Summary:
App Names can contain special characters. Escape them (and unescape later) so that device ID strings don't break.
Simpler is better - thanks to mweststrate for keeping me out of the rabbit hole
## Changelog
Escape app name to account for special characters in name
Pull Request resolved: https://github.com/facebook/flipper/pull/1268
Test Plan:
Added test to test character escaping: `cd desktop && yarn test -- -i app/src/utils/__tests__/clientUtils.node.tsx`
closes https://github.com/facebook/flipper/issues/1252
Reviewed By: passy
Differential Revision: D22066263
Pulled By: jknoxville
fbshipit-source-id: c820d055700ca6f55d35265528ce874eeb159216
Summary:
JS/TS api:
- migrate to TS
- some refactoring (get rid of bridge, make client abstract)
Implementation isn't full yet, things to be implemented:
- let plugins connect on init command from Flipper
- implement Responder
Further plans:
- make fully compatible with react-native api without breaking changes
Reviewed By: mweststrate
Differential Revision: D21839377
fbshipit-source-id: 9e9fe4ad01632f958b59eb255c703c6cbc5fafe2
Summary:
Adds a check for idb to flipper-doctor.
This depends on the flipper settings, to know where to look for idb, so I've made it so you can pass settings into the doctor when running it. When run from the command line, you don't get the settings. This is because settings loading currently depends on redux so I haven't extracted it into its own package.
Not that this will notify ALL open source iOS users with a doctor warning because they don't have idb installed. The error message says you can disable it in settings, which will silence this warning.
CHANGELOG: The open source version now works with physical iOS devices.
Reviewed By: passy
Differential Revision: D21883086
fbshipit-source-id: f28c43487e88a6c07ef3cc3da2764026726c9f18
Summary:
If the analytics plugin was in the foreground, and messages arrived in quick succession, some messages would not be processed.
Although the code was tested, there were not enough assertions to make sure the loop was correct. coverage !== correctness {emoji:1f605}
This fixes T68101450
Changelog: Fixed regression where analytics messages where lost
Reviewed By: jknoxville
Differential Revision: D21929679
fbshipit-source-id: c9fe2b18a249e40085d99914a809abf14fa7cf8f
Summary: Use interface PluginDetails everywhere where plugins are handled and removed PluginDefinition type which was effectively a subset of PluginDetails
Reviewed By: mweststrate
Differential Revision: D21927456
fbshipit-source-id: 434ebeef955b922cc11757e78fbba8dec05f1060
Summary: Moved plugin installation utilities to "plugin-lib" module. There are no functional changes in this diff, just refactoring so that plugin installation utils can be re-used by different modules.
Reviewed By: passy
Differential Revision: D21927387
fbshipit-source-id: 340516a544f7cfdcc15d94660dcb74a012054531
Summary:
See previous diff, let's store the collapsed state of sidebar sections in local storage.
Introduced a reusable hook to take care of that.
Changelog: Device plugins are now expanded by default, and the expand / collapse state will now be remembered across restarts
Reviewed By: passy
Differential Revision: D21903394
fbshipit-source-id: a3c0231acc0aa0877522ec328eedd09cb11aedb1
Summary:
Part of https://github.com/facebook/flipper/issues/262
Use the user-configured idb install location instead of a hardcoded one.
Reviewed By: passy
Differential Revision: D21860236
fbshipit-source-id: 5c604d7b6361e7c93ab49d8a03a437dfce880ac1
Summary:
See previous diff.
Achieves the same optimization as in the mentioned diff, but this time by only debouncing the messages as they arrive over the socket, and not the state updates caused by Redux directly. This means that plugin rendering won't be debounced anymore until we address this more fundamentally.
With this change there is a double level buffering:
1. A buffer that stores all incoming messages (that are not replies to requests)
2. A buffer that stores messages we are interested in in the plugin queue, unless the plugin is active (this we already had).
This still fixes the issue that too chatty plugins cause to many updates foreground plugin (the problem we tried to fix originally), without debouncing plugin rendering if it is needed to update for any other reason.
Another nice benefit is that previously every received message would trigger a store update in Redux which would cause all connected components to evaluate their subscriptions (and then bail out in the typical case). Now we will only update the redux store every 200 ms.
Changelog: Foreground plugins will burn less CPU when they're very chatty
Reviewed By: jknoxville
Differential Revision: D21858849
fbshipit-source-id: c72352e569a8a803bbedffb71b17b11fcefee043
Summary:
I've outlined the tasks required to get iOS device support working for open source users [here](https://github.com/facebook/flipper/issues/262).
This is the first step. It publishes the same code we use internally to GitHub, but in a state where it is only "available" for non-public builds. This will not change any behavior, but means that together with the community, we can start adapting it to suit everyone, and then eventually flip "available" to true for everyone.
Reviewed By: passy
Differential Revision: D21740193
fbshipit-source-id: 586c79ad850f67da330c10a007605ff25a187544
Summary:
Ugh, lockdown brain. This was supposed to be included in the test I added before
to get to 100% coverage. Well, it's here now.
Reviewed By: mweststrate
Differential Revision: D21550904
fbshipit-source-id: 044a11d38e211c6f57cce220adc8c42241a2043a
Summary:
This happened during startup:
{F237135281}
From `tmp` docs:
> IMPORTANT NOTE: template no longer accepts a path. Use the dir option instead if you require tmp to create your temporary filesystem object in a different place than the default tmp.tmpdir.
Reviewed By: mweststrate
Differential Revision: D21570416
fbshipit-source-id: 170886d0fda5e21cb23fe836fcde33eb457a4c1b
Summary: This was for V1 (D16990925) and is no longer referenced anywhere.
Reviewed By: nikoant
Differential Revision: D21501270
fbshipit-source-id: fef30f38c917afcd3d4150a0165cd0a59ad42b6a
Summary:
Both export methods behaved slightly differently:
* Export to file returned earlier when running in background, which skiped the tracking of export termination;
* Share link always showed notification, even when on foreground;
* Share link status "hanged" - the share was never unset.
This change makes both consistent:
* Always track export finish - whether its in foreground or background;
* Only show notificaiton if running in background;
* Always reset the share/status.
On top of this it also:
* Normalizes the screenshot status to terminate in '...' as all the others;
* Only copies export URL to clipboard when exporting link if running in the background.
Reviewed By: passy
Differential Revision: D21425095
fbshipit-source-id: 9864a63269df6bd05ab065ff0e5d9f17b9ac6db6
Summary:
This diff fixes an issue where we don't want to have the Navigation plugin be disabled as background plugins like all other plugins, so that the breadcrumb navigation keeps working.
Yet another hack concerning the super useful Navigation plugin. On a positive note, since connection management for background is not entirely managed by the Desktop and not the native said, these kind of exceptions are fairly easy to make :)
Reviewed By: passy
Differential Revision: D21089537
fbshipit-source-id: 209954ff35c95e066fe688a60ad46ccfc3835c44
Summary:
Background for this diff: https://fb.quip.com/KqEfAlKYlgme
Some plugins don't respect that stuff (livefeed and graphql), but for others it seems to work fine.
This is just a PoC, there are some present bugs concerning the combination of selecting and bg plugins
Questions to investigate:
- [x] make sure that LiveFeed and GraphQL disconnect properly. There might be more plugins that need that
- [x] verifiy that we don't loose one of the original goals of background plugins, e.g. QPL collecting and sending data from device start. Does this still work as intended after this change?
- [x] how can we observe / measure improvements? Are dev builds more responsive after this? Is the layout inspector smoother for example because no QPL plugins are interweaved?
- [x] how is forward and backward compatibility?
- If Flipper is updated, but device not: No change I think, as getBackgroundPlugins() will return an empty set, and background plugins are initiated as usual, so old behavior
- If device is updated, but Flipper not, background plugins won't be started until they are selected. This is a degradation, but hopefully explainable.
- [x] Verify QPL buffer is not unbounded
- [x] Share architecutre changes with team
For Graphql updates: D20943455
Added runtime stats to monitor network traffic (sadly had to redo that since scuba couldn't handle the data format used at first, so probably will hold of landing this diff a week to make sure we can see some effects)
Follow up work:
[x] wait until we released the stat tracking before we release this, to be able to measure the effect?
[x] make sure graphql fix lands
[ ] use side effects abstraction
[ ] fix other background plugins (android only) or fix it in a generic way:
{F234394286}
Changelog: Background plugins will no longer receive a Flipper connection if they are disabled. This should significantly reduce the overall load of Flipper both on the device and desktop when unused plugins are disabled used, which could otherwise generate 10MB/s of network traffic certain scenarios. All plugins *should* be able to handle to this gracefully, but since this is quite a fundamental change, reach out to the Flipper team when in doubt!
Reviewed By: jknoxville
Differential Revision: D20942453
fbshipit-source-id: b699199cb95c1b3e4c36e026b6dfaee7d1652e1f
Summary:
Pull Request resolved: https://github.com/facebook/flipper/pull/1052
Fixed tests failing on Windows because jest auto-mocks works differently there
Reviewed By: passy
Differential Revision: D21237618
fbshipit-source-id: 31c7e92b7f8ae84c1e65fd37428204452b3f1b00
Summary: Fixes a regression (D20679687) of the main menu not loading immediately after application start
Reviewed By: passy, priteshrnandgaonkar
Differential Revision: D21227817
fbshipit-source-id: 37e4ddfcb73de3eac04d6162a3e028864d3e9e7f
Summary: Added eslint rule "no-extraneous-imports" which disallow using modules which are not listed as dependencies in the corresponding package.json. Fixed a bunch of reported errors after the rule applied.
Reviewed By: passy
Differential Revision: D21186848
fbshipit-source-id: 0af9ac4b3fffdfd0ab7c23ae4ff12a3f5989d5e9
Summary:
This diff adds supportsMethod as a closure which can be used to verify a method is implemented on the client side.
It will be used in the next diff
Reviewed By: jknoxville
Differential Revision: D21176415
fbshipit-source-id: fe16d966c58d36558034ce4ade8f58f8031aab18
Summary:
I was trying to plot some graphs based on our background plugin stats, grouped per plugin, but discovered that we can't exploded or group on json object keys, so separated the data into separate `plugin-stats-plugin` events, as is done with the `time-spent` event.
Also fixed:
* grouping stats per plugin, rather than per plugin method, which was to fine-grained
* added `bytesReceived` field to the cumulative event `plugin-stats`
Reviewed By: jknoxville
Differential Revision: D21046016
fbshipit-source-id: 1043612064921cf6427d5b3bbee10b76776df39e
Summary:
D20920582 was unfortunately not enough to fix ability for Flipper to connect to packages that are running as services without an `<application>` tag in the manifest.
The issue there is that the push of `sonarCA.crt` succeeded due to erroneous changes to the target package.
This change ensures both pull and push are covered.
Reviewed By: jknoxville
Differential Revision: D21024673
fbshipit-source-id: cad86ba9ae3b02a99187d3f6bd3b500d2a4b971a
Summary:
Pull Request resolved: https://github.com/facebook/flipper/pull/998
After this diff all the default plugins (which are distributed with Flipper) will be included into the main app bundle instead of bundling each of them separately and then loading from file system. This is done by auto-generating plugins index in build-time and importing it from Flipper app bundle, so Metro can follow these imports and bundle all the plugins to the app bundle.
This provides several benefits:
1) reduced Flipper bundle size (~10% reduction of zipped Flipper archive), because Metro bundles each of re-used dependencies only once instead of bundling them for each plugin where such dependency used.
2) Faster Flipper startup because of reduced bundle and the fact that we don't need to load each plugin bundle from disk - just need to load the single bundle where everything is already included.
3) Metro dev server for plugins works in the same way as for Flipper app itself, e.g. simple refresh automatically recompiles bundled plugins too if there are changes. This also potentially should allow us to enable "fast refresh" for quicker iterations while developing plugins.
4) Faster build ("yarn build --mac" is 2 times faster on my machine after this change)
Potential downsides:
1) Currently all the plugins are identically loaded from disk. After this change some of plugins will be bundled, and some of them (third-party) will be loaded from disk.
2) In future when it will be possible to publish new versions of default plugins separately, installing new version of such plugin (e.g. with some urgent fix) will mean the "default" pre-built version will still be bundled (we cannot "unbundle" it :)), but we'll skip it and instead load new version from disk.
Changelog: Internals: include default plugins into the main bundle instead producing separate bundles for them.
Reviewed By: passy
Differential Revision: D20864002
fbshipit-source-id: 2968f3b786cdd1767d6223996090143d03894b92
Summary:
Pull Request resolved: https://github.com/facebook/flipper/pull/969
Added all public plugins as workspaces to the root package.json. This means all these plugins will use the single root yarn.lock and installation of their dependencies will be faster. This also means that plugins can declare dependencies to other local packages included into workspaces and they will be symlinked automatically.
Changelog: Internals: plugins added as "yarn workspaces" into the root package.json to simplify dependency management between them
Reviewed By: mweststrate
Differential Revision: D20805231
fbshipit-source-id: e85c62d3195d1ea3c5c60def6ca12318a2b53466
Summary: When an android package is not an application, the `run-as` command will fail. In this case, the package might be an operating system service package. In this case, it may be possible for the user to run adb as root. Note that Flipper does not restart adbd via `adb root` on behalf of the user; the command is simply retried without `run-as`.
Reviewed By: jknoxville
Differential Revision: D20920582
fbshipit-source-id: 8db86084c3c3a61d8322edb1e34fdfdf48d0412d
Summary:
Measure how many byte we receive per plugin, and add this to the plugin stats that are collected
Will add a graph to the flipper dashboard, and probably a small visualization in a next diff as well.
Reviewed By: priteshrnandgaonkar
Differential Revision: D20917583
fbshipit-source-id: bb341531ecf8492080af82c56e73c0ec608f7b36
Summary:
* Fixed issue where the test wouldn't be reliable if it fired to quickly since the process has started
* Increased all test timings with a factor 10, to make the test less sensitive to system load
Reviewed By: passy
Differential Revision: D20914978
fbshipit-source-id: a3870e6374e61cf9ec1b11da529077876ef85bf8
Summary: This diff adds the capability to show the plugin names for which flipper failed to fetch metadata. These changes are done for both export as URL and export as File. This diff also fixes the logging for the export as a file and logs the result object in scuba.
Reviewed By: jknoxville
Differential Revision: D20724860
fbshipit-source-id: 4c9591267ca05045e0ed084804d96851c9d7636d
Summary:
Before we dropped all messages if the client connected before the device has been registered. Which happens typically on iOS (it can actually take very long, will investigate that as well, but not in this diff).
However, we don't need to wait for the device to process messages, we just need its serial, which we already know (given that we use that query id to find the matching device in the first place).
I think this will greatly improved perceived stability for iOS
Reviewed By: jknoxville
Differential Revision: D20734892
fbshipit-source-id: f98e8d31558ef606b9a8287e03fc41ab6c3a087d
Summary:
This diff introducing the sideEffect abstractions, that creates a fundamental solution for the problem in D20619226: If a subscription to the store errors, the entire store can't be updated anymore, bringing the entire app to a halt.
Applying this abstraction will be done in a next diff.
To essential idea of sideEffect is that it fixes a few problems:
1. I decouples and throttles the effect so that no arbitrary expensive burden is added to every store update
2. It makes sure that a crashing side effect doesn't crash the entire store update
3. It helps with tracing and monitoring perf problems
4. It puts the side effect behind a selector so that the side effect is only triggered if a relevant part of the store changes, like we do for components.
Note that if some subscription _must_ be handled synchronously, than that logic should be in a reducer, not in a subscription or side effect. Luckily we don't have any examples of that in our code base.
This abstraction might actually be intesting to be shared wider for fun and profit.
Reviewed By: passy
Differential Revision: D20625872
fbshipit-source-id: adaf8356950594d50e6a99a17a862f757c3777db
