Summary:
These are bad since the app might not be foregrounded anymore and it also keeps the timing module awake. We could fix the latter but the former would require someone adding AlarmManager support.
Open to better/more helpful ideas for the warning message but there's not a ton we can tell them to do right now.
Reviewed By: achen1
Differential Revision: D4716273
fbshipit-source-id: c5d3a3ce8af53253b240477f2bde38094a138a02
Summary:
This changeset adds a test that verifies that the duplicate modules use case is broken. The goal is to acknowledge the problem for now, and when we update `jest-haste`, we'll simply fix what needs to be fixed in that test.
See also, https://github.com/facebook/jest/pull/3107
Reviewed By: davidaurelio
Differential Revision: D4673431
fbshipit-source-id: 05e09bdf61a2eddf2e9d1e32a33d32065c9051f1
Summary: This is breaking because it affects the contract for onBatchComplete, but modules really shouldn't (and probably aren't) depending on it being called without any actual native module method calls having happened.
Reviewed By: javache
Differential Revision: D4715656
fbshipit-source-id: 53ddd4a26c9949de86f5111d214b3e5002ca2e94
Summary:
Because `TextInput` only render the `defaultValue` of `string`, when using other types of values, it will render empty('') without any Warnings.
Closes https://github.com/facebook/react-native/pull/11478
Differential Revision: D4340132
Pulled By: lacker
fbshipit-source-id: aedb96d49277836000c0adc53007c97db6363253
Summary:
I've noticed many pull requests are opened with a very short description that fails to explain the motivation behind the PR, and many more forego providing a test plan. With this PR, I am shortening the template in order to provide concise instructions that provide the essential steps above the fold in the "Open a pull request" composer window.
- We need people to open a PR against master, not stable. The exact reason for this is not important and providing an explanation takes away from other more important points.
- Test plans are essential, but their requirement appears below the fold in the current template.
- More PRs could use tests.
- Make a point of asking PR authors to follow up on CI test failures.
- The composer does not parse Markdown into HTML. Markdown is pretty readable as is, but using reference links instead of inline links should help with readability.
I observed that it will only display the first 8 lines of the PR template above the fold. Seeing t
Closes https://github.com/facebook/react-native/pull/12958
Differential Revision: D4718294
Pulled By: mkonicek
fbshipit-source-id: b19a29e5ed73fb78d09c7de17625b1883590075c
Summary:
Only a small amendment, but took me a little bit to figure out why this wasn't working.
Closes https://github.com/facebook/react-native/pull/12706
Differential Revision: D4656700
fbshipit-source-id: d6038581aa70e96a2be775a7a9786e8c7e64f762
Summary:
I've updated the documentation that gives an example of Podfile adding a note that from React Native 0.42.0, users have to explicitly include Yoga in their Podfile.
I didn't want to modify existing `Podfile` from this documentation because they refer to `Podfile` from existing projects that do not use RN 0.42.0. So I just added a note and an example to explain how to include Yoga.
This PR is to fix the following issue: https://github.com/facebook/react-native/issues/12670
Closes https://github.com/facebook/react-native/pull/12728
Differential Revision: D4716172
Pulled By: hramos
fbshipit-source-id: 12a6e61b0d426c776a7cc2dea3ac22a50ffe56ef
Summary:
See https://github.com/facebook/react-native/issues/10845
onSizeChanged is enqueueing a runnable from the ui thread that references a shadow view by id on the native modules thread. Since the shadow thread 'runs ahead' of the UI thread (e.g. it has a newer state of the world than the UI thread), this isn't safe. By the time that code block runs, it's possible that an update from JS has already removed that view from the shadow hierarchy (a change which would then propagate to the native hierarchy on the UI thread).
Reviewed By: AaaChiuuu
Differential Revision: D4706521
fbshipit-source-id: 0915f081068709b895f70b2edce12163b39e5456
Summary:
Disabled earlier today, a fix for HMR landed so let's test it works.
Closes https://github.com/facebook/react-native/pull/12943
Differential Revision: D4712940
Pulled By: jeanlauliac
fbshipit-source-id: 38cbfb04a0a16a0a59e8eb4c9c26adb60ab5d635
Summary:
The breakage fixed by changeset [1] could have been identified earlier if we had typing on `attachHMRServer`, so I spent some time on that. This has revealed in turn a few functions across the codebase that were incorrectly typed, and that are now fixed.
[1] packager: attachHMRServer.js: fix callsite of Server#getModuleForPath()
Reviewed By: davidaurelio
Differential Revision: D4706241
fbshipit-source-id: fc4285245921ae45d5781a47d626fc0559dba998
Summary:
This fixes ```align-content: center``` and ```align-content: flex-end``` when the child exceeds the parents size. See #476. It also fixes those layouts if the child has ```margin: auto``` set.
Closes https://github.com/facebook/yoga/pull/477
Differential Revision: D4697833
Pulled By: emilsjolander
fbshipit-source-id: d081ec7ea559a5f2bd3271c3a4dc272960beddfa
Summary:
If user slide picker when picker item is zero, `UIPickerViewDelegate` will call `pickerView:didSelectRow:inComponent` row=0, `_items[row][@"value"]` will crash.
Closes https://github.com/facebook/react-native/pull/12187
Differential Revision: D4709882
Pulled By: mkonicek
fbshipit-source-id: 772c819d4eaef41ac983287877bda2918f40b1a7
Summary:
The ReactCxx project was not updated when RCTMap was removed in 48f30eca7e, this fixes it. It was also missing the systemJSCWrapper.cpp file.
**Test plan**
Tested that UIExplorerCxx now builds
cc mkonicek
Closes https://github.com/facebook/react-native/pull/12919
Differential Revision: D4709498
Pulled By: mkonicek
fbshipit-source-id: 995a1a914ab0a3227b1219c575a84d136800ff19
Summary:
**Motivation**
I needed to do this today. It's explained in code in the repository, but not included in the actual documentation. As React Native continues to grow, this is something that will be used more and more by the community.
**Test plan (required)**
No test plan required, this is simply a documentation change to add things that are only explained in code comments currently.
Closes https://github.com/facebook/react-native/pull/12938
Differential Revision: D4709339
Pulled By: mkonicek
fbshipit-source-id: a56c7573c3fa25a4059657b95b482b641ff229ff
Summary:
Don't crash in `StackTraceHelper.convertJsStackTrace` on missing `lineNumber`
I've seen the native Android app crash upon displaying a stack trace because of this bug.
Incidental and unrelated change: Refactor `StackTraceHelper.formatFrameSource` to use `StringBuilder`
Closes https://github.com/facebook/react-native/pull/12920
Differential Revision: D4709273
fbshipit-source-id: 95b3eb303f259460e33d8ee32a2db652360c645d
Summary:
Create a JSInstance superinterface which doesn't include all
the lifecycle stuff.
Reviewed By: AaaChiuuu
Differential Revision: D4614410
fbshipit-source-id: 16047bbcb1bb69bf36a0a13ef68f3a6aa396a991
Summary: In a previous changeset, Server#getModuleForPath() started returning `Promise<Module>` instead of `Module`, but the callsites in HMR haven't been updated, causing it to break. This would have been caught if `attachHMRServer.js` was using flow, that I'm considering doing in a following up diff. This would also have been caught if we had better integration testing of HMR. Good news however,it was caught by the OSS e2e test, that covers Hot Reloading.
Reviewed By: davidaurelio
Differential Revision: D4705937
fbshipit-source-id: fe787bc6ae50024759c7f7aeed747394fdce9aa1
Summary: The problem with `bundleOpts` is that it discards Flow typing, so it prevents reinforcing the integration of `Bundler` into `Server`. This changeset removes the `bundleOpts` to solve that issues. Instead, it makes the options explicit so that there is less uncertaintly. I love making options explicit, because they force callsites to take a consicious decision about what is really needed, making them more robust. They also expose oddities that probably needs refatoring, for example having a `resolutionRequest` in the bundle options does not seem correct, it should be an implementation details. Likewise, `onProgress` should probably be exposed differently, as it does not affect the content of the bundle itself.
Reviewed By: davidaurelio
Differential Revision: D4697729
fbshipit-source-id: d543870ba024e7588c10b101fa51429c77cc5ddc
Summary:
* Any animation can be looped on the javascript thread
* Only basic animations supported natively at this stage, loops run
using the native driver cannot contain animations of type sequence,
parallel, stagger, or loop
Motivation: We need a spinner in our app that is displayed and animated while the javascript thread is tied up with other tasks. This means it needs to be offloaded from the javascript thread, so that it will continue to run while those tasks are churning away.
I originally submitted PR #9513, which has served our needs, but brentvatne pointed out a better way to do it. Had hoped his suggestion would be implemented by janicduplessis or another fb employee, but after 5 months I thought I'd give it another push.
I've put together an implementation that basically matches the suggested API. Let me know what you think, and whether others can pick it up from here and get it in to core.
Personal Motivation: I am leaving my current organisation on Feb 10th, so am trying to clean thing
Closes https://github.com/facebook/react-native/pull/11973
Differential Revision: D4704381
fbshipit-source-id: 42a2cdf5d53a7c0d08f86a58485f7f38739e6cd9
Andy Street
mlanter
Jean Lauliac
Jakub Woyke
Ashok Menon
Bin Yue
Andrew Jack
Hector Ramos
Mike Adams
Maxime Aoustin
Marc Horowitz
Martin Konicek
Adam Ernst
Charles Dick
Lukas Wöhrl
Liu Zhanhong
ShiHui
Janic Duplessis
sunnylqm
tfallon@mail.depaul.edu
Petter Hesselberg
David Aurelio
Alexey Lang
Kyle Mathews
Daniel Woelfel
Jonathan Keljo
Aaron Chiu
Theo Yaung
Edwin
Héctor Ramos
Valentin Shergin
