Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 10,883 |
["docs/docs/reference-react-component.md"]
|
React.Component render Fragments docs typo in example: JSX tag is closed prematurely
|
**Do you want to request a *feature* or report a *bug*?**
bug
**What is the current behavior?**
https://facebook.github.io/react/docs/react-component.html#fragments
In the render Fragments example, the `li` JSX tag is closed prematurely.
```jsx
render() {
return [
<li key="A"/>First item</li>,
<li key="B"/>Second item</li>,
<li key="C"/>Third item</li>,
];
}
```
**If the current behavior is a bug, please provide the steps to reproduce and if possible a minimal demo of the problem via https://jsfiddle.net or similar (template: https://jsfiddle.net/ebsrpraL/).**
—
**What is the expected behavior?**
```jsx
render() {
return [
<li key="A">First item</li>,
<li key="B">Second item</li>,
<li key="C">Third item</li>,
];
}
```
**Which versions of React, and which browser / OS are affected by this issue? Did this work in previous versions of React?**
React 16.0.0 documentation website on Chrome Android. Previously array Fragments weren't supported.
|
https://github.com/facebook/react/issues/10883
|
https://github.com/facebook/react/pull/10885
|
a947d3f92fba61ba9107861dde9e4ea89ad87442
|
9f6f7b3520a6c1c57489feb5e55abbc0047aee10
| 2017-09-27T16:43:34Z |
javascript
| 2017-09-27T18:28:04Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 10,868 |
["docs/docs/portals.md"]
|
Grammar mistake on portals docs
|
**Do you want to request a *feature* or report a *bug*?**
Bug
**What is the current behavior?**
Current text:
> 116 Open up the browser DevTools
> 117 to observe that the button
> 118 is not a child the div
> 119 with onClick handler.
Note "**is not a child the div with onClick handler**" is missing a word.
**If the current behavior is a bug, please provide the steps to reproduce and if possible a minimal demo of the problem via https://jsfiddle.net or similar (template: https://jsfiddle.net/ebsrpraL/).**
https://github.com/facebook/react/blame/master/docs/docs/portals.md#L118
**What is the expected behavior?**
> Open up the browser DevTools to observe that the button is not a **direct** child **of** the div with **the** onClick handler.
**Which versions of React, and which browser / OS are affected by this issue? Did this work in previous versions of React?**
React v16.
|
https://github.com/facebook/react/issues/10868
|
https://github.com/facebook/react/pull/10870
|
aad09701929fc03c4d997af8a6e1bed0d3f4ac3a
|
7811677a4986ca0821c28bfc5e16d7efb38ead6a
| 2017-09-27T12:26:16Z |
javascript
| 2017-09-27T16:04:11Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 10,865 |
["src/renderers/dom/fiber/ReactDOMFiberComponent.js"]
|
The IE11 document.implementation.createHTMLDocument API requires the title argument to be specified
|
**Do you want to request a *feature* or report a *bug*?**
bug
**What is the current behavior?**
React interactivity lost, React crash?
**If the current behavior is a bug, please provide the steps to reproduce and if possible a minimal demo of the problem via https://jsfiddle.net or similar (template: https://jsfiddle.net/ebsrpraL/).**
Run with react-dom.development.js in IE11, not sure exactly what the exact repo steps are the fix I've attempted for IE11 was to simply pass an empty string.
~~~
var normalizeHTML = function (parent, html) {
if (!testDocument) {
testDocument = document.implementation.createHTMLDocument(""); // fix for IE11 here
}
var testElement = parent.namespaceURI === HTML_NAMESPACE ? testDocument.createElement(parent.tagName) : testDocument.createElementNS(parent.namespaceURI, parent.tagName);
testElement.innerHTML = html;
return testElement.innerHTML;
};
}
~~~
Source https://github.com/facebook/react/blob/e932ad68bed656eed5295b61ba74e5d0857902ed/src/renderers/dom/fiber/ReactDOMFiberComponent.js#L133-L146
MDN docs [here](https://developer.mozilla.org/en-US/docs/Web/API/DOMImplementation/createHTMLDocument).
**What is the expected behavior?**
Well, that React does not crash in dev mode and that interactively continues to work.
**Which versions of React, and which browser / OS are affected by this issue? Did this work in previous versions of React?**
`[email protected]` IE11
|
https://github.com/facebook/react/issues/10865
|
https://github.com/facebook/react/pull/10921
|
96fde8a09ad525f28910de2c95f0eee2d824d939
|
857ed18b06e2deab66f92eea87860baf56931769
| 2017-09-27T07:53:02Z |
javascript
| 2017-09-28T17:34:33Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 10,831 |
["src/renderers/__tests__/ReactStatelessComponent-test.js", "src/renderers/shared/fiber/ReactDebugCurrentFiber.js", "src/renderers/shared/fiber/ReactFiberBeginWork.js", "src/renderers/shared/fiber/ReactFiberCompleteWork.js", "src/renderers/shared/fiber/ReactFiberContext.js", "src/renderers/shared/fiber/ReactFiberReconciler.js", "src/renderers/shared/fiber/ReactFiberScheduler.js"]
|
Warning: Stateless function components cannot be given refs. Attempts to access this ref will fail.null after updating to React 16
|
**Do you want to request a *feature* or report a *bug*?**
Bug
**What is the current behavior?**
Page load throws console error: `Warning: Stateless function components cannot be given refs. Attempts to access this ref will fail.null`
```
// Chrome console
6VM598:27 Warning: Stateless function components cannot be given refs. Attempts to access this ref will fail.null
window.console.error @ VM598:27
printWarning @ warning.js:36
warning @ warning.js:60
mountIndeterminateComponent @ react-dom.development.js:10488
beginWork @ react-dom.development.js:10647
performUnitOfWork @ react-dom.development.js:12621
workLoop @ react-dom.development.js:12730
callCallback @ react-dom.development.js:1319
invokeGuardedCallbackDev @ react-dom.development.js:1358
invokeGuardedCallback @ react-dom.development.js:1215
performWork @ react-dom.development.js:12848
batchedUpdates @ react-dom.development.js:13292
performFiberBatchedUpdates @ react-dom.development.js:1666
stackBatchedUpdates @ react-dom.development.js:1657
batchedUpdates @ react-dom.development.js:1671
batchedUpdatesWithControlledComponents @ react-dom.development.js:1684
dispatchEvent @ react-dom.development.js:1894
```
**What is the expected behavior?**
No console error. Or at least some way to trace which is the stateless component that is causing the issue.
**Which versions of React, and which browser / OS are affected by this issue? Did this work in previous versions of React?**
`"react": "^16.0.0-rc.3",`
`Chrome Version 60.0.3112.113 (Official Build) (64-bit)`
`macOS Sierra 10.12.6`
- This error only started to appear after upgrading from React 15 -> 16.
|
https://github.com/facebook/react/issues/10831
|
https://github.com/facebook/react/pull/10915
|
47a6ac525dcb3bf3297588fa83adab4beee37e66
|
83a536e62264b7baf0bb5d83835821e6f4c45006
| 2017-09-26T14:12:57Z |
javascript
| 2017-09-28T19:40:26Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 10,772 |
["src/renderers/dom/fiber/ReactDOMFiberEntry.js", "src/renderers/dom/shared/__tests__/ReactDOMServerIntegration-test.js"]
|
Unexpected SSR difference warning with SVG <filter> primitives
|
**Do you want to request a *feature* or report a *bug*?**
Bug
**What is the current behavior?**
I've started recieving this warning about my SVGs which is using [feGaussianBlur](https://developer.mozilla.org/en-US/docs/Web/SVG/Element/feGaussianBlur):
`Warning: Expected server HTML to contain a matching <feGaussianBlur> in <filter>.`
**What is the expected behavior?**
Should be no warnings because there are no difference.
**Which versions of React, and which browser / OS are affected by this issue? Did this work in previous versions of React?**
Such behavior started right after updating to `React 16.0.0-rc.3` and never happened with `React 15.6.1`
|
https://github.com/facebook/react/issues/10772
|
https://github.com/facebook/react/pull/11174
|
18f408fb6b307631623328faff89afc5ec11d562
|
f42dfcdb943eb2c5299139820f626e8a81561165
| 2017-09-21T16:05:03Z |
javascript
| 2017-10-10T18:34:37Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 10,739 |
["fixtures/dom/src/components/fixtures/input-change-events/RadioGroupFixture.js", "fixtures/dom/src/components/fixtures/input-change-events/index.js", "packages/react-dom/src/client/ReactDOMFiberComponent.js", "packages/react-dom/src/client/ReactDOMFiberInput.js", "packages/react-dom/src/client/inputValueTracking.js"]
|
Recent radio input onChange changes break expected behavior
|
### Background ###
I see from [this issue thread](#1471) that recent changes were pushed in `15.6.x` that have – to my mind – broken some of the expected behavior when it comes to firing `onChange` events in a group of radio button inputs.
I am aware of the conversation in #1471, but the problem it's describing is not exactly the same as this so far as I can tell.
Note that the problem I'm about to describe *does not occur* in React <=`15.4.x`.
### Expectations ###
We would expect an individual radio button's `onChange` to fire in these two conditions (among others):
1. A radio input is clicked and it's state changes
2. A label for any radio input is clicked, causing (1).
Indeed, native elements work precisely this way:
```html
<html>
<head></head>
<body>
<input type="radio" id="radio1" name="radiogroup"/>
<label for="radio1">One</label>
<input type="radio" id="radio2" name="radiogroup"/>
<label for="radio2">Two</label>
</body>
<script>
var r1 = document.getElementById('radio1');
var r2 = document.getElementById('radio2');
var handler = function(event){
console.log(event);
};
r1.addEventListener('change', handler);
r2.addEventListener('change', handler);
</script>
</html>
```
If you click the above radio buttons and observe the console output, you'll see that the event is firing as expected: only when a desired input's state is being changed to something new.
[Here is a Codepen example](https://codepen.io/darth-cheney/pen/PJZjEq/)
### React Expectations ###
The following React component will work as expected in `15.4.x` but not `15.6.x`:
```jsx
const RadioTest = (props) => {
let handler = (event) => {
console.log(event);
}
return(){
<div>
<input
type="radio"
name="radiogroup1"
id="radio1"
onChange={handler}/>
<label htmlFor="radio1">One</label>
<input
type="radio"
name="radiogroup1"
id="radio2"
onChange={handler}/>
<label htmlFor="radio2">Two</label>
</div>
}
};
```
In `15.4.x` React, this the inputs in this component will work as you'd expect and as the native DOM elements work. [This Codepen shows it working properly with 15.4.1](https://codepen.io/darth-cheney/pen/yzeXKo/)
In `15.6.x`, the `onChange` event **seems to only fire once**. [Here is another Codepen showing it doesn't work](https://codepen.io/darth-cheney/pen/MEKoxV/) when 15.6.1 is imported.
It is, of course, entirely possible that I am doing something incorrectly (and have been doing so for a couple of years now), but the breaking change to my components only came with recent updates.
|
https://github.com/facebook/react/issues/10739
|
https://github.com/facebook/react/pull/11028
|
9b36df86c6ccecb73ca44899386e6a72a83ad445
|
e0e91310400b616e370c38469b401f968de61966
| 2017-09-18T19:40:55Z |
javascript
| 2017-11-16T14:08:14Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 10,672 |
["src/renderers/dom/fiber/ReactDOMFiberComponent.js", "src/renderers/dom/fiber/ReactDOMFiberEntry.js", "src/renderers/dom/shared/__tests__/ReactDOMServerIntegration-test.js", "src/renderers/shared/fiber/ReactFiberBeginWork.js", "src/renderers/shared/fiber/ReactFiberCompleteWork.js", "src/renderers/shared/fiber/ReactFiberHydrationContext.js", "src/renderers/shared/fiber/ReactFiberReconciler.js", "src/renderers/shared/fiber/ReactFiberScheduler.js"]
|
Wrong SSR diff warning due to non lowercase attributes
|
**Do you want to request a *feature* or report a *bug*?**
Bug
**What is the current behavior?**
I got this warning: `Warning: Extra attributes from the server: preserveAspectRatio,viewBox` during client side hydratation when my react tree contains mixed case attributes such as these svg ones.
**What is the expected behavior?**
No warnings because there are no difference between server and client attributes.
**Which versions of React, and which browser / OS are affected by this issue? Did this work in previous versions of React?**
**React 16.0.0-rc.2**
This worked on 15.*
This seems to happen because the attribute name is put to lowercase here: https://github.com/facebook/react/blob/6552519a211c3e43e3f6cf5b2237c90ae515d423/src/renderers/dom/fiber/ReactDOMFiberComponent.js#L901
but not in the set here:
https://github.com/facebook/react/blob/6552519a211c3e43e3f6cf5b2237c90ae515d423/src/renderers/dom/fiber/ReactDOMFiberComponent.js#L915
As the set removal is done with the lowercased attribute:
https://github.com/facebook/react/blob/6552519a211c3e43e3f6cf5b2237c90ae515d423/src/renderers/dom/fiber/ReactDOMFiberComponent.js#L1011
it stays in the `extraAttributeNames` set and I got wrongly the previous warning.
|
https://github.com/facebook/react/issues/10672
|
https://github.com/facebook/react/pull/10676
|
65b9ad94aaf62eb61b7b3888362c2e7c0b06697a
|
c55ffb37f1b8972dc954b8949e7fca6e8bce6c91
| 2017-09-11T16:26:28Z |
javascript
| 2017-09-13T15:31:40Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 10,637 |
["src/renderers/noop/ReactNoopEntry.js", "src/renderers/shared/fiber/ReactFiberReconciler.js", "src/renderers/shared/fiber/ReactFiberScheduler.js"]
|
Remove performWithPriority from scheduler
|
We don't actually use this anywhere, so let's remove it.
https://github.com/facebook/react/blob/f1d00156df5453bc62bd5668fefdcac4ca67e90e/src/renderers/shared/fiber/ReactFiberScheduler.js#L1504-L1512
|
https://github.com/facebook/react/issues/10637
|
https://github.com/facebook/react/pull/10638
|
8b7082eced4405fdd9243c5f4495148935d3e7d6
|
7d77d795c2c647effbea40272de3834bb38f4e52
| 2017-09-07T17:27:18Z |
javascript
| 2017-09-08T00:35:06Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 10,598 |
["src/renderers/dom/shared/__tests__/ReactDOMServerIntegration-test.js", "src/renderers/shared/server/ReactPartialRenderer.js"]
|
Returning plain string from component renders collapsed text nodes on server
|
Observed this on [email protected]
```js
import React from 'react';
import { renderToString } from 'react-dom/server';
const B = () => 'b';
console.log(renderToString(<div>{'a'}<B />{'c'}</div>));
```
This renders as `<div data-reactroot="">a<!-- -->bc</div>`. If you try to hydrate this you'll get the error: Text content did not match. Server: "bc" Client: "b".
|
https://github.com/facebook/react/issues/10598
|
https://github.com/facebook/react/pull/11109
|
7d3b44bf3772bd0d8d18834881f8f528a842f010
|
32ec797274ae2b956bab45aa335be50026ce1209
| 2017-09-02T22:53:32Z |
javascript
| 2017-10-05T19:05:00Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 10,590 |
["fixtures/attribute-behavior/AttributeTableSnapshot.md", "src/renderers/dom/shared/__tests__/ReactDOMAttribute-test.js", "src/renderers/dom/shared/__tests__/ReactDOMComponent-test.js", "src/renderers/dom/shared/__tests__/ReactDOMServerIntegration-test.js", "src/renderers/dom/shared/hooks/ReactDOMUnknownPropertyHook.js"]
|
Warn for unknown non-lowercase attributes
|
It seems problematic to allow arbitary non-lowercase attribute on HTML nodes. Because the server has no ability to keep them case sensitive (HTML isn't). So they will end up different on the client.
We should probably warn if the attribute is not a known one *and* is not lowercase.
|
https://github.com/facebook/react/issues/10590
|
https://github.com/facebook/react/pull/10699
|
c55ffb37f1b8972dc954b8949e7fca6e8bce6c91
|
c99bad9483309b4cc89bad004450cd668d34b044
| 2017-09-01T01:53:10Z |
javascript
| 2017-09-13T17:13:13Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 10,553 |
["docs/docs/reference-events.md"]
|
[Documentation] Add reference to onInvalid form event
|
**Do you want to request a *feature* or report a *bug*?**
Bug
**What is the current behavior?**
React has an `onInvalid` form event that isn't documented at https://facebook.github.io/react/docs/events.html#form-events
**If the current behavior is a bug, please provide the steps to reproduce and if possible a minimal demo of the problem via https://jsfiddle.net or similar (template: https://jsfiddle.net/84v837e9/).**
https://facebook.github.io/react/docs/events.html#form-events
**What is the expected behavior?**
`onInvalid` should be added to the list of form events
**Which versions of React, and which browser / OS are affected by this issue? Did this work in previous versions of React?**
This documentation error has been around since v15, when the event was added. https://facebook.github.io/react/blog/2016/04/07/react-v15.html
|
https://github.com/facebook/react/issues/10553
|
https://github.com/facebook/react/pull/10554
|
dd9db8b9d2de09f4b9e45fd03db58126253cb633
|
f5547ee3722104e6d41856ce846b5660c551544d
| 2017-08-27T02:29:38Z |
javascript
| 2017-08-29T15:11:45Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 10,468 |
["fixtures/dom/src/components/Header.js", "fixtures/dom/src/components/fixtures/form-state/ControlledFormFixture.js", "fixtures/dom/src/components/fixtures/form-state/index.js"]
|
DOM Fixture Form Coverage: Restored State and Auto-fill
|
Follow up from https://github.com/facebook/react/pull/10456. Presently, we don't have a manual fixture that ensures that form state is captured when the browser provides input on behalf of the user.
- [ ] Restored form state
- [ ] Autocomplete state
**Other notes**
As a reference point, I'd like to find some documentation around how restored state works. I think I might be using the wrong search terms. Does anyone have a link handy?
|
https://github.com/facebook/react/issues/10468
|
https://github.com/facebook/react/pull/17951
|
3e9251d605692e6db6103e4fca9771ac30a62247
|
2078aa9a401aa91e97b42cdd36a6310888128ab2
| 2017-08-16T12:21:43Z |
javascript
| 2020-02-04T14:53:35Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 10,458 |
["scripts/rollup/build.js", "scripts/rollup/header.js"]
|
License headers are missing on built bundles
|
Our built files are missing license headers.
cjs has no headers:
https://unpkg.com/[email protected]/cjs/react.development.js
https://unpkg.com/[email protected]/cjs/react-dom.development.js
umd has headers identifying the bundle name but not the license:
https://unpkg.com/[email protected]/umd/react.development.js
https://unpkg.com/[email protected]/umd/react-dom.development.js
Both should have license headers similar to this:
```
/**
* React v15.6.1
*
* Copyright 2013-present, Facebook, Inc.
* All rights reserved.
*
* This source code is licensed under the BSD-style license found in the
* LICENSE file in the root directory of this source tree. An additional grant
* of patent rights can be found in the PATENTS file in the same directory.
*
*/
```
We likely just need to change the code in getBanner in scripts/rollup/build.js. https://github.com/facebook/react/blob/b4a3e3b34594759c90aef1694d2166b2d1126fa1/scripts/rollup/build.js#L78
You can test by running `yarn build` and looking in build/packages.
cc @trueadm
|
https://github.com/facebook/react/issues/10458
|
https://github.com/facebook/react/pull/10490
|
f1d00156df5453bc62bd5668fefdcac4ca67e90e
|
8b7082eced4405fdd9243c5f4495148935d3e7d6
| 2017-08-15T06:24:18Z |
javascript
| 2017-09-07T18:38:16Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 10,438 |
["docs/docs/reference-react-component.md"]
|
React.Component Docs - displayName
|
In https://facebook.github.io/react/docs/react-component.html we have the following statement:
> **displayName**
> The `displayName` string is used in debugging messages. JSX sets this value automatically; see JSX in Depth.
I'm almost sure it's incorrect. Neither React or JSX sets `displayName` automatically (another error: in the docs "JSX in Depth" doesn't even mention `displayName`).
Maybe we could change it to something like:
> **displayName**
> The `displayName` string is used in debugging messages. Usually you don't need to set it explicitly because it's inferred from the function or class name that creates the component. You might want to set it explicitly if you want to display a different name for debugging purposes or when you create a Higher-Order Component, see [Wrap the Display Name for Easy Debugging](https://facebook.github.io/react/docs/higher-order-components.html#convention-wrap-the-display-name-for-easy-debugging) for details.
Makes sense?
|
https://github.com/facebook/react/issues/10438
|
https://github.com/facebook/react/pull/10451
|
11a1543b9ececd1fe8dbaf45df1e3ed3de5a8d9f
|
617f8810ba8e5ce6ba2f2917e06eea3425b2e0d5
| 2017-08-11T07:47:52Z |
javascript
| 2017-09-08T23:45:25Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 10,420 |
["fixtures/dom/src/components/fixtures/selects/index.js", "src/renderers/dom/shared/eventPlugins/ChangeEventPlugin.js", "src/renderers/dom/shared/wrappers/__tests__/ReactDOMSelect-test.js"]
|
<select>'s onChange event fires with incorrect selectedIndex, value
|
**Do you want to request a *feature* or report a *bug*?**
Report a bug.
**What is the current behavior?**
`<select>` element's `onChange` event is fired with incorrect `selectedIndex` and `value`, like it hasn't actually changed. This happens only if the element has been rendered under another React DOM root, see the example in the link below.
**If the current behavior is a bug, please provide the steps to reproduce and if possible a minimal demo of the problem**
https://jsfiddle.net/martinsb/wd54vmmj/7/
**What is the expected behavior?**
`onChange` callback handlers should receive events with those values that user has actually selected. In the test case above, for example, when choosing "Two", I expect to have the following console output: `Change event, selectedIndex: 1, value: 2`
**Which versions of React, and which browser / OS are affected by this issue? Did this work in previous versions of React?**
The bug is observed with `react` and `react-dom` packages version `16.0.0-beta.3`. Latest stable version (`15.6.1`) works just fine.
|
https://github.com/facebook/react/issues/10420
|
https://github.com/facebook/react/pull/10444
|
2cd0ecdaac3ddb5c5b5a0036e0467bf9f6107a2a
|
3bc64327f01a2c9fd1529934ebf4a484c55d7e94
| 2017-08-09T07:12:14Z |
javascript
| 2017-08-12T02:02:10Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 10,415 |
["src/renderers/dom/fiber/ReactDOMFiberComponent.js", "src/renderers/dom/fiber/ReactDOMFiberEntry.js", "src/renderers/dom/shared/DOMNamespaces.js", "src/renderers/dom/shared/__tests__/ReactServerRendering-test.js", "src/renderers/dom/shared/setInnerHTML.js", "src/renderers/dom/shared/utils/__tests__/setInnerHTML-test.js", "src/renderers/dom/stack/client/DOMLazyTree.js", "src/renderers/dom/stack/client/ReactDOMComponent.js", "src/renderers/shared/server/ReactPartialRenderer.js"]
|
Server side rendering: "Warning: <linearGradient /> is using uppercase HTML. Always use lowercase HTML tags in React."
|
**Do you want to request a *feature* or report a *bug*?**
Report a bug.
**What is the current behavior?**
Whenever I render SVG elements that are camel cased (e.g. `<linearGradient />`) on the server side, I get the following warning:
```
Warning: <linearGradient /> is using uppercase HTML. Always use lowercase HTML tags in React.
```
**What is the expected behavior?**
Warnings should be consistent with the client side rendering, if possible. In this case, I'd expect no warning, since using camel cased SVG element names is pretty standard.
For client side rendering, there seems to be a check for the current namespace:
https://github.com/facebook/react/blob/efcac24af26afba90e2ef525509ca40f3bd6f270/src/renderers/dom/fiber/ReactDOMFiberComponent.js#L344-L352
For server side rendering, there seems to be currently no such check in place:
https://github.com/facebook/react/blob/81706eeb7a51c1cea7fd8e66733a4f4618155ef5/src/renderers/shared/server/ReactPartialRenderer.js#L537-L544
**Which versions of React, and which browser / OS are affected by this issue? Did this work in previous versions of React?**
Hitting this on 16.0.0-beta.5. Was not hitting it on 15.6.1, the version I was previously on.
Thanks in advance.
|
https://github.com/facebook/react/issues/10415
|
https://github.com/facebook/react/pull/10452
|
9921e572722c0fa9d3cda65317f6de60f5d0919f
|
5e85f0065b80b6fbfa9715bc1e6f190ad719ce82
| 2017-08-09T00:32:12Z |
javascript
| 2017-08-13T23:17:45Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 10,408 |
["fixtures/dom/src/components/fixtures/date-inputs/switch-date-test-case.js"]
|
Broken "date inputs" test DOM fixture
|
1. Open "Date Inputs" test harness: http://localhost:3000/date-inputs
1. Click the little (x) to clear the input
> **RangeError: Invalid time value**
> src/components/fixtures/date-inputs/switch-date-test-case.js:43
This fixture is broken for 15.6.1 and 16 beta 3.
|
https://github.com/facebook/react/issues/10408
|
https://github.com/facebook/react/pull/10412
|
9166cf9b8d21b51098f5bf44c09925657ca93929
|
64287af8d1f9f89f4493644ddc9c205dddf36c44
| 2017-08-08T16:05:28Z |
javascript
| 2017-08-09T15:52:26Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 10,407 |
["src/renderers/__tests__/EventPluginHub-test.js", "src/renderers/dom/fiber/ReactDOMFiberComponent.js", "src/renderers/dom/fiber/__tests__/ReactDOMFiber-test.js", "src/renderers/dom/shared/__tests__/ReactDOMComponent-test.js", "src/renderers/shared/shared/event/EventPluginHub.js"]
|
Warn early if event listener is not a function
|
We used to throw if event listener is not a function in Stack:
```js
ReactDOM.render(
<h1 onClick={{}}>Hello World!</h1>,
document.getElementById('container')
);
```
<img width="565" alt="screen shot 2017-08-08 at 3 41 08 pm" src="https://user-images.githubusercontent.com/810438/29077502-2646412e-7c50-11e7-8c99-3b7f6c0a2589.png">
Fiber silently ignores this.
I think we should throw. This would let us add support for objects (e.g. for passive listener configuration) if we want to in a minor rather than a major.
|
https://github.com/facebook/react/issues/10407
|
https://github.com/facebook/react/pull/10453
|
6ab2869a20e5e9a6f1f0ac02da34692dac9c1c9d
|
34780da5c8d702daf84a46ba436a3c580b95a3cb
| 2017-08-08T14:43:25Z |
javascript
| 2017-08-23T12:38:48Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 10,406 |
["scripts/rollup/results.json", "src/renderers/dom/ReactDOMStackEntry.js", "src/renderers/dom/fiber/ReactDOMFiberEntry.js", "src/renderers/dom/fiber/__tests__/ReactDOMFiber-test.js", "src/renderers/dom/shared/checkReact.js"]
|
Show a better message if React is not loaded
|
Currently if React is not loaded, ReactDOM shows this:
<img width="664" alt="screen shot 2017-08-08 at 3 36 52 pm" src="https://user-images.githubusercontent.com/810438/29077217-784c7624-7c4f-11e7-88bc-b055407523a9.png">
We should throw a nicer error message in the entry point I think.
|
https://github.com/facebook/react/issues/10406
|
https://github.com/facebook/react/pull/10449
|
3bc64327f01a2c9fd1529934ebf4a484c55d7e94
|
5ff5700625e8a1b340a0d01ff9c88afb07e00bf6
| 2017-08-08T14:37:53Z |
javascript
| 2017-08-13T18:16:13Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 10,388 |
["docs/docs/reference-react-dom-node-stream.md", "scripts/error-codes/codes.json", "scripts/fiber/tests-passing-except-dev.txt", "scripts/fiber/tests-passing.txt", "scripts/rollup/results.json", "src/renderers/dom/ReactDOMNodeStreamRenderer.js", "src/renderers/dom/ReactDOMServerBrowserEntry.js", "src/renderers/dom/ReactDOMServerNodeEntry.js", "src/renderers/dom/shared/__tests__/ReactDOMServerIntegration-test.js", "src/renderers/dom/shared/__tests__/ReactServerRenderingBrowser-test.js"]
|
Consider renaming renderToStream to renderToNodeStream
|
With the checkin of #10362, the code to render to a stream now looks like:
```
import { renderToStream } from 'react-dom/server'
renderToStream(<div>some text</div>).pipe(response);
```
This is great! I am, however, a little concerned about taking the word "node" out of the mix, because I'm thinking that we may eventually want to render to [**browser** streams](https://www.w3.org/TR/streams-api/), which are a really neat feature that can be used in concert with "service-worker-side rendering". I would propose renaming `renderToStream` to `renderToNodeStream` and `renderToStaticStream` to `renderToStaticNodeStream`. This would give room for `renderToBrowserStream` in the future.
If I make a PR on this, would it have a chance of being accepted?
cc @gaearon & @sebmarkbage
|
https://github.com/facebook/react/issues/10388
|
https://github.com/facebook/react/pull/10425
|
8b1b0720cc856d6b43d35dd29e50f260c7472347
|
4d0891498575f60e12ed80d7e4a7c0e35b8b7975
| 2017-08-04T18:16:30Z |
javascript
| 2017-08-10T19:23:00Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 10,336 |
["src/renderers/shared/ReactDOMFrameScheduling.js"]
|
Delete rAF from ReactDOMFrameScheduling?
|
We don't use rAF for scheduling anything anymore, yet we still enable it in ReactDOMFrameScheduling. It can probably be simplified a bit by not exposing it.
|
https://github.com/facebook/react/issues/10336
|
https://github.com/facebook/react/pull/10337
|
81706eeb7a51c1cea7fd8e66733a4f4618155ef5
|
372445735eda4d7960419947c49499019adb79db
| 2017-08-01T01:58:04Z |
javascript
| 2017-08-02T14:26:37Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 10,328 |
["scripts/error-codes/extract-errors.js", "scripts/rollup/results.json"]
|
Some invariant messages aren't extracted
|
Not sure why but I can see these messages in the beta bundles:
```
ReactDOM.render(): Invalid component element. Instead of passing a string like 'div', pass React.createElement('div') or <div />.
Missing injection for fiber findDOMNode
Missing injection for stack findDOMNode
Element appears to be neither ReactComponent nor DOMNode. Keys: %s
Target container is not a DOM element.
```
I'd like to understand why.
|
https://github.com/facebook/react/issues/10328
|
https://github.com/facebook/react/pull/10348
|
c5d1558a5b7182c621658f3c3ab65c23e84166d8
|
73217a74a08fbca930135ac8d9bc46e461fa9fd3
| 2017-07-30T16:52:48Z |
javascript
| 2017-08-02T09:38:27Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 10,321 |
["fixtures/dom/public/index.html", "fixtures/dom/src/components/fixtures/error-handling/index.js", "src/renderers/shared/fiber/ReactFiberErrorLogger.js", "src/renderers/shared/fiber/ReactFiberScheduler.js", "src/renderers/shared/fiber/__tests__/ReactIncrementalErrorHandling-test.js", "src/renderers/shared/utils/ReactErrorUtils.js"]
|
"Error: null" in the captured error warning due to cross-origin issues
|
Three points from discussion in https://github.com/reactjs/react-redux/issues/756:
* Due to https://bugs.chromium.org/p/chromium/issues/detail?id=701371, environments like CodeSandbox show `Error: null` in our “React caught an error” warning. Let’s consider skipping both as they’re not very useful and look like a bug. `Error: null` looks like a separate error and draws attention away from the real error just above.
* In any case we should probably not print `The error was thrown at:` if we don’t have a JS stack.
* There is a related issue. What do we pass to error boundary if we caught `null`? It seems that most error boundaries will intuitively do something like `if (this.state.error)` or read `error.message`, which `null` will break. There are other falsy values too like `''` and `undefined`. My proposal is we always wrap primitives into our own `Error` objects that we pass to the boundary.
|
https://github.com/facebook/react/issues/10321
|
https://github.com/facebook/react/pull/10353
|
028e0ea9371b890fe6accac6b04ff8430abebd34
|
a650699d2f7e797204f26524038c29b403fa0213
| 2017-07-29T00:24:58Z |
javascript
| 2017-08-03T17:25:53Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 10,307 |
["src/renderers/dom/fiber/ReactDOMFiberComponent.js"]
|
Server rendering warnings too vague
|
**Do you want to request a *feature* or report a *bug*?**
Feature
**What is the current behavior?**
While server rendering React DOM and then hydrating with a different structure, React rightly warns that there are inconsistencies. But when I first saw these warnings I was not sure exactly what was going wrong.
Here are a few of the warnings I have been receiving:
```
Warning: Did not find a matching <button> in <div>.
Warning: Did not find a matching <img> in <div>.
Warning: Did not expect server HTML to contain a <div> in <div>.
```
Only on that last error did I catch on that this was a problem with server-rendering and the inconsistencies between that and my client render (intentional at the moment while testing some things in unpure render functions).
**What is the expected behavior?**
Being more explicit that this was a rendering inconsistency between server and client.
**Which versions of React, and which browser / OS are affected by this issue? Did this work in previous versions of React?**
React v16-beta.1
|
https://github.com/facebook/react/issues/10307
|
https://github.com/facebook/react/pull/10312
|
e1cfd6c082dae2e1472a6a051f6919725d9cfc91
|
7a60a8092144e8ab2c85c6906dd4a7a5815cff1f
| 2017-07-27T11:53:59Z |
javascript
| 2017-07-27T17:02:16Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 10,299 |
["scripts/rollup/results.json", "src/renderers/shared/server/ReactPartialRenderer.js"]
|
React 16 server renderer breaks with style object in prod mode
|
There is a bug in the React 16 beta server renderer, reported by @7rulnik [in the React 16 beta thread](https://github.com/facebook/react/issues/10294#issuecomment-318231160), where the renderer throws an exception when `style` is included and NODE_ENV is `production`.
Given this file:
```
// index.js
const React = require('react')
const ReactServer = require('react-dom/server')
const foo1 = React.createElement("div", {
className: "sign-link",
id: "sign-layout-link",
style: { position: 'absolute' }
});
console.log(ReactServer.renderToString(foo1))
```
If you run `NODE_ENV=production node index.js`, you'll get:
```
TypeError: re is not a function
at /Users/v7rulnik/projects/kupibilet/internals/kupibilet.ru/node_modules/react-dom/cjs/react-dom-server.production.min.js:1:7774
at /Users/v7rulnik/projects/kupibilet/internals/kupibilet.ru/node_modules/fbjs/lib/memoizeStringOnly.js:23:32
at f (/Users/v7rulnik/projects/kupibilet/internals/kupibilet.ru/node_modules/react-dom/cjs/react-dom-server.production.min.js:1:2423)
at b (/Users/v7rulnik/projects/kupibilet/internals/kupibilet.ru/node_modules/react-dom/cjs/react-dom-server.production.min.js:1:3163)
at e.renderDOM (/Users/v7rulnik/projects/kupibilet/internals/kupibilet.ru/node_modules/react-dom/cjs/react-dom-server.production.min.js:1:9665)
at e.render (/Users/v7rulnik/projects/kupibilet/internals/kupibilet.ru/node_modules/react-dom/cjs/react-dom-server.production.min.js:1:8762)
at e.read (/Users/v7rulnik/projects/kupibilet/internals/kupibilet.ru/node_modules/react-dom/cjs/react-dom-server.production.min.js:1:8452)
at Object.T [as renderToString] (/Users/v7rulnik/projects/kupibilet/internals/kupibilet.ru/node_modules/react-dom/cjs/react-dom-server.production.min.js:1:4384)
at repl:1:13
at ContextifyScript.Script.runInThisContext (vm.js:44:33)
```
Weirdly, dev mode does **not** throw an error.
|
https://github.com/facebook/react/issues/10299
|
https://github.com/facebook/react/pull/10300
|
f92343159dc26fe8fb7b3bbd4801cbc202d70e04
|
3abbcc48a539135bffa31d3caaa0601adf4ca42b
| 2017-07-27T04:15:09Z |
javascript
| 2017-07-27T09:19:05Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 10,265 |
["src/renderers/dom/shared/__tests__/ReactDOMServerIntegration-test.js", "src/renderers/dom/shared/hooks/ReactDOMUnknownPropertyHook.js"]
|
Invalid "unknown prop" warnings for SSR
|
Run the [`fixtures/ssr` test harness](https://github.com/facebook/react/tree/master/fixtures/ssr). In the terminal, the following warning is logged:
```
Warning: Unknown prop `onClick` on <a> tag. Remove this prop from the element. For details, see https://fb.me/react-unknown-prop
in a (at Page.js:12)
in p (at Page.js:18)
in div (at Page.js:17)
in Page (at App.js:12)
in div (at App.js:10)
in body (at Chrome.js:17)
in html (at Chrome.js:9)
in Chrome (at App.js:9)
in App (at render.js:18)
```
SSR behavior should match client-side behavior and not warn for a valid attribute.
|
https://github.com/facebook/react/issues/10265
|
https://github.com/facebook/react/pull/10272
|
171149a4b02a399a986c411b264515de45e70521
|
4c46b6c98acee5f55c7fc668c882678028e9de61
| 2017-07-24T16:26:59Z |
javascript
| 2017-07-25T16:01:48Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 10,217 |
["CHANGELOG.md"]
|
Incorrect data in compositionend event when typing Korean on IE11
|
To reproduce:
* In IE11 (on Win7 or Win10) go to https://jsfiddle.net/robbertbrak/84v837e9/164/
* Open the Developer console.
* Switch to the Microsoft Korean IME (standard settings) and switch to Korean input.
* Put the cursor in the contenteditable div and type `여름.` (on a QWERTY keyboard this is typed as `dufma.`).
* Do the same for the input field.
Result: in the developer console a list of composition events and the contents of the data attribute is logged. However, the data of the first `compositionend` event is wrong. It should be 여, not 여르. See screenshot below.
You can see that this is wrong by trying out the same thing on https://dvcs.w3.org/hg/d4e/raw-file/tip/key-event-test.html. As seen in the screenshot below, a plain (non-React) input field emits a `compositionend` event with the proper data.
This occurs with the latest version of React, but I have also seen this behaviour in older versions. An example of where this causes problems is in Draft JS.
|
https://github.com/facebook/react/issues/10217
|
https://github.com/facebook/react/pull/13571
|
550dd1d2eccd4855f1ec5e4155ac89dd43c70eff
|
0156740610473ee9bb5640c786024e38c8380441
| 2017-07-19T06:50:46Z |
javascript
| 2018-09-06T14:06:32Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 10,212 |
["src/renderers/__tests__/ReactComponent-test.js", "src/renderers/dom/ReactDOMNodeStreamRenderer.js", "src/renderers/dom/ReactDOMStringRenderer.js", "src/renderers/dom/shared/__tests__/ReactDOMServerIntegration-test.js", "src/renderers/dom/shared/__tests__/ReactServerRendering-test.js", "src/renderers/shared/server/ReactPartialRenderer.js"]
|
react-dom/server: renderToString fails with array type children
|
React 16 alpha allows returning array in render function but server renderer expects children to always be react elements
```
TypeError: Cannot read property 'toLowerCase' of undefined
at ReactDOMServerRenderer.renderDOM
```
source
```javascript
function renderDOM(element, context) {
var tag = element.type.toLowerCase(); // element is array
```
Fiddle: https://jsfiddle.net/84v837e9/160/
**[email protected]**
|
https://github.com/facebook/react/issues/10212
|
https://github.com/facebook/react/pull/10221
|
ca46c5278fe24dd9f564918d30e454033278e054
|
f732fc677b7f3f14ad7809a736d06705e4742347
| 2017-07-18T22:05:18Z |
javascript
| 2017-07-31T15:59:14Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 10,196 |
["scripts/fiber/tests-passing.txt", "src/renderers/dom/shared/__tests__/inputValueTracking-test.js", "src/renderers/dom/shared/inputValueTracking.js"]
|
Uncontrolled radio fix is breaking master
|
I'm not sure why but https://github.com/facebook/react/pull/10156 seems to have caused issues in production with us. We'll need to either revert or find a way to fix forward. Right now master is broken, and if we sync again before fixing, we'll break things.
|
https://github.com/facebook/react/issues/10196
|
https://github.com/facebook/react/pull/10207
|
197e184859140524e1de92abfc6d678bbc85b340
|
9043ad6b9dbc293b9cca33516bdcb026fbeee4b2
| 2017-07-16T17:32:05Z |
javascript
| 2017-07-18T15:31:04Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 10,188 |
["packages/react-dom/src/__tests__/ReactDOM-test.js", "packages/react-dom/src/__tests__/ReactServerRendering-test.js", "packages/react-dom/src/server/ReactPartialRenderer.js"]
|
Make Uses of ReactDebugCurrentFrame.getCurrentStack Reentrant
|
I realized that the refactor to use ReactDebugCurrentFrame as a decoupled stack frame for error messages is not reentrant: https://github.com/facebook/react/pull/10105/files#r127573520
Fiber is not reentrant but other renderers might be. The synchronous server renderer is atm.
We should add tests for warnings between renderers, such as calling a server-render from within a client render, and update the set/reset callsites to use push/pop instead to account for that.
|
https://github.com/facebook/react/issues/10188
|
https://github.com/facebook/react/pull/13181
|
d64d1ddb570a4fb7d748f4c500ac083886c2f0ec
|
6ebc8f3c07d8aadea127532fd33637cb091a9e66
| 2017-07-15T01:52:38Z |
javascript
| 2018-07-11T18:43:54Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 10,172 |
["scripts/rollup/results.json", "src/renderers/dom/ReactDOMStackEntry.js", "src/renderers/dom/fiber/ReactDOMFiberEntry.js", "src/renderers/dom/shared/ReactDOMClientInjection.js", "src/renderers/dom/shared/ReactDOMInjection.js"]
|
Server renderer shouldn't include event code
|
Since they currently share the injection, server renderer includes a lot of unnecessary client-only code. We should cut it out.
|
https://github.com/facebook/react/issues/10172
|
https://github.com/facebook/react/pull/10173
|
309412d8b6d6ed0cfbb50a6be2f1a99cca36fea5
|
7dd7c1702b0f71ea997c11dc4d72ee030ea8764c
| 2017-07-13T17:05:47Z |
javascript
| 2017-07-13T18:32:27Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 10,140 |
["scripts/fiber/tests-passing.txt", "src/renderers/__tests__/ReactUpdates-test.js", "src/renderers/shared/fiber/ReactFiberScheduler.js"]
|
Fiber infinite recursion in mounting lifecycle
|
here is an app https://gist.github.com/gre/52e28f74eda0beccc17e880efe18c250
There is intentionally a bug in this app: the App component returns `<Demo setVariables={v=>this.setState(..)} variables={..} />` and `Demo` calls `setVariables` in componentDidMount and componentWillReceiveProps (it shouldn't, this is bad user code). What this issue raises is a regression behaviour in React 16 compared to 15.
in React 15 : it renders but you directly get a `Uncaught RangeError: Maximum call stack size exceeded`.
in React 16 : nothing renders, the app never loads and the browser is spinning forever! I think React is stuck in an infinite recursion.
I presume this is because fiber is by nature async. The fact things never ends makes it very hard to investigate and isolate the source of the problem in a large codebase.
should there be some sort of maximum call stack in Fiber ?
> Context: When we migrated our app to latest React Native recently, we had a weird freezing case: the app was completely unresponsive (no button works) but you could still do native actions like scroll a ScrollView. however, even the RN devtool menu action was not working (like the inspector) – probably because it involves JS and React was just stuck in this recursion loop.
In our case, the issue was because calling `relay.setVariables` in lifecycle but with a wrong logic. for some reason our code used to work in the past but now creates this recursion loop.
|
https://github.com/facebook/react/issues/10140
|
https://github.com/facebook/react/pull/10178
|
999df3e7776247ddd78eda891c269196c6dd9992
|
4dabdd2a3048b5883cc71f8c91442a905c838b6d
| 2017-07-11T06:46:11Z |
javascript
| 2017-07-13T22:56:02Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 10,103 |
["packages/react-dom/src/__tests__/ReactCompositeComponent-test.js", "packages/react-reconciler/src/ReactFiberBeginWork.js"]
|
Improve error when you've forgotten to extend React.Component
|
When you forget to extend `React.Component` you get the following error:
```
TypeError: Cannot call a class as a function
at _classCallCheck (null.js:7:99)
at Hello (null.js:11:5)
...react internals...
```
[[Example](http://react.jsbin.com/?html,js,console)]
It would be nice if in dev you could do a check for `Component.prototype.render` and if it exists, provide a better warning along the lines of:
```
Warning: Using a class with a render method as a function, did you forget to extend React.Component?
```
I believe it would happen in [ReactCompositeComponent.js](https://github.com/facebook/react/blob/b840229286ac2a82fa49553ce793cf7b953d1845/src/renderers/shared/stack/reconciler/ReactCompositeComponent.js#L248).
|
https://github.com/facebook/react/issues/10103
|
https://github.com/facebook/react/pull/11168
|
4a43cf6eac39e95be30c86fc0de8a62bc132c4b0
|
7f10fae4c18510f64d7cf07bcbf4a1742387f2e4
| 2017-07-04T05:44:46Z |
javascript
| 2017-10-31T12:02:55Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 10,100 |
["scripts/fiber/tests-passing.txt", "src/renderers/__tests__/ReactCompositeComponent-test.js", "src/renderers/dom/stack/client/ReactReconcileTransaction.js", "src/renderers/dom/test/__tests__/ReactTestUtils-test.js", "src/renderers/native/ReactNativeReconcileTransaction.js", "src/renderers/native/__tests__/ReactNativeMount-test.js", "src/renderers/shared/stack/reconciler/CallbackQueue.js"]
|
setState callback called with null instead of undefined
|
**Do you want to request a *feature* or report a *bug*?**
bug
**What is the current behavior?**
In React v16, the callback function that is an optional parameter to the `setState` function is called and passed a value of null. Previously, in React v15, the function was called with a value of undefined. This change breaks the default parameters feature of JavaScript.
I discovered this behavior while working on a React Native project (hence the example using View & Text).
**If the current behavior is a bug, please provide the steps to reproduce and if possible a minimal demo of the problem via https://jsfiddle.net or similar (template: https://jsfiddle.net/84v837e9/).**
```
import React, {
Component,
} from 'react';
export default class MyComp extends Component {
componentDidMount() {
this.setState({}, (val) => console.log(val))
}
render() {
return (
<View>
<Text>{this.props.climbs}</Text>
</View>
);
}
}
```
**What is the expected behavior?**
The callback function will be called with undefined
**Which versions of React, and which browser / OS are affected by this issue? Did this work in previous versions of React?**
Broken behavior was observed in React 16 alpha 12 and React Native 0.45.1.
I replicated the issue in React 15.4.1 and React Native 0.40.0 and it was working as expected.
|
https://github.com/facebook/react/issues/10100
|
https://github.com/facebook/react/pull/10101
|
5d5589b29509a3d5b41ffd7233198746652c3441
|
50d905b0838857e76f7eb2f0875047c264f4c24e
| 2017-07-03T21:17:44Z |
javascript
| 2017-07-12T17:49:13Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 10,043 |
["src/renderers/dom/shared/CSSProperty.js"]
|
Style object with columns shorthand property wrong expansion
|
**Do you want to request a *feature* or report a *bug*?**
Report a bug.
**What is the current behavior?**
Adding a style object to an element with the `columns` shorthand property expand it the wrong way, it should be left untouched.
```
<div style={{ columns: 2 }}>
```
Expand to:
```
<div style="columns: auto 2px;">
```
Which is a totally different behaviour.
See: https://developer.mozilla.org/en-US/docs/Web/CSS/columns
**If the current behavior is a bug, please provide the steps to reproduce and if possible a minimal demo of the problem via https://jsfiddle.net or similar (template: https://jsfiddle.net/84v837e9/).**
https://codesandbox.io/s/YvYVjQZ9O
**What is the expected behavior?**
```
<div style="columns: 2;">
```
**Which versions of React, and which browser / OS are affected by this issue? Did this work in previous versions of React?**
Latest 15.6.1
|
https://github.com/facebook/react/issues/10043
|
https://github.com/facebook/react/pull/10115
|
8ca40630e635539dc817f97f5c4c5f1065373c05
|
ab4ddf64939aebbbc8d31be1022efd56e834c95c
| 2017-06-26T13:30:23Z |
javascript
| 2017-07-11T10:28:57Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 10,013 |
["src/renderers/dom/shared/setInnerHTML.js", "src/renderers/dom/shared/utils/__tests__/setInnerHTML-test.js"]
|
dangerouslySetInnerHTML in IE11 for svg elements
|
**Do you want to request a *feature* or report a *bug*?**
Bug
**What is the current behavior?**
When using `dangerouslySetInnerHtml` of svg elements, the DOM nodes from the previous render aren't being removed.
**If the current behavior is a bug, please provide the steps to reproduce and if possible a minimal demo of the problem via https://jsfiddle.net or similar (template: https://jsfiddle.net/84v837e9/).**
See this [fiddle](https://jsfiddle.net/csmrvzsb/1/)
**What is the expected behavior?**
The DOM should only contain the last rendered value for `dangerouslySetInnerHTML`.
**Which versions of React, and which browser / OS are affected by this issue? Did this work in previous versions of React?**
React - 15.5.4+
Browser - IE11
Did this work in previous versions of React? Not sure
After digging around I found #6982 . Which made me think this is happening because of [this line](https://github.com/facebook/react/blob/master/src/renderers/dom/shared/setInnerHTML.js#L37) in `setInnerHTML.js` where it looks like this
```js
// IE does not have innerHTML for SVG nodes, so instead we inject the
// new markup in a temp node and then move the child nodes across into
// the target node
if (node.namespaceURI === DOMNamespaces.svg && !('innerHTML' in node)) {
reusableSVGContainer =
reusableSVGContainer || document.createElement('div');
reusableSVGContainer.innerHTML = '<svg>' + html + '</svg>';
var svgNode = reusableSVGContainer.firstChild;
while (svgNode.firstChild) {
node.appendChild(svgNode.firstChild);
}
}
```
Because there's a call to `node.appendChild(svgNode.firstChild);` it will never remove all of the previous nodes but only add the new ones.
But that's just a guess...
|
https://github.com/facebook/react/issues/10013
|
https://github.com/facebook/react/pull/11108
|
3019210df2b486416ed94d7b9becffaf254e81c4
|
6ad6dcd11271bbbd355481b2ab058f0d47507efa
| 2017-06-20T16:38:38Z |
javascript
| 2017-10-10T10:54:47Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 9,988 |
["fixtures/dom/public/react-loader.js", "fixtures/dom/src/components/TestCase.js", "fixtures/dom/src/components/fixtures/input-change-events/RadioGroupFixture.js", "fixtures/dom/src/components/fixtures/input-change-events/index.js", "fixtures/dom/yarn.lock", "src/renderers/dom/fiber/ReactDOMFiberComponent.js", "src/renderers/dom/shared/__tests__/inputValueTracking-test.js", "src/renderers/dom/shared/inputValueTracking.js", "src/renderers/dom/stack/client/ReactDOMComponent.js"]
|
Regression: onChange doesn't fire with defaultChecked and radio inputs
|
**Do you want to request a *feature* or report a *bug*?**
Bug
**What is the current behavior?**
In React 15.6.1, this behaviour is changed; In 15.5.4, it fires the change event reliably.
15.6.1 - https://codesandbox.io/embed/VPA42ZnRo
15.5.4 - https://codesandbox.io/embed/JZ0mnE5oy
You'll need to have the console open to get the debugger statement.
In 15.6.1, the first change fires, but all subsequent changes do not fire. In 15.5.4, all changes fire.
**Which versions of React, and which browser / OS are affected by this issue? Did this work in previous versions of React?**
React 15.6.1 vs React 15.5.4; Chrome latest stable.
|
https://github.com/facebook/react/issues/9988
|
https://github.com/facebook/react/pull/10156
|
2dcdc3c633809d176ca6716dca1bddcd7d835be9
|
999df3e7776247ddd78eda891c269196c6dd9992
| 2017-06-16T12:00:38Z |
javascript
| 2017-07-13T20:02:31Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 9,962 |
["docs/warnings/refs-must-have-owner.md", "scripts/fiber/tests-passing.txt", "src/renderers/__tests__/multiple-copies-of-react-test.js", "src/renderers/shared/fiber/ReactChildFiber.js"]
|
String Refs Without An Owner Should Fail Gracefully
|
Occurs in Fiber with multiple instances of the React package (which can't share current owner).
|
https://github.com/facebook/react/issues/9962
|
https://github.com/facebook/react/pull/10151
|
5495e495de0bd15cd96a76ee0d7ca823cc09cc71
|
309412d8b6d6ed0cfbb50a6be2f1a99cca36fea5
| 2017-06-14T15:19:06Z |
javascript
| 2017-07-13T17:41:32Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 9,712 |
["fixtures/dom/src/components/fixtures/number-inputs/NumberInputDecimal.js", "fixtures/dom/src/components/fixtures/number-inputs/index.js", "scripts/fiber/tests-passing.txt", "src/renderers/dom/fiber/wrappers/ReactDOMFiberInput.js", "src/renderers/dom/shared/wrappers/__tests__/ReactDOMInput-test.js", "src/renderers/dom/stack/client/wrappers/ReactDOMInput.js"]
|
BUG: input type="number"
|
```
render () {
<input type="number" value={this.props.value} />
}
```
for example, when the original value is '.98', we need format it to '0.98'. but it is not take effect。
```javascript
class Test extends React.Component {
state = {
value: '.98',
}
changeValue = () => {
alert(0.98);
this.setState({
value: '0.98',
});
}
render() {
return <div>
<input type="number" value={this.state.value} />
<button onClick={this.changeValue}>changeValue</button>
</div>;
}
}
ReactDOM.render(<Test />, mountNode);
```
|
https://github.com/facebook/react/issues/9712
|
https://github.com/facebook/react/pull/9714
|
3daabf1efe49ee74d8c7bce4a7dabebfbba9c9db
|
8ab56e5c8b8f797a6b2bdd4188b4082c55a5190b
| 2017-05-18T04:00:42Z |
javascript
| 2017-06-09T01:27:17Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 9,639 |
["scripts/fiber/tests-passing.txt", "src/renderers/dom/shared/ReactBrowserEventEmitter.js", "src/renderers/dom/shared/ReactDOMEventListener.js", "src/renderers/dom/shared/ReactDOMInjection.js", "src/renderers/dom/shared/__tests__/ReactDOMEventListener-test.js", "src/renderers/dom/test/ReactTestUtils.js"]
|
Rename ReactEventListener to ReactDOMEventListener
|
Currently, ReactEventListener is a DOM-specific module but it has a generic sounding name that is confusing to contributors (i.e., confusing to me).
Let's rename it to ReactDOMEventListener. This should be pretty easy – just rename the file and update (and re-sort) the imports.
# @philipp-spiess is working on this task.
|
https://github.com/facebook/react/issues/9639
|
https://github.com/facebook/react/pull/9640
|
831fd6940ed65c390e8ef1965532add97802cca8
|
c0e32a1b03b99de1399f5abf8b7cb402e88d0609
| 2017-05-09T21:05:49Z |
javascript
| 2017-05-09T23:17:03Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 9,635 |
["scripts/fiber/tests-passing.txt", "src/renderers/__tests__/refs-test.js", "src/renderers/shared/stack/reconciler/ReactCompositeComponent.js"]
|
No error when element with ref created in another elements constructor in dev environment.
|
When creating an Element with a ref from inside of a constructor an error is thrown in production mode, but not in dev mode.
Here is a minimal sample:
https://jsfiddle.net/84v837e9/35/
An error, "Only a ReactOwner can have refs", should happen in dev and production modes.
This is happening in React 15.4
|
https://github.com/facebook/react/issues/9635
|
https://github.com/facebook/react/pull/10025
|
b4a3e3b34594759c90aef1694d2166b2d1126fa1
|
7f6b9409195a280ef3e30ad4cbeb9fa5a18d5cdd
| 2017-05-09T16:29:28Z |
javascript
| 2017-08-15T09:21:49Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 9,627 |
["package.json", "src/isomorphic/classic/__tests__/createReactClassIntegration-test.js", "yarn.lock"]
|
isMounted behavior different with `create-react-class`
|
**Do you want to request a *feature* or report a *bug*?**
Bug
**What is the current behavior?**
Calling `this.isMounted()` in `componentWillUnmount` in prior versions would return `true`. Now it returns `false`.
I believe this was untested behavior before, but the new tests that were added may check the wrong value: https://github.com/facebook/react/blob/master/src/isomorphic/classic/class/__tests__/create-react-class-integration-test.js#L417 Changing this line to test for `true` will exhibit the behavior.
The fix would be to defer setting the `__isMounted` flag to `false` until after all mixins and the `componentWillUnmount` method were called on the component.
**Which versions of React, and which browser / OS are affected by this issue? Did this work in previous versions of React?**
15.5.x and anything using `create-react-class` is broken. Working correctly in 15.4.x with `React.createClass`.
|
https://github.com/facebook/react/issues/9627
|
https://github.com/facebook/react/pull/9629
|
e817e8c95aa400728db1306e377e6b5ceb3f0293
|
7b05946776d2c66ade493f757a99d638c555c4b2
| 2017-05-09T00:31:55Z |
javascript
| 2017-06-20T22:37:01Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 9,601 |
["docs/docs/optimizing-performance.md"]
|
[docs] Update "Profiling Components with Chrome Timeline"
|
Google Chrome 58 just dropped and there is a shiny new **Performance** tab instead of the old **Timeline** tab in devTools. You can read all about it [here](https://developers.google.com/web/updates/2017/03/devtools-release-notes), but there is really not much to read except for:
> ### _"The Timeline panel has been renamed to the Performance panel, to better reflect its purpose."_
As a result, the docs for `optimizing-performance.md` needs to be updated [here](https://github.com/facebook/react/blob/master/docs/docs/optimizing-performance.md) to reflect these new changes.
I tested by profiling a component in our React app at work, seems to work the same so it's really just a matter of updating the docs. I will submit a PR with the update 👍
|
https://github.com/facebook/react/issues/9601
|
https://github.com/facebook/react/pull/9602
|
4b53dd0e6b892b6de6f991d168b1ebe5f99068b5
|
5cf571839e5a01804b175e9da34315a02ffcef28
| 2017-05-04T03:00:55Z |
javascript
| 2017-05-21T19:28:25Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 9,589 |
["packages/react/index.js"]
|
Ensure flat bundles don't duplicate code with weak minifiers
|
If the minifier doesn't eliminate constant expressions (like Uglify does), with the way our flat bundles currently work you would ship two bundles.
We should warn prominently in this case, as you would effectively ship a double bundle to production.
|
https://github.com/facebook/react/issues/9589
|
https://github.com/facebook/react/pull/10446
|
5ba1803ba2bf8581c2605247f3cf50aebacce1c4
|
34092a0f242c8059fd4cec4970f72d9326f2824d
| 2017-05-02T22:34:08Z |
javascript
| 2017-08-16T18:44:31Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 9,577 |
["src/renderers/__tests__/ReactComponent-test.js", "src/renderers/shared/fiber/ReactChildFiber.js"]
|
[Fiber] returning functions from render does not throw
|
https://jsfiddle.net/3apqnhmw/
```jsx
class Foo extends React.Component {
render() {
return <div>Foo</div>
}
}
class App extends React.Component {
render() {
return Foo
}
}
```
When `App` returns `Foo` instead of `<Foo />` accidentally, Fiber does not catch this.
|
https://github.com/facebook/react/issues/9577
|
https://github.com/facebook/react/pull/10376
|
6f28ecf0cb6630050cdae022b54dd0e6b75e1634
|
efa71484b334153958d8a0f35310c3c4ee72c731
| 2017-05-01T22:26:33Z |
javascript
| 2017-08-04T14:51:47Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 9,500 |
["scripts/prettier/index.js"]
|
yarn prettier runs a too long command for OS windows
|
CLI `yarn prettier` return error too long command.
The problem is only on windows OS. Checked on windows 10 x64 CLI cmd
|
https://github.com/facebook/react/issues/9500
|
https://github.com/facebook/react/pull/11458
|
e2e7fcce7e137e6e6fbc23cfadea1a0abaa0e6ce
|
40fbed572112a4ac90566fc822e33eb8cb4986c3
| 2017-04-22T22:03:35Z |
javascript
| 2017-11-05T21:06:08Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 9,479 |
["docs/_js/live_editor.js"]
|
Pretty-print error on React website
|
Just noticed that since we switched to Babel, compile error is ugly if you make a mistake on https://facebook.github.io/react/ interactive examples:
<img width="339" alt="screen shot 2017-04-21 at 4 53 33 pm" src="https://cloud.githubusercontent.com/assets/810438/25285719/23c3a592-26b3-11e7-976f-1f0b8ead7ed3.png">
We should fix it to preserve the newlines. It should be an easy fix.
You’d need to follow the instructions in the `docs` folder to get them running, then figure out a fix and send a PR. Please comment here if you plan to work on this so that two people don’t work on the same thing.
---
## Update: @tsriram is looking at this.
|
https://github.com/facebook/react/issues/9479
|
https://github.com/facebook/react/pull/9538
|
0e3280e2c6f4eb78cd9cd024e0345e1197f36734
|
70c01963d9a5c8d6789559334bbf558aee976632
| 2017-04-21T15:55:18Z |
javascript
| 2017-04-27T10:49:34Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 9,421 |
["docs/docs/installation.md"]
|
[DOCS] The new tabs on installation page doesn't work with link
|
Opening the link https://facebook.github.io/react/docs/installation.html#adding-react-to-an-existing-application displays the "Try React" tab.
|
https://github.com/facebook/react/issues/9421
|
https://github.com/facebook/react/pull/9422
|
855b8d3526a55e690108e8ede9091c14b6f62466
|
4c8c5debac5074427d5e6ce19896d4062e078a83
| 2017-04-13T20:06:26Z |
javascript
| 2017-04-13T20:51:19Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 9,317 |
["docs/docs/reference-react-component.md"]
|
improving doc
|
Here
https://facebook.github.io/react/docs/react-component.html#constructor
it says
> Beware of this pattern, as it effectively "forks" the props and can lead to bugs.
Nice but it would be **better** if docs explains what bugs it can lead to.
|
https://github.com/facebook/react/issues/9317
|
https://github.com/facebook/react/pull/9404
|
446f186279f3b5197d098c2f145ca809dd090d24
|
5cd44d21e9eb1e5014b9ae2109268600ce1c360f
| 2017-04-02T12:54:24Z |
javascript
| 2017-04-12T23:56:22Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 9,275 |
["docs/docs/installation.md"]
|
Confusingly structured Installation section
|
Hey guys, I want to bring out some issue in docs, Install section. https://facebook.github.io/react/docs/installation.html. This happened to me and one of my colleagues and I guess others may get confused too.
For example I want to use create-react-app approach for installing react.
This means I need to follow ONLY this section right: https://facebook.github.io/react/docs/installation.html#creating-a-single-page-application
But after my colleague followed that, he then thought had do follow things like this too:
https://facebook.github.io/react/docs/installation.html#enabling-es6-and-jsx
In other words the installation docs, doesn't say that if you follow the first link I gave for example above, following second link is not necessary as create-react-app includes Babel for example.
Isn't it?
I think the doc has to mention that it is an OR approach, either use this section, or this, etc.
|
https://github.com/facebook/react/issues/9275
|
https://github.com/facebook/react/pull/9401
|
30d6c598c6fa06e087ffa2ff32356a66917c68de
|
855b8d3526a55e690108e8ede9091c14b6f62466
| 2017-03-28T19:44:24Z |
javascript
| 2017-04-13T19:54:12Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 9,239 |
["scripts/fiber/tests-passing.txt", "src/isomorphic/modern/class/__tests__/ReactPureComponent-test.js", "src/renderers/shared/fiber/ReactFiberClassComponent.js", "src/renderers/shared/stack/reconciler/ReactCompositeComponent.js"]
|
Warn in dev if shouldComponentUpdate is defined on PureComponent
|
Currently, if you define `shouldComponentUpdate` on `React.PureComponent` nothing bad happens and it acts like `React.Component` https://jsfiddle.net/jekmdfva/
We should probably warn in dev since it defaults the point of `PureComponent`
|
https://github.com/facebook/react/issues/9239
|
https://github.com/facebook/react/pull/9240
|
5518bd44a9bdd573ec33f9f2af9d91ea90dedc90
|
ec527cc8342c9c80dfd3e396c4280346a7603450
| 2017-03-22T22:07:47Z |
javascript
| 2017-04-21T14:55:16Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 9,230 |
["packages/react-dom/src/__tests__/DOMPropertyOperations-test.js", "packages/react-dom/src/__tests__/ReactDOMComponent-test.js", "packages/react-dom/src/__tests__/ReactDOMServerIntegrationAttributes-test.js", "packages/react-dom/src/__tests__/utils/ReactDOMServerIntegrationTestUtils.js", "packages/react-dom/src/client/DOMPropertyOperations.js", "packages/react-dom/src/client/ReactDOMComponent.js", "packages/react-dom/src/server/ReactDOMServerFormatConfig.js", "packages/react-dom/src/shared/DOMProperty.js"]
|
Boolean attributes on Web Components
|
When setting boolean attributes on Web Components
`<x-search somebool name={this.props.name} />`
they get rendered as attribute/value pairs instead
`<x-search somebool="true" name={this.props.name}>`
This causes problems with some AMP components, for eg. `<amp-iframe />` which has a `resizable` attribute that gets rendered as `resizable="true"`. This results in non valid AMP content and developers crying with blood tears on their keyboards :)
> The attribute 'resizable' in tag 'amp-iframe' is set to the invalid value 'true'. (see https://www.ampproject.org/docs/reference/components/amp-iframe)
Is there any reason for the current behaviour? Will changing this break something else?
|
https://github.com/facebook/react/issues/9230
|
https://github.com/facebook/react/pull/24541
|
6e2f38f3a4d58f11bbe86ca6f938c27767366967
|
82c64e1a49239158c0daa7f0d603d2ad2ee667a9
| 2017-03-21T15:09:51Z |
javascript
| 2022-05-20T17:10:43Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 9,224 |
[".github/ISSUE_TEMPLATE.md"]
|
Create a new JSFiddle for issue template
|
Right now the current JSFiddle (https://jsfiddle.net/reactjs/69z2wepo/) can be a pain to use. It doesn't have any syntax highlighting and defaults to using `React.createClass` which in almost every case is not what people are going to be reproducing issues with.
|
https://github.com/facebook/react/issues/9224
|
https://github.com/facebook/react/pull/9289
|
686010037d10adde53b8ac926a7bfef0ab2ab8b1
|
8d6d3e8c05106903b8fb62ac6a6519bf270d48b3
| 2017-03-20T17:45:01Z |
javascript
| 2017-03-30T16:43:16Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 9,216 |
["docs/docs/addons-animation.md", "docs/docs/addons-create-fragment.md", "docs/docs/addons-perf.md", "docs/docs/addons-test-utils.md", "docs/docs/addons-two-way-binding-helpers.md"]
|
[Documentation][Add-Ons] Incorrect Prev Next links
|
Articles in Add-Ons have incorrect Prev Next links. For example the first `Performance Tools` has Prev link to the `Keyed Fragments` which is actually the fourth. `Immutability Helpers` does not have Prev Next links at all. I think the easiest way is just to remove the Prev Next links from Add-Ons articles. It will allow to add/update/remove articles without need to update Prev/Next links. What do you think?
|
https://github.com/facebook/react/issues/9216
|
https://github.com/facebook/react/pull/9226
|
f5144121f97253496248d26e61f1e69497e0a02a
|
6dfb65a1219a2f41ff643d489c4041dbea69d153
| 2017-03-19T08:30:14Z |
javascript
| 2017-04-21T14:31:17Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 9,141 |
["docs/tutorial/tutorial.md"]
|
Misleading wording in Documentation
|
Under "[Lifting State Up](https://facebook.github.io/react/tutorial/tutorial.html#lifting-state-up)":
> The `onClick` doesn't have any special meaning here, but it's popular to name handler props starting with `on` and their implementations with `handle`.
is misleading because it is special and references the SyntheticEvent handlers. Instead, the documentation should give a brief description of why `onClick` was used and reference the section on [SyntheticEvent](https://facebook.github.io/react/docs/events.html) to allow the users to see available handlers other than `onClick`.
|
https://github.com/facebook/react/issues/9141
|
https://github.com/facebook/react/pull/9427
|
37f9e35ad93e1cb1042a7bc3d8b78290dfcecec6
|
53a3939fb015504d006cc4f2d3ace67133164f09
| 2017-03-09T17:30:48Z |
javascript
| 2017-04-18T15:27:45Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 8,894 |
["src/renderers/shared/stack/reconciler/instantiateReactComponent.js"]
|
Closure Compiler: wrong assignment order in `instantiateReactComponent`
|
When consuming React from source with the Google Closure Compiler, [this bit](https://github.com/facebook/react/blob/eca5b1d48e71218800156ce474ec79b990b09fbd/src/renderers/shared/stack/reconciler/instantiateReactComponent.js#L26-L32) of code isn't being run in the correct order.
It seems like the Google Closure Compiler transformation is changing the function to a `var` assignment. It messes up the behavior because the original `instantiateReactComponent` function is hoisted. My thought is it's related to the fact that the `instantiateReactComponent` function is exported from that module.
Anyway, this produces an error because the assignment to `ReactCompositeComponentWrapper.prototype` is now in the wrong order (assignment -> function definition, when we want definition -> assignment).
This may also be related or the root cause of #8887.
The most straightforward solution is to bring the `Object.assign` call after the function is declared, which I've verified works correctly. I'm happy to submit a PR with the described approach.
|
https://github.com/facebook/react/issues/8894
|
https://github.com/facebook/react/pull/8895
|
fa1087b42c9eb4c769f6d094a22377f9055c5a69
|
43130597021cbf9332675f4ca53e4f49c7001ca1
| 2017-01-30T22:45:32Z |
javascript
| 2017-02-06T16:02:13Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 8,887 |
["src/renderers/dom/shared/DOMPropertyOperations.js"]
|
Usage of `this` in DOMPropertyOperations module breaks build in Closure Compiler under advanced optimizations
|
**Do you want to request a *feature* or report a *bug*?**
Report a bug
**What is the current behavior?**
Referring to object's method from a another method within same object, `this.deleteValueForProperty(node, name);`, [here](https://github.com/facebook/react/blob/c78464f8ea9a5b00ec80252d20a71a1482210e57/src/renderers/dom/shared/DOMPropertyOperations.js#L141), breaks React build in Closure Compiler under advanced optimizations.
**If the current behavior is a bug, please provide the steps to reproduce and if possible a minimal demo of the problem via https://jsfiddle.net or similar (template: https://jsfiddle.net/reactjs/69z2wepo/).**
Build a React app using Closure Compiler in advanced mode. (Unfortunately I can not provide a source of the app where I got this bug.)
**What is the expected behavior?**
An app built with Closure Compiler in advanced mode should not throw `this.deleteValueForProperty is not a function` exception
**Which versions of React, and which browser / OS are affected by this issue? Did this work in previous versions of React?**
React v15.4.2
Changing `this.deleteValueForProperty` to `DOMPropertyOperations.deleteValueForProperty` solves this issue. There are other methods of `DOMPropertyOperations` object which are using `deleteValueForProperty` properly via `DOMPropertyOperations.deleteValueForProperty` call.
If this makes sense, I'll prepare a PR.
|
https://github.com/facebook/react/issues/8887
|
https://github.com/facebook/react/pull/8892
|
466bb4ffb90206968e20dddc054fc285a4ae32fb
|
59aac010dd049719394c9cc4b70417302ce73c14
| 2017-01-30T15:13:44Z |
javascript
| 2017-01-31T00:32:26Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 8,875 |
["docs/css/react.scss"]
|
your website page not working properly in ie10
|
**Do you want to request a *feature* or report a *bug*?**
I have attached a screen short. check please. 80% people in the world use ie built in browser.
**What is the current behavior?**
**If the current behavior is a bug, please provide the steps to reproduce and if possible a minimal demo of the problem via https://jsfiddle.net or similar (template: https://jsfiddle.net/reactjs/69z2wepo/).**
**What is the expected behavior?**
**Which versions of React, and which browser / OS are affected by this issue? Did this work in previous versions of React?**
|
https://github.com/facebook/react/issues/8875
|
https://github.com/facebook/react/pull/8886
|
ec936dd848da3b7fe3c451cb46921ff2dd5e42ab
|
466bb4ffb90206968e20dddc054fc285a4ae32fb
| 2017-01-27T02:43:44Z |
javascript
| 2017-01-31T00:32:07Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 8,789 |
["grunt/config/browserify.js", "package.json", "yarn.lock"]
|
Missing object-assign licence (please check others)
|
**Do you want to request a *feature* or report a *bug*?**
bug
**What is the current behavior?**
Missing copyright licence
**If the current behavior is a bug, please provide the steps to reproduce and if possible a minimal demo of the problem via https://jsfiddle.net or similar (template: https://jsfiddle.net/reactjs/69z2wepo/).**
See https://markssoftware.com/2017/01/14/npm-is-lying-to-you-and-facebook-is-stealing-copyright/
**What is the expected behavior?**
Not a lawyer, Facebook should know how to show MIT according to their policies.
**Which versions of React, and which browser / OS are affected by this issue? Did this work in previous versions of React?**
Checked current version 15.4.2? All OSes and this is probably in previous versions.
|
https://github.com/facebook/react/issues/8789
|
https://github.com/facebook/react/pull/8803
|
06399b8ce3fa587531e491749ba502c193b09062
|
c6a7dc7f24b498d1c2cfb3aa2c04139b33012b80
| 2017-01-14T18:39:32Z |
javascript
| 2017-01-20T01:29:06Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 8,761 |
["docs/docs/reference-events.md", "src/renderers/dom/fiber/ReactDOMFiberComponent.js", "src/renderers/dom/shared/ReactBrowserEventEmitter.js", "src/renderers/dom/shared/eventPlugins/SimpleEventPlugin.js", "src/renderers/dom/stack/client/ReactDOMComponent.js", "src/renderers/shared/shared/event/EventConstants.js"]
|
onToggle Event name not recognized
|
**report a *bug*?**
**What is the current behavior?**
onToggle prop attached to a details html element does not attach the event handler , instead shows a warning "Warning: Unknown prop `onToggle` on `<details>` tag. Remove this prop from the element. For details, see https://fb.me/react-unknown-prop
in details (created by Details) ...
**If the current behavior is a bug, please provide the steps to reproduce and if possible a minimal demo of the problem via https://jsfiddle.net or similar (template: https://jsfiddle.net/reactjs/69z2wepo/).**
No time right now
**What is the expected behavior?**
-
**Which versions of React, and which browser / OS are affected by this issue? Did this work in previous versions of React?**
Latest
WorkAround:
Use onClick instead, (does not work fire event handler when details is opened with keyboard !)
|
https://github.com/facebook/react/issues/8761
|
https://github.com/facebook/react/pull/8831
|
269555e35c3196ebdeb9dcaaab51f1dce4a56462
|
ad133f5b3dbf01331f819f3000ab52097bf7d598
| 2017-01-12T14:12:21Z |
javascript
| 2017-02-03T10:25:09Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 8,676 |
["docs/_layouts/default.html"]
|
Docs look weird on small widths
|
Note the accidental top bar:
<img width="401" alt="screen shot 2017-01-03 at 15 23 29" src="https://cloud.githubusercontent.com/assets/810438/21612335/a23b8808-d1c8-11e6-8006-17bd5fa9f07a.png">
Is this a misplaced tag? Also can see this when any doc page is loading.
|
https://github.com/facebook/react/issues/8676
|
https://github.com/facebook/react/pull/8681
|
41588058d4724e681969bfe89e678850cffc4b70
|
6fa0b0a271510fec902314fcaa3e1f363b185ff6
| 2017-01-03T15:24:10Z |
javascript
| 2017-01-04T12:48:53Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 8,596 |
["src/shared/utils/PooledClass.js"]
|
Unused Function fiveArgumentPooler
|
There is no reference to any use of [fiveArgumentPooler](https://github.com/facebook/react/blob/32f5b034ed229d048f76ae74e18d270edc801dbf/src/shared/utils/PooledClass.js#L68) in the codebase. Deleted it without issue. Sending a Pull Request
|
https://github.com/facebook/react/issues/8596
|
https://github.com/facebook/react/pull/8597
|
ab19dd02921e96ebf42fce6717c0af6de60c68bd
|
b106ca0c8e5c1514b9fbfc69bd60c0c18ad47214
| 2016-12-18T04:41:02Z |
javascript
| 2016-12-20T04:53:04Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 8,529 |
["packages/react-dom/src/__tests__/__snapshots__/ReactTestUtils-test.js.snap", "packages/react-dom/src/events/DOMTopLevelEventTypes.js", "packages/react-dom/src/events/SimpleEventPlugin.js"]
|
Feature Request: Support auxclick event (onAuxClick)
|
Version 15.4.1
As starting Chrome 55, there is a new event `auxclick` to handle middle click, and `click` doesn't trigger by middle click anymore
https://developers.google.com/web/updates/2016/10/auxclick
|
https://github.com/facebook/react/issues/8529
|
https://github.com/facebook/react/pull/11571
|
75491a8f4b5924424b94198e7a56b3519f6430c6
|
ac72388563c4b7010f191aa9138a2ea81969a399
| 2016-12-08T07:12:28Z |
javascript
| 2018-08-03T18:57:34Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 8,497 |
["src/renderers/dom/shared/__tests__/ReactDOMComponent-test.js", "src/renderers/dom/shared/utils/assertValidProps.js"]
|
Add component stack to the contentEditable warning
|
*feature*
>Warning: A component is `contentEditable` and contains `children` managed by React. It is now your responsibility to guarantee that none of those nodes are unexpectedly modified or duplicated. This is probably not intentional.
It would help heaps if I would know _which_ component is causing this.
|
https://github.com/facebook/react/issues/8497
|
https://github.com/facebook/react/pull/11208
|
066f02281b5907101e57dd70e2c5c2ed9f748ca2
|
a3a10db22c70eb562ce0dc6e7c268642a85c1c01
| 2016-12-05T03:10:57Z |
javascript
| 2017-10-18T12:59:43Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 8,459 |
["scripts/fiber/tests-passing.txt", "src/renderers/testing/ReactTestRenderer.js", "src/renderers/testing/__tests__/ReactTestRenderer-test.js"]
|
[react-test-renderer] componentWillUnmount is not called for inner component instance
|
**Do you want to request a *feature* or report a *bug*?**
bug
**What is the current behavior?**
componentWillUnmount for inner instances is never called.
Here are 2 (jest) tests, the first works, not the second.
```js
import React from "react";
import renderer from "react-test-renderer";
test("top level componentWillUnmount works", () => {
let count = 0;
class A extends React.Component {
componentWillUnmount() {
count ++;
}
render() {
return <div />;
}
}
const inst = renderer.create(<A />);
inst.unmount();
expect(count).toEqual(1);
});
test("inner componentWillUnmount works", () => {
let count = 0;
class A extends React.Component {
componentWillUnmount() {
count ++;
}
render() {
return <div />;
}
}
const inst = renderer.create(
<div>
<A />
</div>
);
inst.unmount();
expect(count).toEqual(1);
});
```
**What is the expected behavior?**
The two test above should pass. All component that are unmounted should have the proper lifecycle, whatever the nesting is.
**Which versions of React, and which browser / OS are affected by this issue? Did this work in previous versions of React?**
React 15.4.0
|
https://github.com/facebook/react/issues/8459
|
https://github.com/facebook/react/pull/8512
|
3def431543f9ea9e872ff98966c53a7f00f292e9
|
931cad5aae77659332ec771c124258edf139d568
| 2016-11-30T10:38:31Z |
javascript
| 2016-12-14T15:57:21Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 8,409 |
["src/renderers/dom/fiber/ReactDOMFiberEntry.js"]
|
React DevTools Download should not appear in chrome-extension://
|
This is a bug report.
**Expected Behavior**
React should not suggest me to download the DevTools extension in a `chrome-extension://` page, because the extension has no permission to inject the devtool.
**Actual Behavior**
`Download the React DevTools and use an HTTP server (instead of a file: URL) for a better development experience: https://fb.me/react-devtools`
**Reproducing**
1. Create a Chrome extension with an `option` or `popup` page that uses React
2. Load the extension to Chrome and open up that page.
3. Observe console.
**Version**
Google Chrome 56.0.2924.3
React 15.4.0
|
https://github.com/facebook/react/issues/8409
|
https://github.com/facebook/react/pull/10359
|
982380863272b900e0500631b829408b96220aae
|
8890db707bd9cf535ccb5cbbfa87a7f4564fe399
| 2016-11-24T07:19:54Z |
javascript
| 2017-08-02T19:04:49Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 8,395 |
["fixtures/dom/src/components/fixtures/text-inputs/index.js", "packages/react-dom/src/client/ReactDOMFiberInput.js"]
|
Firefox validation triggers on input component render
|
When using native form validation and loading a page with a form containing an input with the `required` prop set, it triggers browser validation immediately. This produces Firefox's native error highlighting. This is because Firefox appears to run validation when the DOM is used to set a value on an existing input component.
The behavior can be seen here: http://codepen.io/anon/pen/zowOzo
Angular has [run into this already](https://github.com/ggershoni/angular.js/commit/cc311740c6b680c68a88c44fecd97ede1f18f831).
|
https://github.com/facebook/react/issues/8395
|
https://github.com/facebook/react/pull/12925
|
aa85b0fd5ffc92de38720c29833a54c67285abfb
|
8aeea5afa2d688968545b1cc5a30a48ddc0fa503
| 2016-11-23T13:58:23Z |
javascript
| 2018-05-29T13:48:58Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 8,379 |
["packages/react-call-return/src/ReactCallReturn.js", "packages/react-dom/src/server/ReactPartialRenderer.js", "packages/react-reconciler/src/ReactChildFiber.js", "packages/react-reconciler/src/ReactPortal.js", "packages/react/src/React.js", "packages/react/src/ReactChildren.js", "packages/react/src/ReactElement.js", "packages/react/src/ReactElementValidator.js", "packages/shared/ReactSymbols.js", "scripts/rollup/results.json"]
|
React app not rendering in IE11 and below
|
**Do you want to request a *feature* or report a *bug*?**
BUG
**What is the current behavior?**
App doesn't render in IE11 and below.
**If the current behavior is a bug, please provide the steps to reproduce and if possible a minimal demo of the problem via https://jsfiddle.net or similar (template: https://jsfiddle.net/reactjs/69z2wepo/).**
- Build React app with Webpack
- Launch IE11 or below (not an emulator)
- You will see a blank screen or half-compiled React app with the following error:
**What is the expected behavior?**
- To render as it does in Chrome, Firefox, etc.
**Which versions of React, and which browser / OS are affected by this issue? Did this work in previous versions of React?**
We've temporarily shrink-wrapped our dependencies so that we can continue to serve our app. We're starting to believe that it's coming from a dependency that `react`, `apollo` or `webpack` may be relying on...
Shrink-wrapped versions of our dependencies (left is the new one that's broken and the right one is the old one that continues to work across all browsers) https://www.diffchecker.com/SyaJUcsk
|
https://github.com/facebook/react/issues/8379
|
https://github.com/facebook/react/pull/11629
|
d1f6fbd22a7ccce6403ad35651994207c366a7ab
|
1cb6199d22af6f6ba2f55e4db18ed2f4216aaaf2
| 2016-11-22T14:42:34Z |
javascript
| 2017-11-22T18:08:22Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 8,301 |
["grunt/config/browserify.js"]
|
react-dom 15.4.0 does not work in AMD enviroment
|
If I try to load react-dom with RequireJS, I get an error: "Mismatched anonymous define()".
Looking at the react-dom from the 15.4.0 distribution (https://cdnjs.cloudflare.com/ajax/libs/react/15.4.0/react-dom.js), I notice that there seem to be two UMD wrappers at the top.
This is probably related to the change in the packaging structure (#7164 and #7168).
|
https://github.com/facebook/react/issues/8301
|
https://github.com/facebook/react/pull/8374
|
6beb87eb73cb7a5e390c47614b5f796afdfa3f65
|
a3ba48bf724de4c23d7c32f1b381476bcd3bb6fa
| 2016-11-16T15:29:48Z |
javascript
| 2016-11-23T01:16:45Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 8,268 |
["docs/community/support.md"]
|
Add #reactjs irc note to support page.
|
**Do you want to request a *feature* or report a *bug*?**
Docs change
I think there used to be a pointer to the freenode #reactjs channel in the [support page](https://facebook.github.io/react/community/support.html), but now only Reactiflux is linked. Could the note be added back in? The channel still has 500+ users in it most hours of the day, with several of us who frequently answer questions, and a pretty good flow of conversation and questions throughout the day.
|
https://github.com/facebook/react/issues/8268
|
https://github.com/facebook/react/pull/8270
|
b9f65d9b1dcb751bb38f8deec6340a5b5d295d28
|
9d201918bf4432fd3e231088d4cdb8146a1c7861
| 2016-11-11T12:48:58Z |
javascript
| 2016-11-11T21:21:46Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 8,260 |
["packages/react-dom/src/__tests__/ReactErrorLoggingRecovery-test.js", "packages/react-reconciler/src/ReactFiberErrorLogger.js", "packages/react-reconciler/src/__tests__/ReactIncrementalErrorLogging-test.internal.js", "packages/react-reconciler/src/__tests__/ReactIncrementalErrorLogging-test.js", "scripts/jest/matchers/toWarnDev.js", "scripts/jest/setupEnvironment.js", "scripts/jest/setupTests.js", "scripts/jest/shouldIgnoreConsoleError.js"]
|
Make sure that jest tests fail if an error is thrown within a jsdom event handler
|
In a few situations in our unit tests, we use .dispatchEvent on a jsdom DOM node in order to use the jsdom event system:
https://github.com/facebook/react/blob/e612826650ff68e73bff4d71db01fea8147ee1fd/src/renderers/dom/shared/wrappers/__tests__/ReactDOMInput-test.js#L127
But it appears that if the event handler throws (ex: change the onChange handler in that test to throw), then the error is silently ignored and does not cause the test to fail.
jsdom does make some attempt to report an error:
https://github.com/tmpvar/jsdom/blob/9.8.3/lib/jsdom/living/events/EventTarget-impl.js#L235
We should hook into this in our jest config (scripts/jest/test-framework-setup.js) and see if we can fail any test that triggers this exception behavior. From the implementation of reportException it looks like we might be able to do an onerror handler; if that doesn't work then hooking into their "virtual console" may be possible.
|
https://github.com/facebook/react/issues/8260
|
https://github.com/facebook/react/pull/13401
|
69e2a0d732e1ca74f6dc5df9d0ddd0bf24373965
|
b2adcfba32d133aabdbc182bba1001e729060abe
| 2016-11-10T19:19:30Z |
javascript
| 2018-08-15T16:44:46Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 8,238 |
["scripts/fiber/tests-passing.txt", "src/renderers/shared/fiber/ReactFiberCommitWork.js", "src/renderers/shared/fiber/ReactFiberUpdateQueue.js", "src/renderers/shared/fiber/__tests__/ReactIncremental-test.js"]
|
[Fiber] Handle errors in callbacks
|
During the commit phase (when updates are flushed to the DOM), errors that are thrown inside `componentDidMount` and `componentDidUpdate` are trapped so they can be handled later, after the tree has been committed.
An error thrown in a `setState` callback should be trapped and handled the same way.
Here's the relevant section of code: https://github.com/facebook/react/blob/master/src/renderers/shared/fiber/ReactFiberCommitWork.js#L302-L330
`callCallbacks` calls all the `setState` callbacks that are part of the update queue (a linked list of updates). If any of those callbacks throw an error, we should catch the error and continue calling the rest of the callbacks.
This is a good first issue for someone interested in contributing to React/React Fiber. Please read [Fiber Principles: Contributing to Fiber](https://github.com/facebook/react/issues/7942) before submitting a PR.
|
https://github.com/facebook/react/issues/8238
|
https://github.com/facebook/react/pull/8242
|
9d201918bf4432fd3e231088d4cdb8146a1c7861
|
077822e9d0f8a26bd505ee38e6353449ac8a845b
| 2016-11-08T19:03:52Z |
javascript
| 2016-11-12T20:51:36Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 8,209 |
["docs/docs/thinking-in-react.md"]
|
Documentation: Sample is still using string refs
|
Sample code "Step 5: Add Inverse Data Flow" in Quick Start/Thinking in React is still using old string refs.
```
handleChange() {
this.props.onUserInput(
this.refs.filterTextInput.value,
this.refs.inStockOnlyInput.checked
);
}
render() {
return (
<form>
<input
type="text"
placeholder="Search..."
value={this.props.filterText}
ref="filterTextInput"
onChange={this.handleChange}
/>
<p>
<input
type="checkbox"
checked={this.props.inStockOnly}
ref="inStockOnlyInput"
onChange={this.handleChange}
/>
{' '}
Only show products in stock
</p>
</form>
);
}
```
I think it's better (easier to understand) to use a new syntax.
```
handleChange() {
this.props.onUserInput(
this.filterTextInput.value,
this.inStockOnlyInput.checked
);
}
...
ref={(input) => this.filterTextInput = input}
...
ref={(input) => this.inStockOnlyInput = input}
```
|
https://github.com/facebook/react/issues/8209
|
https://github.com/facebook/react/pull/8815
|
c8a41672ede96c629a04a67da750f01493e62f6c
|
5d96162b57824502497e49edf01348c60532758f
| 2016-11-04T22:45:22Z |
javascript
| 2017-01-17T18:17:32Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 8,134 |
["package.json"]
|
React does not support node 7.x for development
|
**What is the current behavior?**
`AssertionError: Current node version is not supported for development, expected "7.0.0" to satisfy "4.x || 5.x || 6.x".`
**Steps to reproduce**
1. Install node 7.x
2. Clone [React repo](https://github.com/facebook/react)
3. run `npm i`
**What is the expected behavior?**
React support node v7 for development
|
https://github.com/facebook/react/issues/8134
|
https://github.com/facebook/react/pull/8135
|
1efb7a4e354703010c2001d769b516c18ed4726f
|
9dc7a9594af963c4e5311e95d344e809a75f66ee
| 2016-10-28T21:15:51Z |
javascript
| 2016-10-29T10:24:12Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 8,080 |
["scripts/fiber/tests-passing.txt", "src/isomorphic/classic/types/ReactPropTypes.js", "src/isomorphic/classic/types/__tests__/ReactPropTypes-test.js", "src/renderers/dom/shared/utils/ReactControlledValuePropTypes.js", "src/shared/types/ReactPropTypesSecret.js", "src/shared/types/checkReactTypeSpec.js"]
|
Re-add the warning about PropTypes
|
After a discussion with @spicyj we decided to keep the warning even though the functions will throw in production. The warning is the only way to learn about that.
We should also document how to pass the secret through in case you know what you're doing.
|
https://github.com/facebook/react/issues/8080
|
https://github.com/facebook/react/pull/8903
|
994a0c8b0c5a2104d57bf99edb9c6667bb58a281
|
1d90894f93d9564cdb6aec9ca3e3da7e1eb01b5a
| 2016-10-24T19:42:52Z |
javascript
| 2017-02-08T16:29:19Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 8,034 |
["docs/docs/lists-and-keys.md"]
|
docs: list and key examples override Number
|
**Do you want to request a _feature_ or report a _bug_?**
bug
**What is the current behavior?**
In #8027 some [example](http://react-new-docs-temporary-preview.surge.sh/react/docs/lists-and-keys.html#extracting-components-with-keys)s use `Number` as component name
``` javascript
function Number(props) {
// ...
}
```
Maybe it is a good idea to rename this component to `NumberItem` or something.
|
https://github.com/facebook/react/issues/8034
|
https://github.com/facebook/react/pull/8039
|
54dfe0b43caf52e9519787ac2dc8b1acb02f9d7e
|
7340fa9d933781747be0c3ac9850e264590fe5cd
| 2016-10-21T20:25:21Z |
javascript
| 2016-10-22T20:44:28Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 7,975 |
["docs/contributing/implementation-notes.md"]
|
Docs - Implementation notes - replaceChild args in worng order
|
The order of the arguments is wrong
``` js
// how it is now
prevNode.parentNode.replaceChild(prevNode, nextNode);
// how it should be
prevNode.parentNode.replaceChild(nextNode, prevNode);
```
https://facebook.github.io/react/contributing/implementation-notes.html#updating-composite-components
https://developer.mozilla.org/en/docs/Web/API/Node/replaceChild
|
https://github.com/facebook/react/issues/7975
|
https://github.com/facebook/react/pull/7989
|
01753cde852ef152ed1da844b27b37d82f8f6faa
|
cf37402338f77929e84dfba863f73ad44c7f6259
| 2016-10-14T13:05:31Z |
javascript
| 2016-10-22T20:59:54Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 7,927 |
["src/renderers/shared/stack/reconciler/instantiateReactComponent.js", "src/shared/utils/getNextDebugID.js", "src/test/ReactShallowRenderer.js"]
|
Changing state too quickly cause error with shallow render
|
**Do you want to request a _feature_ or report a _bug_?**
Report a bug
**What is the current behavior?**
Running `ReactTestUtils.createRenderer().render()` causes react to throw an exception:
```
Warning: Exception thrown by hook while handling onSetChildren: Invariant Violation: Expected onBeforeMountComponent() parent and onSetChildren() to be consistent (3 has parents 0 and 2).
Invariant Violation: Expected onBeforeMountComponent() parent and onSetChildren() to be consistent (3 has parents 0 and 2).
```
I have the same issue when using both `Component` and `PureComponent`.
I do _not_ see the error when using `ReactTestUtils.renderToDocument()`.
**If the current behavior is a bug, please provide the steps to reproduce and if possible a minimal demo of the problem via https://jsfiddle.net or similar (template: https://jsfiddle.net/reactjs/69z2wepo/).**
Could not figure out how to use `ReactTestUtils` in that JSFiddle example even though react-with-addons is loaded.
Here's the code I used in my project to reproduce it:
``` jsx
import React, { Component, PureComponent } from 'react'
import { render } from 'react-dom'
import ReactTestUtils from 'react-addons-test-utils'
class Root extends Component {
componentWillMount() {
this.testBreak()
setTimeout(this.testBreak.bind(this), 10)
}
testBreak() {
var i = 0, l = 50
while(i <= l) {
this.setState({ test: `this will break testing ${++i}` })
}
}
render() {
return (
<div>
{/* You need more than one nested element for this to break */}
<div>First div</div>
<div>Second div</div>
</div>
)
}
}
render(
<Root />
, document.getElementById('root'))
class TestReactElement extends PureComponent {
render() { return (
<div>{this.props.name}</div>
)}
}
const testRender = (name) => {
let renderer = ReactTestUtils.createRenderer()
renderer.render(<TestReactElement name={name} />)
}
testRender('Test 1')
testRender('Test 2')
testRender('Test 3')
```
It works fine so long as I run `testRender()` once. As soon as it's run 2 or more times, the error pops up.
In my project, TAP output is being listened to from a `console.log` wrapper and `state` is being updated when a new one comes in. So while my reproducible example is contrived, it's real-world bug when running tests in my project.
**What is the expected behavior?**
The expected behavior is that no error occurs in the console when running multiple shallow test renders while at the same time React is updating the state of an element rendered to the DOM.
**Which versions of React, and which browser / OS are affected by this issue? Did this work in previous versions of React?**
`15.3.0` and `15.3.2`. Windows 10 64-bit v1607 (and updates up to October 10th, 2016). I've not tried this in previous versions of React. Verified it displays the error in Chrome `54.0.2840.50 beta-m (64-bit)` and Firefox `50.0b5`.
|
https://github.com/facebook/react/issues/7927
|
https://github.com/facebook/react/pull/8097
|
e3688d1fa3a69a2a893b79759eb759ac9a51a8a9
|
6eebed0535a5e28effb7200783cede7a4bdab7ec
| 2016-10-10T10:07:15Z |
javascript
| 2016-10-27T09:31:10Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 7,925 |
["src/renderers/dom/fiber/ReactDOMFiber.js", "src/renderers/dom/fiber/__tests__/ReactDOMFiber-test.js", "src/renderers/noop/ReactNoop.js", "src/renderers/shared/fiber/ReactFiberCommitWork.js", "src/renderers/shared/fiber/ReactFiberReconciler.js", "src/renderers/shared/fiber/ReactFiberRoot.js", "src/renderers/shared/fiber/ReactFiberScheduler.js", "src/renderers/shared/fiber/__tests__/ReactIncremental-test.js"]
|
[Fiber] Umbrella for remaining features / bugs
|
This is an umbrella issues for remaining fiber issues. More can be found by running the unit tests with the `useFiber` flag in `ReactDOMFeatureFlags` turned on
See #8830 for additional tasks beyond the scope of the initial Fiber release
<details>
<summary>Phases 1–6</summary>
**Phase 1: Infrastructure**
- [x] Set up infra to know which tests are failing @spicyj
- [x] ReactTestUtils features
- [x] reactComponentExpect @gaearon #8258
- [x] scry etc. @gaearon #8257
**Phase 2: Smaller / Initial Tasks**
- [x] ReactDOMFiber.render - Simple synchronous case. [@sebmarkbage] #8029
- [x] String refs using owner. [@acdlite] #8099
- [x] Pass the correct previous state to componentDidUpdate. #7941
- [x] componentWillMount, componentWillReceiveProps, componentWillUpdate life-cycles. [@sebmarkbage] #8015
- [x] componentDidUpdate should not fired if shouldComponentUpdate returns false. [@sebmarkbage] #8028
- [x] Instance should be recreated when already started work gets resumed (no componentWillReceiveProps without didMount). [@sebmarkbage] #8015
- [x] Fix module pattern in mountIndeterminateComponent. [@sebmarkbage] #8015
- [x] findDOMNode. #8083 [@sebmarkbage]
- [x] Switching between types doesn't track deletions (updateSlot doesn't return null). [@sebmarkbage / @gaearon] #8085
- [x] PureComponent flag. [@acdlite] #8118
- [x] Ensure that scheduling works synchronously by default but batched in DOM events. (Initial default priority.) [@acdlite] #8127
- [x] unstable_batchedUpdates [@acdlite] #8127
- [x] setState in componentDidMount/componentDidUpdate should always be synchronous at the end of the batch. Even if work is deferred. [@acdlite] #8206
- \- [x] Make sure setState + error boundaries work in it [@gaearon / @acdlite] #8193
- [x] Unmount failed trees before attempting to recover [@acdlite] #8304
**Phase 3: Larger Tasks**
- [x] Allow testing Fiber in Facebook web codebase [@spicyj]
- [x] Error boundaries. [@gaearon]
- [x] Context [@gaearon #8272]
- [x] DOM attributes/properties update. [@sebmarkbage]
- [x] DOM events - Plug in the event system. [@sebmarkbage]
- [x] renderContainerIntoSubtree - Preserve state and such. [@spicyj #8368]
- [x] [fb] Bug: Switching from text children to regular children won't clear the old text content. (Not an easy fix since the side-effect of the parent is scheduled after the child insertions.) [@acdlite / @sebmarkbage] #8331
- [x] Recover from errors during commit phase, including errors thrown from host environment. Should only use a single try-catch block. [@acdlite]
- [x] [fb] SVG mode switch (based on if a parent is SVG or HTML). [@gaearon #8417]
- [x] [fb] setState behavior when called in all possible life-cycles including sibling life-cycles and parents and children. [@acdlite]
**Phase 4: Uncovered Bugs**
- [x] Root container is wrong in commit phase (https://github.com/facebook/react/pull/8568#issuecomment-266848256) [@sebmarkbage] #8607
- [x] Update state/props object pointers even if bailout happens. This will currently cause an unnecessary componentDidUpdate. Solve that somehow. [@acdlite] #8655
- [x] ~~Validate if inputValueTracking does indeed need to be cleaned up on unmount~~ I don't think it needs to but we should revert anyway
- [x] autoFocus doesn't work. Todo in ReactDOMFiberComponent. [@bvaughn] #8646
- [x] [fb] When the last effect renders, there's an infinite loop. [@acdlite]
- [x] [fb] Defer clearing of initial nodes until work is actually begun. (Ensure that unmounting and rerendering in the same batch works.) [@acdlite]
- [x] Updates on different roots in componentWillMount (ReactUpdates-test) [@acdlite]
- [x] Refs should update even if shouldComponentUpdate returned false. However, componentDidUpdate should not fire. [@acdlite] #8685
- [x] [fb] Make top level render and unmount synchronous by default even when they're in a batch. (Even for updates? No, not for updates.) [@acdlite] #8634
- [x] [fb] Uncaught TypeError: Cannot read property 'children' of null https://github.com/facebook/react/blob/705c9bcfd2745749a603b72e04528095021a673b/src/renderers/shared/fiber/ReactFiberBeginWork.js#L213 (probably due to context change, based on the caller) We should probably either avoid rerendering or set pendingProps to memoizedProps in this case. [@sebmarkbage] #8613
- [x] Does replaceState remove any previous callbacks scheduled? [@acdlite]
- [x] Root fiber should use `updateQueue` instead of `pendingProps`. [@acdlite]
- [x] [fb] Ensure that we get the current props for controlled components in restoreStateIfNeeded. [@spicyj] #8443
- [x] [fb] Iterable children. (E.g. immutable.js / Set etc.) [@gaearon #8446]
- [x] [fb] Reuse HostText when bailing out [@sebmarkbage] #8371
- [x] [fb] Ensure that renderSubtreeIntoContainer twice enables the context to pass through the middle subtree. I.e. two nested layer with a single context provider at the very top. [@spicyj] #8407
- [x] Make HostContainer append/insert/remove child instead of always updating all. Detach .return when a child gets unmounted (removed). [@sebmarkbage] #8400
- [x] Handle selection restoration #8401 #8353
- [x] Swap out the active Fiber when a host node is updated - including if listeners change so that we can know which listener is active. _This needs to remain even when prepareUpdate is fixed._
- [x] Remove traversal of `.output`. #8406
- [x] findDOMNode and findAllInRenderedTree are broken. If an insertion happens and we happen to look at the previous work in progress, then it will not have any deletes or removes. So it looks like it is current. I assume the same can happen for a deletion after it is done. [@sebmarkbage] #8450
- [x] ~~[fb] Batch different roots together into one commit (ReactUpdates-test)~~ Won't fix. Use Portals instead. #8508
- [x] Error handling when an error is thrown in detachRef callbacks. [@acdlite]
- [x] Ensure that if we walk up .return in TreeTraversal we always get the current Fiber for props, or possibly just walk up the parentNode tree instead. [@sebmarkbage] #8491
- [x] ~~Ensure that when we switch on tag names, we cover mixed/upper case. E.g. `React.createElement('INPUT')`.~~ Won't fix. Instead, we'll warn if upper case is used on HTML tags. [@sebmarkbage] #8563
- [x] Some scheduling issue is causing ReactCompositeComponentNestedState-test to fail. It performs the first setState before the second one. Will any of the Task priority stuff fix this? (@acdlite)
- [x] [fb] Scheduling an update during render doesn't work. We mostly already warn if setState is called from render, but it might required that componentWillMount (and others) calling a callback which sets state on a different component works. Currently the complete phase resets both the updateQueue and pending priority. This can cause infinite loops to happen. Apparently still used. We need to either warn/log better and upgrade everyone or support it in Fiber. [@acdlite]
- [x] Top-level context push/pop isn't properly matched up; lots of tests fail if you patch in [this](https://gist.github.com/342ebd8df0d08143381dd38c01d13d15). #8568 [@bvaughn]
- [x] [fb] Reentrant mounting in synchronous mode [@spicyj #8623]
**Phase 5**
- [x] Feature flag to disable Fiber-only features [@acdlite] #8797
- [x] Polyfills or alternate paths for ~~`Map`, `requestAnimationFrame`~~ and `requestIdleCallback`. [@sebmarkbage] #8833
- [x] isMounted - technically not deprecated yet. #8083 [@sebmarkbage]
- [x] Top level render callbacks. Second argument to render. #8102 [@koba04]
- [x] Ensure that server rendering works using Stack when Fiber is enabled. [#8372]
- [x] Declarative portal API @gaearon #8386
- [x] Add tests for event bubbling (This will need to track #8117 for changes to how top level event listeners get attached.)
- [x] React Test Renderer support. (@iamdustan #8628)
- [x] React ART support. #8521 (@bvaughn)
- [x] Land basic React Native support in React repo. #8560 (@bvaughn)
- [x] Mark root renders during mount (perf testing) (see tests for "marks top-level updates") [@bvaughn] #8687
- [x] Merge callbackList onto the updateQueue instead of a separate field. [@acdlite] #8728
- [x] Separate priority level for state updates. #7457 [@acdlite]
- [x] Fix coroutine issue where Fiber is passed in user space and could become stale. [@sebmarkbage] #8840
- [x] DOM Dev Tools [@gaearon]
**Phase 6: Unit tests and known bugs push**
- [x] [fb] Fix `findDOMNode` bug when used in certain cases from `componentWillUnmount`. [@sebmarkbage, @acdlite] #8897
- [x] [fb] Provide ability to block event bubbling in portals [@spicyj]
- [x] Unit tests: ReactFragment-test invariants [@acdlite] #8890
- [x] Unit tests: ReactDOMProduction-test invariant [@acdlite] #8907
- [x] Unit tests: ReactEmptyComponent-test invariant [@acdlite] #8908
- [x] Ensure we replace `throw new Error` with `invariant` calls and they have sensible messages [@acdlite] #8926
- [x] Unit tests: ReactMount-test [@acdlite] #8919
- [x] Unit tests: ReactMultiChild-test (remove Map as children support in Stack) [@acdlite] #8911
- [x] Unit tests: ReactMultiChildText-test [@acdlite]
- [x] RN: Fix event bubbling regression. [@spicyj/@bvaughn]
- [x] Make sure Babel doesn't generate bad code (e.g. IIFE for `let`/`const`), maybe fork Babel plugin to throw [@spicyj]
- [x] RN: Land React update in fbsource with Fiber disabled [@sebmarkbage]
- [x] Verify Fiber works in IE on Facebook (Map / Set: open IE10/11 and make sure it works) [@spicyj]
- [x] [fb] Figure out Ads Image Cropper issue [@spicyj]
- [x] Unit tests: ReactDOMComponent-test [@acdlite] #8948
- [x] Unit tests: ReactComponentLifeCycle-test warnings (ex: fDN in render) [@acdlite] #8949
- [x] Allow assigning this.state in cWRP with warning [@acdlite] #9040
- [x] Unit tests: ReactStatelessComponent-test warning [@trueadm] #9043
- [x] Unit tests: refs-test [@trueadm] #9045
- [x] Make a new manual propTypes checker. [@acdlite] #9004
- [x] Unit tests: ReactCompositeComponent-test warnings (ex: setState in render) [@acdlite] #8950
- [x] Ensure "break on all/uncaught exceptions" works as expected in browsers [@acdlite] #8961
- [x] Investigate what modules and unit tests fail if we remove Stack [@trueadm] #9069
- [x] Quick investigation into FB require.js perf [@trueadm]
- [x] Quick investigation into Fiber bundle perf, cutting out extra dev-only code [@trueadm] #9096
- [x] [fb] Move .hidden check into hostConfig (https://fburl.com/m8juo29d, https://fburl.com/538fgu7g)
</details>
-----
**Phase 6.1: 15.5 Release**
- [ ] Warn (throw?) when doing an update on a container that was manually emptied outside of React (in stack, this mounted a brand-new tree; in fiber, it tries to apply an update and usually fails) [~~@keyanzhang~~, @sebmarkbage took over this] (blocks beta)
- [x] Extract propTypes into a separate npm package. Get `prop-types` package name. [@acdlite]
- [x] Unit tests: ReactContextValidator-test -- Pass the correct previous `context` to `componentDidUpdate` (or deprecate feature and remove test). [@bvaughn] #8631
- [x] [fb] Better fix to ReactART.Text.
- [x] Update warning when calling propTypes directly to refer user to `checkPropTypes` API
- [x] Docs updates for test utils and removed add-ons [@flarnie]
**Phase 6.2: React Native Fiber**
- [x] Reimplement Shallow Renderer without Stack dependencies [@lelandrichardson/@trueadm] #8982
- [ ] RN: ensure we keep all important invariants (such as that [text must be wrapped](https://github.com/facebook/react-native/blob/9f3e928f8335a51cd87593a150726e564afd2fa4/Libraries/Renderer/src/renderers/native/ReactNativeTextComponent.js#L32-L36) in `<Text>`) (does not block anything but we should do this)
- [ ] Ensure that we call the FB specific warning module instead of the `fbjs/lib/warning` one in RN. [@spicyj] (doesn't block anything OSS, internal to FB)
- [x] Convert propTypes callers to use checkPropTypes. [@acdlite]
- [x] Deploy forwarding modules to react/lib/ReactCurrentOwner, react/lib/ReactComponentTreeHook, fbjs/lib/invariant, fbjs/lib/warning, fbjs/lib/ExecutionEnvironment, fbjs/lib/performanceNow, fbjs/lib/emptyObject, fbjs/lib/emptyFunction, fbjs/lib/shallowEqual. So that we can keep using providesModule in this repo until we have flat bundles. Or better yet: Do the inverse and switch to using CommonJS naming convention in www. [@sebmarkbage, @bvaughn]
- [ ] See if we can codemod `createStrictPropTypeChecker` away and use `exactShape` instead. (doesn't block OSS release)
- [x] Release a new version of react-redux whose peer is compatible with 16-alpha. Update RN to use it. https://github.com/reactjs/react-redux/pull/629
- [x] Enable `prepareNewChildrenBeforeUnmountInStack` feature flag (one week after landing sync).
- [x] ~~Codemod Flow errors yielded by properly typing `findNodeHandle`.~~ (we added types and `$FlowFixMe` comments; good enough for now)
- [x] Redesign the host component "type", the type of host component refs and integration with `NativeMethodsMixin` to be upgrade compatible. [@bvaughn]
- [x] Unblocks: RN Dev Tools [@gaearon] https://github.com/facebook/react-devtools/pull/528
- [x] Unblocks: RN Inspector (see _known issues_ section of PR #8560) [@gaearon] https://github.com/facebook/react-native/pull/12421
- [ ] Make sure instanceProps in `ReactNativeComponentTree` doesn't leak. (@sebmarkbage, @bvaughn)
**Phase 6.3: Flat Bundles/Rollup**
- [x] Make React Perf work with Fiber (ReactPerf-test) [@gaearon] (Decided not to in favor of deeper timeline integration)
- [x] Unit tests: ReactComponentTreeHook-test [@gaearon]
- [x] Unit tests: ReactHostOperationHistoryHook-test [@gaearon]
- [x] Make a list of all internal React modules required by FB code
- [x] Come up with a better strategy for dealing with DEV code and make sure we don't ship it in prod
- [x] Change www sync script https://fburl.com/juffnc9d to run rollup using commonjs source
- [x] Shimming the required files (ex: ReactDOMInjection, warning)
- [x] Exporting a single bundle that exposes the internals we need
- [x] Make fowarding modules for ReactDOM/etc as well as internal modules
- [ ] Convert OSS repo to ES6 modules (does not block release; nice-to-have follow up)
- [x] Make separate dev and non-dev build in one file for www (`if (__DEV__) { ... all of React ... } else { ... all of React }`)
- [x] Convert OSS repo to rollup
- [x] Replace current Grunt/Gulp build system with a unified build system (maybe purely using Rollup)
- [ ] Add Closure Compiler with ADVANCED and deal with mangling properly (does not block release; nice-to-have follow up)
**Phase 7: Async-Compatible** (does not block release)
- [ ] Fix incremental regression (if it is a regression)
- [ ] [fb] Fix resumeMountClassInstance so that `newProps` gets passed to construct class. Otherwise `this.props` can be `null`.
- [ ] RN: Aborting async work or unmounting a tree due to an error leaks native views
- [ ] Add cross-renderer support for portals (necessary for async ART)
- [ ] ReactDOMFiber.render - Return the root instance from render even if priority is deferred. Test and cover incremental cases
- [ ] Decide on replacing priorities with deadlines to fix starvation
**Phase 8: Server-Side Rendering**
- [x] What are we doing?
- [x] Unit tests: ReactRenderDocument-test (SSR) -- Reviving server rendered markup (including rendering into `document` and shadow DOM containers)
**Phase 9: Improvements** (does not block release)
- [ ] The `getPublicInstance` internal API should consistently be used before exposing any stateNodes.
- [ ] See https://github.com/facebook/react/pull/8751#discussion_r119161506
- [x] Key-warnings on fragments. [@flarnie]
- [x] Don't mark "update" effect on newly mounted components unless they have a componentDidMount life-cycle.
- [ ] Add support to Enzyme for Fiber (https://github.com/airbnb/enzyme/pull/1007)
- [ ] Currently bailing out on equal props/state equality means that we still copy all of our children, and traverse all of them. Meaning that a parent with many children will still do a lot of work even if it bails out / "skips". If "pendingWorkPriority" on referred only to children, we would know if any children has additional work. [@acdlite] #8716
- [x] Unit tests: ReactDOMTextComponent-test -- .normalize() case
- [ ] Reconsider the ContentReset flag. Is there a cleaner solution for clearing out text content / innerHTML? Such as changing the insertion order
- [ ] Reconsider storing masked contexts on instances
- [ ] Revert inputValueTracking and replace with different comparison strategy. getter/setters is slow and complicated. Subscription model isn't helpful for Fiber's goals. The principle of over listening to lots of events and diffing the value is sound. <s>We haven't shipped in OSS yet so not too hard to fix</s> -- This will probably get pushed back to 16/17
- [ ] Fiber drops the lazy DOM insertion order. Is there a solution that isn't slow in IE11? Is it worth fixing or should we leave IE11 slow and wait for a fix in the browser? (related: #8417)
- [ ] Follow up from previous task (https://github.com/facebook/react/pull/8728#discussion_r95643611): use a linked list for `callbackList` [@acdlite] https://github.com/facebook/react/pull/8752
- [ ] Ensure that it is possible to resume "incomplete" parents without rerendering. A shallow reuse. Store it on memoizedProps. Resuming "completed" parents is already possible. [@acdlite] #8716
- [ ] Move callback invariant closer to usage as requested in https://github.com/facebook/react/pull/8639#issuecomment-271944608
- [ ] Replace some invariants with warnings to help code size https://github.com/facebook/react/pull/8677#issuecomment-270335705
|
https://github.com/facebook/react/issues/7925
|
https://github.com/facebook/react/pull/8102
|
2ba571c24618b97049898532e2e220897f9b4c43
|
2ef12084e4bdaca980bee0f404cada005519ffbb
| 2016-10-10T08:04:34Z |
javascript
| 2016-10-26T17:19:19Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 7,880 |
["grunt/tasks/npm-react-native.js", "grunt/tasks/npm-react-test.js", "packages/react-test-renderer/package.json"]
|
ReactTestRenderer 15.4.0-rc.1 missing files
|
Installing [email protected] from npm just brings down an index.js file, the referenced lib folder does not exist.
|
https://github.com/facebook/react/issues/7880
|
https://github.com/facebook/react/pull/7888
|
2a2dd7689f90bb18d3e732f068e7b030b2d2cc8f
|
077d660a2727e3d3d6629d3a62693f8509100139
| 2016-10-05T06:45:16Z |
javascript
| 2016-10-05T22:32:38Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 7,856 |
["src/isomorphic/classic/element/ReactElementValidator.js", "src/isomorphic/classic/element/__tests__/ReactElementValidator-test.js", "src/isomorphic/modern/element/__tests__/ReactJSXElementValidator-test.js"]
|
Show component stack for invalid type warning during element creation
|
When you do `<Foo />` but Foo is not a valid string or function, we show an error message:
https://github.com/facebook/react/blob/92c84a6f37245f4fbe3ce5dc50845856201637fa/src/isomorphic/classic/element/ReactElementValidator.js#L196
We include the owner name ("Please check the render method of Bar.") but not the full parent stack which we show for some of our other warnings like key warnings and prop types. We should call into ReactComponentTreeHook to get the full stack if possible.
|
https://github.com/facebook/react/issues/7856
|
https://github.com/facebook/react/pull/8495
|
fbfecd13ce03d0386ea93b73cc645ccc09bbbed3
|
fb7e49439feecd401029f24bf401f22b235c3947
| 2016-10-03T23:14:48Z |
javascript
| 2017-01-09T15:26:01Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 7,803 |
["src/renderers/dom/shared/ReactDOMComponentTree.js"]
|
Upgrading to React v15.2.0 causes my development build to crash IE9
|
My application crashes IE9 when I upgrade to React v15.2.0! It works fine with v15.1.0 and all earlier versions.
My diff looks like this
```
package.json
- "react-addons-perf": "15.1.0",
+ "react-addons-perf": "15.2.0",
- "react-addons-shallow-compare": "15.1.0",
+ "react-addons-shallow-compare": "15.2.0",
- "react-addons-test-utils": "15.1.0",
+ "react-addons-test-utils": "15.2.0",
- "react": "15.1.0",
+ "react": "15.2.0",
- "react-addons-css-transition-group": "15.1.0",
+ "react-addons-css-transition-group": "15.2.0",
- "react-addons-transition-group": "15.1.0",
+ "react-addons-transition-group": "15.2.0",
- "react-dom": "15.1.0",
+ "react-dom": "15.2.0",
```
Below is a GIF of the crash, I recorded this using BrowserStack (Win7, IE9). I start the recording with Google loaded and dev tools open, I then navigated to my apps URL. As you can see IE9 just crashes hard with nothing logged to console. All other browsers load my app just fine, this only seems to crash IE9!
Did you guys drop support for IE9 in 15.2.0? If not, any ideas what might be causing this?
|
https://github.com/facebook/react/issues/7803
|
https://github.com/facebook/react/pull/8018
|
fa4710fe5136e5feceed09b54c2ff49d012b8c15
|
6ce8f1f93c1d8758fcaa819ab18c70a7dd65b6aa
| 2016-09-24T20:09:36Z |
javascript
| 2016-11-10T16:02:48Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 7,784 |
["src/renderers/dom/shared/HTMLDOMPropertyConfig.js", "src/renderers/dom/shared/__tests__/ReactDOMComponent-test.js"]
|
support slot attribute (ShadowDOM v1) on DOM elements
|
**Do you want to request a _feature_ or report a _bug_?**
feature ( I guess? )
**What is the current behavior?**
React doesn't support `slot` attribute which comes with shadowDom v1, so it will throw error and because of that, web component isn't working
code:
``` js
class App extends Component {
render() {
return (
<div className="App">
<div className="App-header">
<img src={logo} className="App-logo" alt="logo" />
<h2>Welcome to React</h2>
</div>
<p className="App-intro">
To get started, edit <code>src/App.js</code> and save to reload.
</p>
<div>
<fancy-tabs background hello="World!">
<button slot="title">Tab 1</button>
<button slot="title" selected>Tab 2</button>
<button slot="title">Tab 3</button>
<section>content panel 1</section>
<section>content panel 2</section>
<section>content panel 3</section>
</fancy-tabs>
</div>
</div>
);
}
}
```
rendered html with error
<img width="919" alt="screen shot 2016-09-21 at 2 05 47 pm" src="https://cloud.githubusercontent.com/assets/1223799/18717136/4d2fe7d2-801f-11e6-8205-795576144626.png">
**If the current behavior is a bug, please provide the steps to reproduce and if possible a minimal demo of the problem via https://jsfiddle.net or similar (template: https://jsfiddle.net/reactjs/69z2wepo/).**
**What is the expected behavior?**
Support new DOM attribute standard
**Which versions of React, and which browser / OS are affected by this issue? Did this work in previous versions of React?**
React 15.3.2 / Chrome 53 / OSX El Capitan
|
https://github.com/facebook/react/issues/7784
|
https://github.com/facebook/react/pull/8061
|
c1cca0c7d8e0cad2e4dfb88cdd3a095b8acc1c51
|
346d50e2ba0cccf25605791e9739ddb0cf5ddd04
| 2016-09-21T15:20:34Z |
javascript
| 2016-11-05T16:34:54Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 7,782 |
["docs/_posts/2013-06-05-why-react.md"]
|
jsfiddle from official website not working
|
**Do you want to request a _feature_ or report a _bug_?**
bug
**What is the current behavior?**
http://jsfiddle.net/fv6RD/3/ does not work, the code is not executed, there are two files loaded from the Facebook servers but they both return a 5xx error
**If the current behavior is a bug, please provide the steps to reproduce and if possible a minimal demo of the problem via https://jsfiddle.net or similar (template: https://jsfiddle.net/reactjs/69z2wepo/).**
Go to https://facebook.github.io/react/docs/why-react.html and then click the link at the bottom to go to https://facebook.github.io/react/blog/2013/06/05/why-react.html and then click on the link to jsfiddle in the text `Check out this jsFiddle to see an example of reconciliation in action.`
**What is the expected behavior?**
that there is some generated output in the result pane
**Which versions of React, and which browser / OS are affected by this issue? Did this work in previous versions of React?**
Chrome (latest stable, Mac OS), Safari (Mac OS)
|
https://github.com/facebook/react/issues/7782
|
https://github.com/facebook/react/pull/7783
|
00ac1797d75cea4b42d46c0161ee77acec5125f3
|
13692d59add7cfffe06e69f8d4fe2488d0a55514
| 2016-09-21T12:12:02Z |
javascript
| 2016-09-21T12:42:49Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 7,740 |
["scripts/fiber/tests-passing.txt", "src/renderers/shared/stack/reconciler/ReactCompositeComponent.js", "src/renderers/shared/stack/reconciler/ReactOwner.js", "src/renderers/shared/stack/reconciler/ReactReconciler.js", "src/renderers/shared/stack/reconciler/ReactRef.js", "src/renderers/testing/ReactTestMount.js", "src/renderers/testing/ReactTestRenderer.js", "src/renderers/testing/__tests__/ReactTestRenderer-test.js"]
|
`react-test-renderer` and refs
|
**Do you want to request a _feature_ or report a _bug_?**
Bug
**What is the current behavior?**
It's not possible to test component that use `ref` with the `react-test-renderer` utilitiesTesting: the refs are always `null`.
``` javascript
/* @flow */
import React from 'react';
export default class Foo extends React.Component {
/* the future refs */
bar;
componentDidMount() {
console.log(this.bar); // this.bar is null
this.bar.doThings() // So this fail
}
render() {
return (
<div ref={(c) => { console.log('ref cb', c); this.bar = c; }}> {/* The callback is call but, `c` is null*/}
<p>Hello World</p>
</div>
);
}
}
```
``` javascript
import React from 'react';
import renderer from 'react-test-renderer';
it('should have valide ref', () => {
const foo = renderer.create(<Foo />);
expect(foo.toJSON()).toMatchSnapshot();
});
```
**What is the expected behavior?**
The ref should be usable.
**Which versions of React, and which browser / OS are affected by this issue? Did this work in previous versions of React?**
- `[email protected]`
- `[email protected]`
- `[email protected]`
Only tested with these versions.
|
https://github.com/facebook/react/issues/7740
|
https://github.com/facebook/react/pull/8261
|
da021ca4aeb5d83414cabbe370b65412a41deb1a
|
e43aaab2547870bc80c0b778604c5ee55b1d87f0
| 2016-09-15T13:13:31Z |
javascript
| 2016-11-11T14:35:18Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 7,735 |
["src/renderers/art/ReactART.js", "src/renderers/dom/shared/ReactDOMComponent.js", "src/renderers/native/ReactNativeBaseComponent.js", "src/renderers/shared/stack/reconciler/ReactMultiChild.js", "src/renderers/shared/stack/reconciler/ReactOwner.js", "src/renderers/testing/ReactTestEmptyComponent.js", "src/renderers/testing/ReactTestMount.js", "src/renderers/testing/ReactTestReconcileTransaction.js", "src/renderers/testing/ReactTestRenderer.js", "src/renderers/testing/ReactTestTextComponent.js"]
|
Fix Flow issues introduced in #7716
|
We missed this because Flow wasn't being run properly on Travis but now that it is, it's clear that #7716 introduced a bunch of new Flow failures. Most of these shouldn't be too bad and I like to say "you break it you buy it" so this one is for you @Aweary :)
There were a couple other pre-existing Flow issues that @vjeux is going to fix, so mostly this should just be the TestRenderer issues.
|
https://github.com/facebook/react/issues/7735
|
https://github.com/facebook/react/pull/7736
|
864bc7b939ca8a93583087d1ff596cd8049b4162
|
c78464f8ea9a5b00ec80252d20a71a1482210e57
| 2016-09-14T22:09:12Z |
javascript
| 2016-09-15T16:44:56Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 7,718 |
["src/isomorphic/classic/element/ReactElement.js", "src/isomorphic/classic/element/__tests__/ReactElement-test.js", "src/isomorphic/classic/element/__tests__/ReactElementClone-test.js"]
|
Bug: False Positive "plain object" Warning
|
There is a bug in the logic for determining if `props` is a plain object. This bug results in a false positive, flooding the console with hundreds of instances of this warning: "Expected props argument to be a plain object. Properties defined in its prototype chain will be ignored.".
The code that's failing is in [ReactElement.js, Line 210 in the v15.3.1 branch](https://github.com/facebook/react/blob/v15.3.1/src/isomorphic/classic/element/ReactElement.js#L210) (line number is probably different in master)
> config.**proto** == null || config.**proto** === Object.prototype,
This fails when the props object is created in one iframe, but uses an instance of React from another frame. Each browser window/tab/frame has it's own instance of the Object constructor. For instance, `document.querySelector('iframe').contentWindow.Object === Object` always results in `false`.
Check out this [demo](https://jsfiddle.net/05ugxo5f/4/), which shows the false-positive warning message.
|
https://github.com/facebook/react/issues/7718
|
https://github.com/facebook/react/pull/7724
|
9eba80825f0931794e4553db1806e78de3a218de
|
b2297ae6c3117ca9b3e15aedaf73d9187eb88e6c
| 2016-09-13T17:42:15Z |
javascript
| 2016-09-14T02:15:07Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 7,645 |
["src/renderers/shared/stack/reconciler/ReactCompositeComponent.js", "src/renderers/testing/__tests__/ReactTestRenderer-test.js"]
|
String refs cause incorrect warning in ReactTestRenderer
|
If you attach a `ref` to an element using a string (`ref='foo'`) and use `ReactTestRenderer` you get the `"Stateless function components cannot be given refs"` warning
See https://github.com/facebook/react/issues/7371#issuecomment-237816352
This is because `getPublicInstance` returns `null` and `attachRef` requires that the instance be non-null in the warning invariant.
I think we can either return a simple instance instead of `null` or use some additional check in `attachRef` to see if we're dealing with `ReactTestRenderer`. The first option seems reasonable, given the comment in the current code:
``` js
ReactTestComponent.prototype.getPublicInstance = function() {
// I can't say this makes a ton of sense but it seems better than throwing.
// Maybe we'll revise later if someone has a good use case.
return null;
};
```
cc @spicyj
|
https://github.com/facebook/react/issues/7645
|
https://github.com/facebook/react/pull/7658
|
6a525fdc4c90303c4e14c3aee26f8faa50e77c27
|
06ea71d3fdd614e12c659c2f08f371c6b01fe880
| 2016-09-02T16:31:14Z |
javascript
| 2016-09-06T08:49:20Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 7,630 |
["fixtures/dom/src/components/fixtures/input-change-events/RadioNameChangeFixture.js", "fixtures/dom/src/components/fixtures/input-change-events/index.js", "packages/react-dom/src/__tests__/ReactDOMInput-test.js", "packages/react-dom/src/client/ReactDOMFiberComponent.js", "packages/react-dom/src/client/ReactDOMFiberInput.js"]
|
Radio buttons are not correctly checked when using multiple lists of radio buttons
|
**Do you want to request a _feature_ or report a _bug_?**
Bug
**What is the current behavior?**
When there are two lists of radio buttons and they are conditionally shown the correct radio button isn't always checked. In the [provided JSFiddle](https://jsfiddle.net/atimmer/krphcc8e/3/) I have created two lists and put a button before them to toggle between the two lists. You can see when switching to the second list of radio buttons none of the radio buttons is checked even though the second one should be checked based on the state.
**What is the expected behavior?**
The correct radio button should be checked. The following JSFiddle shows the behaviour with React 15.3.0 where it still worked: https://jsfiddle.net/atimmer/Lzjs56sn/1/.
**Which versions of React, and which browser / OS are affected by this issue? Did this work in previous versions of React?**
✅ React 15.3.0: It works.
❌ React 15.3.1: It is broken.
|
https://github.com/facebook/react/issues/7630
|
https://github.com/facebook/react/pull/11227
|
3f405da614694f333a0315d3c769d144872efe28
|
70abda5b92b8aae4dc0255a8ca00c2025ce05c3d
| 2016-09-01T14:18:26Z |
javascript
| 2017-11-19T15:44:35Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 7,544 |
["src/renderers/dom/client/wrappers/ReactDOMInput.js", "src/renderers/dom/client/wrappers/__tests__/ReactDOMInput-test.js"]
|
Warn if checkbox changes controlledness
|
We should be warning if a controlled component becomes uncontrolled, or vice versa.
I thought this was fixed in https://github.com/facebook/react/issues/5821, but apparently it was either not fixed for checkbox components, or maybe I subsequently broke it in https://github.com/facebook/react/pull/7003 (or similar). Regardless, we should fix it.
Demo of case that should warn: https://jsfiddle.net/s4742gkk/
|
https://github.com/facebook/react/issues/7544
|
https://github.com/facebook/react/pull/7603
|
5331fd00bc3936d19916d934a0fb0e549499ca4e
|
ba84b5b0a7047115e60a8fc68d563782a5756fba
| 2016-08-23T00:29:10Z |
javascript
| 2016-08-29T18:23:30Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 7,492 |
["src/umd/ReactUMDEntry.js", "src/umd/ReactWithAddonsUMDEntry.js"]
|
ReactComponentTreeHook should not end up in the production build
|
I noticed it’s getting included in production builds again (but it shouldn’t be).
Filing this to track it because we specifically fixed it recently, and we’re regressing on this.
Likely related to #7168.
|
https://github.com/facebook/react/issues/7492
|
https://github.com/facebook/react/pull/7653
|
9d9af63a6a9276390e3acafc25893d518cd59eba
|
a09d158a7cc66634ae5c92649193c83fa69963c1
| 2016-08-13T20:33:21Z |
javascript
| 2016-09-03T21:32:20Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 7,482 |
["grunt/config/browserify.js"]
|
Master UMD builds don’t work as CommonJS using Webpack alias config
|
Not sure what’s up yet but putting this up so I don’t forget to check it later.
In the past, I could put this into Webpack config:
``` js
resolve: {
alias: {
react: 'react/dist/react',
'react-dom': 'react-dom/dist/react-dom',
},
}
```
Many people do that to improve build times.
This works with 0.14.x and 15.3.0 but not in master.
Most likely related to recent @sebmarkbage changes.
It fails like this:
<img width="760" alt="screen shot 2016-08-12 at 14 55 55" src="https://cloud.githubusercontent.com/assets/810438/17624533/e8b7c366-609c-11e6-8fe6-9adb951f12a8.png">
Seems like it claims to be a UMD shim but assumes global environment (whereas a true UMD shim would detect environment and fallback to `require` when available).
cc @zpao
|
https://github.com/facebook/react/issues/7482
|
https://github.com/facebook/react/pull/7840
|
f88c1d38bef076a25fad3f55cff9374307e3b750
|
92c84a6f37245f4fbe3ce5dc50845856201637fa
| 2016-08-12T13:57:22Z |
javascript
| 2016-10-03T22:49:10Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 7,424 |
["src/isomorphic/classic/element/ReactElement.js", "src/renderers/dom/server/__tests__/ReactServerRendering-test.js", "src/renderers/shared/ReactDebugTool.js", "src/renderers/shared/hooks/ReactChildrenMutationWarningHook.js", "src/renderers/shared/stack/reconciler/__tests__/ReactComponent-test.js"]
|
False-positive mutation warning
|
Given the following test case
``` jsx
function TestCmp() {
return <span>{NaN}</span>;
}
ReactDOM.render(<TestCmp />, document.getElementById('app'));
```
Produces the following warning:
```
Warning: Component's children should not be mutated.
in span (created by TestCmp)
in TestCmp (created by Unknown)
```
This is probably caused by [this mutation check](https://github.com/facebook/react/blob/328fc75bc910032ff37dc9d93429fb14b476bd0d/src/renderers/shared/hooks/ReactChildrenMutationWarningHook.js#L34), which was introduced in 15.3.0, but it doesn't work for NaN, because NaN != NaN
|
https://github.com/facebook/react/issues/7424
|
https://github.com/facebook/react/pull/7455
|
616e468987fc655bf5e7bc62b97fc86d3b5917e7
|
38c4ade6cc2588e28090385ec989997363bcb782
| 2016-08-04T12:52:04Z |
javascript
| 2016-09-13T13:25:31Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 7,406 |
["src/isomorphic/classic/element/ReactElement.js", "src/renderers/dom/server/__tests__/ReactServerRendering-test.js", "src/renderers/shared/ReactDebugTool.js", "src/renderers/shared/hooks/ReactChildrenMutationWarningHook.js", "src/renderers/shared/stack/reconciler/__tests__/ReactComponent-test.js"]
|
Memory leak in React 15.3.0 non-production server side rendering
|
A memory leak appears to have been introduced to server side rendering in React 15.3.0 when `NODE_ENV != production`.
With `NODE_ENV=development`, the size of the heap grows proportionally to the number of requests received. This was confirmed in both Node v6.3.0 and v0.12.3. Also it was **confirmed to work correctly in React 15.2.0**.
_To reproduce_: Start the app configured for server-side rendering with `NODE_ENV=development` (or just not production). Inspect the V8 heap before and after hitting your server with a number of requests. I used [heapdump](https://github.com/bnoordhuis/node-heapdump) to take the snapshots and [wrk](https://github.com/wg/wrk) to hit my app with a bunch of requests. Load the snapshots into chrome dev tools profiler. Here is what I see:
<img width="1278" alt="screen shot 2016-08-02 at 12 47 34 pm" src="https://cloud.githubusercontent.com/assets/7452924/17343291/b8922606-58b1-11e6-9891-b6dbc22edc29.png">
The culprit appears to be the [react children mutation warning](https://github.com/facebook/react/blob/328fc75bc910032ff37dc9d93429fb14b476bd0d/src/renderers/shared/hooks/ReactChildrenMutationWarningHook.js#L18). `onBeforeComponentMount` saves a reference to a given React Element and all of its props in the `elements` object, but the cleanup steps `onComponentHasMounted` or `onComponentHasUpdated` are never called on the server and thus the `elements` object is ever growing.
**Which versions of React, and which browser / OS are affected by this issue? Did this work in previous versions of React?**
react - 15.3.0
react-dom - 15.3.0
node - 6.3.0 || 0.12.3
express - ^4.0.0
|
https://github.com/facebook/react/issues/7406
|
https://github.com/facebook/react/pull/7455
|
616e468987fc655bf5e7bc62b97fc86d3b5917e7
|
38c4ade6cc2588e28090385ec989997363bcb782
| 2016-08-02T20:26:19Z |
javascript
| 2016-09-13T13:25:31Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 7,349 |
["src/renderers/dom/__tests__/ReactDOMProduction-test.js", "src/renderers/shared/ReactDebugTool.js", "src/renderers/shared/__tests__/ReactPerf-test.js", "src/renderers/shared/stack/reconciler/ReactCompositeComponent.js"]
|
Did not expect componentDidMount timer to start while render timer is still in progress for another instance
|
Not sure this should be filed as a bug, or if this is intended behaviour, in which case it would be more of a feature request, for something that we lost since the last few versions.
Which is more specific error logging.
Since the last few version this error shows up too many times, without any usable stack-trace:
`There is an internal error in the React performance measurement code. Did not expect componentDidMount timer to start while render timer is still in progress for another instance.`
This mainly occurs when a prop or state is not defined. This wouldn't be a big problem if I could stacktrace it, but in this case I cannot lead it back to any related component, I have to go off on what I changed last.
I do use `redux` and the stacktrace does go back to createStore and then connect, but this should not be related since this error shows up regardless of which element has an undefined property.
It is also often accompanied by:
`Uncaught TypeError: Cannot read property '_currentElement' of null`
How do I debug this error? Or are there plans to improve this?
|
https://github.com/facebook/react/issues/7349
|
https://github.com/facebook/react/pull/7548
|
0a248ee7b953d5c207b87d2840bd4295891760e2
|
a229cdba7fd00799d82c1bab23704220d47e6331
| 2016-07-25T14:57:30Z |
javascript
| 2016-08-24T18:18:32Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 7,337 |
["docs/_posts/2016-07-22-create-apps-with-no-configuration.md", "docs/img/blog/create-apps-with-no-configuration/compiled-successfully.png", "docs/img/blog/create-apps-with-no-configuration/compiled-with-warnings.png", "docs/img/blog/create-apps-with-no-configuration/created-folder.png", "docs/img/blog/create-apps-with-no-configuration/failed-to-compile.png", "docs/img/blog/create-apps-with-no-configuration/npm-run-build.png"]
|
Stop using imgur for images in blog post
|
https://facebook.github.io/react/blog/2016/07/22/create-apps-with-no-configuration.html is using imgur for image hosting. We should just put those on our site, like we have for previous posts.
While making sure we maintain control, it will also ensure that those images are served over HTTPS (which they aren't currently). We could fix that by updating the img srcs but we should probably just host them anyway.
<img width="355" alt="image" src="https://cloud.githubusercontent.com/assets/8445/17066395/c3d3c6aa-4ffb-11e6-8179-871b4f6cf699.png">
|
https://github.com/facebook/react/issues/7337
|
https://github.com/facebook/react/pull/7339
|
c0b7d818727af75b76349f41f2c11530ef5a6902
|
7614c12ed71e7a76128da62e8eac7835502b6626
| 2016-07-22T18:03:24Z |
javascript
| 2016-07-22T19:36:07Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 7,317 |
["docs/docs/addons-animation.md"]
|
ReactCSSTransitionGroup's transition*Timeout props are required but undocumented
|
When a `ReactCSSTransitionGroup` is used without timeout props, it [warns](https://github.com/facebook/react/blob/master/src/addons/transitions/ReactCSSTransitionGroup.js#L28-L33):
> Warning: Failed propType: transitionLeaveTimeout wasn't supplied to ReactCSSTransitionGroup: this can cause unreliable animations and won't be supported in a future version of React. See https://fb.me/react-animation-transition-group-timeout for more information. Check the render method of `Constructor`.
for each missing timeout prop. However, that URL simply points to [the Getting Started section of the Animation page](https://facebook.github.io/react/docs/animation.html#getting-started). The code listed there does use timeout props, but nothing on the page explains what those props do or why they're necessary. In fact, the word "timeout" only appears in the names of those props in code examples.
Could we get some explanation in the docs, please?
|
https://github.com/facebook/react/issues/7317
|
https://github.com/facebook/react/pull/8124
|
b3b13919c7f86fc42eeb20693294e031754c0d30
|
e1b140225a5881d831e3ac7e9128106a27039fec
| 2016-07-20T14:15:35Z |
javascript
| 2016-10-27T13:21:34Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 7,302 |
["src/isomorphic/classic/types/ReactPropTypes.js", "src/isomorphic/classic/types/__tests__/ReactPropTypes-test.js"]
|
PropTypes: More informative warning in createElementTypeChecker
|
**Do you want to request a _feature_ or report a _bug_?**
Enhancement
**What is the current behavior?**
Element type checker is not as informative as the other ones.
**What is the expected behavior?**
It would be nice to provide the actual type of the prop passed similarly to other type checkers. I.e., instead of
> Invalid prop `Prop` supplied to `Component`, expected a single ReactElement.
say
> Invalid prop `Prop` of type `type` supplied to `Component`, expected a single ReactElement.
**Which versions of React, and which browser / OS are affected by this issue? Did this work in previous versions of React?**
Seeing this on 15.2.1 but I think it's also applicable to previous versions.
I know the general intention is to move away from prop types, but if there isn't a specific reason for the current behavior I can look into making this more pretty.
|
https://github.com/facebook/react/issues/7302
|
https://github.com/facebook/react/pull/7319
|
5e3959e071b0750713bddf13e05319cd272a7b05
|
8bcea5310e4ac21c7c4f0552c01a270c4746b08b
| 2016-07-19T09:58:44Z |
javascript
| 2016-07-20T18:16:55Z |
closed
|
facebook/react
|
https://github.com/facebook/react
| 7,263 |
["src/renderers/dom/shared/HTMLDOMPropertyConfig.js"]
|
support for referrerPolicy attribute
|
`Warning: Unknown prop `referrerPolicy` on <a> tag.`
`referrerPolicy` is defined by the WHATWG HTML for e.g. `a` elements (https://html.spec.whatwg.org/multipage/semantics.html#the-a-element). Also see https://w3c.github.io/webappsec-referrer-policy/
|
https://github.com/facebook/react/issues/7263
|
https://github.com/facebook/react/pull/7274
|
f02cbba9fd764540d631a326aaf11f9cab1bdfb1
|
cccef3c68310df3bd611df2a7b98a530645c67c0
| 2016-07-13T08:46:58Z |
javascript
| 2016-07-13T18:56:57Z |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.