Enable ReactNativeHost integration test #5071

vmoroz · 2020-05-30T23:22:11Z

In the Microsoft.ReactNative.IntegratonTests we already had a test for the ReactNativeHost, but it was not running because we have to fix a number of issues. In this PR we are addressing these issue and make the test to run.
To enable the tests the following changes had to be done:

Expose ReactNativeHost API for instance load and unload that return IAsyncAction.
Created an adapter from Mso::Future to IAsyncAction.
Implement Application::Current() property override to return nullptr instead of throwing exceptions if the current application is absent.
Changed native modules that depend on the Application::Current() to check for nullptr and do nothing in such case.
Added collection of IReactPackageProvider to ReactInstanceSettings.
Queue up the CallJSFunction calls while JavaScript is loaded. Otherwise, native modules, that are created while loading JS, lose the call.
Fixed sample applications which are not working in current code.

Fixes #5057
Fixes #5070

Microsoft Reviewers: Open in CodeFlow

dannyvv · 2020-05-31T17:25:04Z

packages/microsoft-reactnative-sampleapps/windows/SampleAppCPP/App.cpp

+/// <param name="e">Details about the launch request and process.</param>
+void App::OnLaunched(LaunchActivatedEventArgs const &e) {
+  super::OnLaunched(e);
+  Frame rootFrame = Window::Current().Content().as<Frame>();


Just observing: Aleksander's PR has window.Activate(); before navigating...
#Resolved

just seeing this now. It's not necessary to call Window::Activate if you call super::OnLaunched. Otherwise you must call it. #Resolved

It works without it. Like Alexander says it is redundant in this context.

In reply to: 432967013 [](ancestors = 432967013)

dannyvv · 2020-05-31T17:26:33Z

packages/microsoft-reactnative-sampleapps/windows/SampleAppCS/MainPage.xaml

@@ -1,3 +1,5 @@
+<!-- Copyright (c) Microsoft Corporation. All rights reserved.
+     Licensed under the MIT License.  -->


[](start = 36, length = 2)

All Xml comments end with double-space...
Or is that the auto-formatter? #Resolved

It was copy & paste. I can fix it.

In reply to: 432967124 [](ancestors = 432967124)

Removed extra space.

In reply to: 433240152 [](ancestors = 433240152,432967124)

dannyvv · 2020-05-31T17:31:08Z

vnext/Microsoft.ReactNative/ReactNativeHost.cpp

+  return ReloadInstance();
+}
+
+IAsyncAction ReactNativeHost::ReloadInstance() noexcept {


To me this seems a bit counter-intuitive...
To me it seems backwards. LoadInstance and UnloadInstance would have implemenataton. and ReloadInstance would just call UnloadInstance and then LoadInstance.

I realize it is an artifact of the m_reactHost api... just calling out if feels 'funny'.
#Resolved

The ReloadInstance is much more powerful than LoadInstance because otherwise requiring to always unload before load would be too difficult and error prone. I would rather have it as the prime API instead of inventing what to do if developers call LoadInstance twice without calling UnloadInstance between them. Should we crash or ignore? Plus, the Reload call could let us optimize some internals if needed.
So, why to have LoadInstance in addition to the ReloadInstance which does the same work? The answer is the "syntactic sugar". The ReloadInstance call looks strange when we just start application. Saying LoadInstance seems to be more logical. Internally the 'unload' part is noop in that case anyway.

In reply to: 432967474 [](ancestors = 432967474)

dannyvv · 2020-05-31T17:43:22Z

vnext/Microsoft.ReactNative.IntegrationTests/ReactNativeHostTests.cpp

+      host.InstanceSettings().UseDeveloperSupport(false);
+      host.InstanceSettings().UseWebDebugger(false);
+      host.InstanceSettings().UseFastRefresh(false);
+      host.InstanceSettings().UseLiveReload(false);


I've noticed that also in the template these are all set at the same time, should we have a single method to toggle these settings (in case we get more in the future) i.e.
UseDevSettings(true|false) #Resolved

It is a good point. We could implement it in a separate PR.

In reply to: 432968392 [](ancestors = 432968392)

I have created Issue #5078 to address it in the future.

In reply to: 433245285 [](ancestors = 433245285,432968392)

dannyvv · 2020-05-31T17:45:21Z

vnext/Mso/future/futureWinRT.h

+
+namespace Mso {
+
+struct AsyncActionFutureAdapter : winrt::implements<


AsyncActionFutureAdapter [](start = 7, length = 24)

Just checking if we have a convention like is ocmmon in C# to have one type per file and match the filename to the type.
Here I see fileName: futureWinRT.h, with type struct AsyncActionFutureAdapter #Resolved

I chose the file name for the 'growth'. In C++ we often have multiple types in the same file.
The WinRT has four async interfaces based on the IAsyncInfo: IAsyncAction, IAsyncOperation, IAsyncActionWithProgress, an IAsyncOperationWithProgress. This file must contain conversion to all four of them and then conversion back from them. I just did a required bare minimum at this point.

In reply to: 432968554 [](ancestors = 432968554)

dannyvv

NickGerleman · 2020-05-31T21:50:14Z

vnext/Desktop/pch.h

 #pragma once

+#ifndef WIN32_LEAN_AND_MEAN
+#define WIN32_LEAN_AND_MEAN


FYI we will already define these for all projects including React.cpp.props. No harm in adding this to the PCH as well though. #Resolved

I will remove it. It is a not related left-over from solving a build error.

In reply to: 432987546 [](ancestors = 432987546)

Removed extra defs from pch.h

In reply to: 433248481 [](ancestors = 433248481,432987546)

NickGerleman · 2020-05-31T21:56:35Z

vnext/Microsoft.ReactNative.Cxx/UI.Xaml.impl.override.h

+namespace winrt::impl {
+
+template <>
+WINRT_IMPL_AUTO(Windows::UI::Xaml::Application)


Can we define our own helper instead of replacing the original? This feels a bit fragile, or unexpected. #Resolved

Yes, I think in this case having a helper would make more sense. I was just played too much with the idea how I can augment the generated consumer classes with custom code instead of having all these manually written ReactContext, ReactDispatcher, and other classes.

In reply to: 432988186 [](ancestors = 432988186)

Defined TryGetCurrentApplication instead.

In reply to: 433249993 [](ancestors = 433249993,432988186)

NickGerleman · 2020-05-31T22:04:37Z

vnext/Microsoft.ReactNative/ReactHost/ReactInstanceWin.cpp

+  {
+    std::scoped_lock lock{m_mutex};
+    if (m_state == ReactInstanceState::HasError || m_state == ReactInstanceState::Unloaded) {
+      jsCallQueue = std::move(m_jsCallQueue);


Why not clear()? #Resolved

To avoid calling destructors under the lock. It is a common technique. I will add a comment.

In reply to: 432988947 [](ancestors = 432988947)

Added comments

In reply to: 433251387 [](ancestors = 433251387,432988947)

NickGerleman · 2020-05-31T22:07:55Z

vnext/Microsoft.ReactNative/ReactNativeHost.idl

  runtimeclass ReactNativeHost {
    ReactNativeHost();

-    IVector<IReactPackageProvider> PackageProviders { get; set; };


What was previous behavior of someone set this? #Resolved

I think it was buggy. It was like giving external code ability to access private variables. As a rule of thumb we must not allow the collections to be set, but rather just modify them. Why? For example, if we decide that the internal storage must be thread-safe (which is not today), then there is no way for us to enforce it if we allow the collection to be set. It means that the API has an unnecessary 'freedom' that puts us in the corner when we want to improve the implementation.

In reply to: 432989273 [](ancestors = 432989273)

NickGerleman · 2020-05-31T22:12:52Z

vnext/Mso/future/futureWinRT.h

+    try {
+      winrt::slim_lock_guard const guard(m_lock);
+      RethrowIfFailed();
+      return 0;


nit: S_OK? #Resolved

The C++/WinRT always uses 0 for some reason.

In reply to: 432989753 [](ancestors = 432989753)

NickGerleman · 2020-05-31T22:16:24Z

vnext/Mso/future/futureWinRT.h

+  }
+
+  uint32_t Id() const noexcept {
+    return 1;


Does the contract for this allow collisions? Any reason we do this instead of a monontonically increasing int? #Resolved

This is what the C++/WinRT uses internally. It seems that the Id property is never used. The code is 'adapted' from their implementation of IAsyncAction for co-routines.

In reply to: 432990053 [](ancestors = 432990053)

packages/microsoft-reactnative-sampleapps/windows/SampleAppCPP/App.cpp

vnext/Microsoft.ReactNative.Cxx/CppWinRTIncludes.h

vnext/Microsoft.ReactNative.Cxx/UI.Xaml.impl.override.h

asklar · 2020-05-31T23:06:28Z

In the Microsoft.ReactNative.IntegratonTests we already had a test for the ReactNativeHost, but it was not running because we have to fix a number of issues. In this PR we are addressing these issue and make the test to run.
To enable the tests the following changes had to be done:

Expose ReactNativeHost API for instance load and unload that return IAsyncAction.

Created an adapter from Mso::Future to IAsyncAction.

Implement Application::Current() property override to return nullptr instead of throwing exceptions if the current application is absent.

Changed native modules that depend on the Application::Current() to check for nullptr and do nothing in such case.

Added collection of IReactPackageProvider to ReactInstanceSettings.

Queue up the CallJSFunction calls while JavaScript is loaded. Otherwise, native modules, that are created while loading JS, lose the call.

Fixed sample applications which are not working in current code.

Fixes #5057
Fixes #5070

Microsoft Reviewers: Open in CodeFlow

I see there are some changes to instance load state, will this change fix #5002 too? #Resolved

vmoroz · 2020-06-01T14:14:55Z

No, this change does not fix #5002, but it gets us closer. As we write more and more tests that exercise the ReactNativeHost API, we can start provide guarantees that it behaves correctly. For example, even the simplest test has shown that we had an issue implementing the CallJSFunction. The issue #5002 indicates that there is something wrong with the ReactViewHost handling. I hope that we can address it by adding more tests.

In reply to: 636544322 [](ancestors = 636544322)

vmoroz · 2020-06-01T14:53:10Z

@JunielKatarn, @acoates-ms, I am disabling a number of Desktop unit tests that prevent many of PRs successful checks. None of these tests can run locally in Visual Studio. If we need them, then they must be fixed/updated to be run successfully locally in Visual Studio first before we require them to pass in CI loop.

asklar · 2020-06-01T16:01:38Z

@JunielKatarn, @acoates-ms, I am disabling a number of Desktop unit tests that prevent many of PRs successful checks. None of these tests can run locally in Visual Studio. If we need them, then they must be fixed/updated to be run successfully locally in Visual Studio first before we require them to pass in CI loop.

@vmoroz I'm actually able to run these tests locally which is why I was trying to figure out why they're failing in the pipeline. I'm assuming these fail for you? are you starting the bundler in \vnext ?

NikoAri · 2020-06-01T17:54:05Z

packages/microsoft-reactnative-sampleapps/windows/SampleAppCS/MainPage.xaml

@@ -1,3 +1,5 @@
+<!-- Copyright (c) Microsoft Corporation. All rights reserved.


All rights reserved. [](start = 42, length = 20)

From:
https://docs.opensource.microsoft.com/changelog.html?q=all%20rights%20reserved

CELA open source policy
The term All Rights Reserved is no longer recommended to be in MIT license copyright headers. #Resolved

It is addressed by PR #5083

In reply to: 433393589 [](ancestors = 433393589)

vmoroz · 2020-06-01T17:58:05Z

packages/microsoft-reactnative-sampleapps/windows/SampleAppCS/App.xaml.cs

-            var frame = Window.Current.Content as Frame;
-            frame.Navigate(typeof(MainPage));
-            Window.Current.Activate();
+            var frame = (Frame)Window.Current.Content;


(Frame)Window.Current.Content; [](start = 24, length = 30)

Note that in C# the 'as' conversion returns null if conversion failed, while the C-style conversion throws an exception. It is different from the C++/WinRT where the 'as' conversion throws an exception while 'try_as' returns null in case of failure.
This is why the code here is changed here to use C-style conversion to match the C++/WinRT 'as()' function call which is a correct functionality since we do not check for null after the call. #Resolved

ghost · 2020-06-01T17:58:53Z

Hello @vmoroz!

Because this pull request has the AutoMerge label, I will be glad to assist with helping to merge this pull request once all check-in policies pass.

p.s. you can customize the way I help with merging this pull request, such as holding this pull request until a specific person approves. Simply @mention me (`@msftbot`) and give me an instruction to get started! Learn more here.

aeulitz · 2020-06-01T18:17:30Z

vnext/Desktop.IntegrationTests/HttpResourceIntegrationTests.cpp


-  TEST_METHOD(RequestGetSucceeds) {
+  // This test always fails because the requested resource does not exist.
+  // TEST_METHOD(RequestGetSucceeds) {


// TEST_METHOD [](start = 2, length = 14)

Does whatever harness this project uses support a SKIPTESTMETHOD annotation? #Resolved

No, this is VS test. We cannot use the SKIPTESTMETHOD here.

In reply to: 433405820 [](ancestors = 433405820)

aeulitz · 2020-06-01T18:19:07Z

vnext/Desktop.IntegrationTests/RNTesterIntegrationTests.cpp

 } // namespace Microsoft::VisualStudio::CppUnitTestFramework

+// None of these tests are runnable
+#if 0


Remove/deactivate TEST_CLASS attribute, but keep it compiling? #Pending

Same for the other tests in the folder.

In reply to: 433406671 [](ancestors = 433406671)

aeulitz · 2020-06-01T18:21:33Z

vnext/Microsoft.ReactNative/Microsoft.ReactNative.vcxproj.filters

      <Filter>Modules</Filter>
    </ClCompile>
-    <ClCompile Include="..\ReactUWP\Modules\DeviceInfoModule.cpp">
-      <Filter>Modules</Filter>


Modules [](start = 6, length = 24)

The PR doesn't contain a corresponding change to the .vcxproj. Is it not needed? #Resolved

It seems that VS has a bug managing the filters file. The file was removed by another PR, but when someone opens the project we see the filter file changed. I do not know a good way to avoid it unless VS fixes the issue. #Resolved

It is done by some other change. VS has a bug that filters file is not updated/saved correctly. When next developer opens VS project they see such random changes. No harm to bring this change along.

In reply to: 433407937 [](ancestors = 433407937)

aeulitz · 2020-06-01T18:26:49Z

vnext/Microsoft.ReactNative/ReactNativeHost.idl

    ReactInstanceSettings InstanceSettings { get; set; };

-    void ReloadInstance();
+    Windows.Foundation.IAsyncAction LoadInstance();


Windows.Foundation.IAsyncAction LoadInstance(); [](start = 4, length = 47)

Can we please add doc-style comments to document intended usage of the API? #Pending

aeulitz · 2020-06-01T18:30:01Z

vnext/Microsoft.ReactNative/ReactHost/React.h

  WaitingForDebugger,
  Loaded,
  HasError,
+  Unloaded,


Can we please add comments detailing how these states relate to which operations on what API components? #Pending

aeulitz · 2020-06-01T19:41:17Z

vnext/Microsoft.ReactNative.Cxx/Microsoft.ReactNative.Cxx.vcxitems

    <ClInclude Include="$(MSBuildThisFileDirectory)UI.Input.h" />
    <ClInclude Include="$(MSBuildThisFileDirectory)UI.Popups.h" />
    <ClInclude Include="$(MSBuildThisFileDirectory)UI.Text.h" />
+    <ClInclude Include="$(MSBuildThisFileDirectory)XamlHelpers.h" />


)XamlHelpers.h [](start = 50, length = 14)

Is this intentional? The change appears to be adding a XamlUtils.h file. #Pending

No, it is another artifact of failed .filters file handling...

In reply to: 433448516 [](ancestors = 433448516)

aeulitz · 2020-06-01T19:43:37Z

vnext/Microsoft.ReactNative.Cxx/CppWinRTIncludes.h

 #include <winrt/Windows.UI.Xaml.h>

+#define XAML_CPPWINRT_NAMESPACE winrt::Windows::UI::Xaml
 namespace xaml = winrt::Windows::UI::Xaml;


For my education: Is this header file specifically meant for Xaml-based consumers? #Resolved

Yes, this is for XAML users. We are going to split up the project to separate the "core" files from XAML. We just need to change some ABI files first.

In reply to: 433449630 [](ancestors = 433449630)

aeulitz · 2020-06-01T19:45:13Z

vnext/Microsoft.ReactNative.IntegrationTests/ReactNativeHostTests.cpp

-  // app into a UWP app). In the long term, the issue should get addressed by splitting the Microsoft.ReactNative.dll
-  // into a "core" and a "UI" DLL, and by having the tests in this project target the "core" DLL.
-  SKIPTESTMETHOD(JsFunctionCall_Succeeds) {
+  TEST_METHOD(JsFunctionCall_Succeeds) {


TEST_METHOD [](start = 1, length = 12)

Awesome! #Resolved

aeulitz · 2020-06-01T19:51:57Z

vnext/Shared/OInstance.cpp

 #else
-      std::string bundlePath = m_devSettings->bundleRootPath + jsBundleRelativePath + ".bundle";
+      std::string bundlePath =
+          (fs::path(m_devSettings->bundleRootPath) / (jsBundleRelativePath + ".bundle")).u8string();


/ [](start = 51, length = 1)

Is this an operator overload for FS paths? #Resolved

Yes, this is how the std::filesystem concatenates paths.

In reply to: 433453526 [](ancestors = 433453526)

JunielKatarn · 2020-06-02T20:52:02Z

vnext/Desktop.IntegrationTests/WebSocketJSExecutorIntegrationTest.cpp


+// None of these tests are runnable
+#if 0
 TEST_CLASS (WebSocketJSExecutorIntegrationTest) {


Wish I had seen this review before.
There are more orthodox ways to disable these tests.
We should aim to run them in UWP too, actually.

#Resolved

Let's discuss it offline.

In reply to: 434168343 [](ancestors = 434168343)

NickGerleman · 2020-06-03T03:07:21Z

Removing the Backport to 0.62 label, since we need to rename it as part of the deprecation in favor of Request Backport which was already added to this PR. Triage of this item is happening in the next round it sounds like.

Khalef1 · 2020-06-08T22:13:56Z

We discussed this in triage. Full change is too large to be backported, but we should backport the CallJSFunction queue only

Khalef1 · 2020-06-08T22:14:45Z

Please note that only part of this change has been approved for backport

* [0.62] Cherry pick PR #5071 for CallJsFunction queuing * Change files Co-authored-by: Andrew Coates <30809111+acoates-ms@users.noreply.github.com>

vmoroz added 3 commits May 30, 2020 16:11

Enable ReactNativeHost integration test

32bc9d3

Change files

cb6e28c

Format code

ab30ce8

vmoroz requested review from a team, acoates-ms, aeulitz and asklar May 30, 2020 23:22

vmoroz requested a review from a team as a code owner May 30, 2020 23:22

vmoroz mentioned this pull request May 31, 2020

Add SampleAppCpp OnLaunched #5074

Closed

dannyvv reviewed May 31, 2020

View reviewed changes

dannyvv approved these changes May 31, 2020

View reviewed changes

Fix Playground-Win32 compilation

a75499f

NickGerleman reviewed May 31, 2020

View reviewed changes

asklar reviewed May 31, 2020

View reviewed changes

packages/microsoft-reactnative-sampleapps/windows/SampleAppCPP/App.cpp Outdated Show resolved Hide resolved

asklar reviewed May 31, 2020

View reviewed changes

vnext/Microsoft.ReactNative.Cxx/CppWinRTIncludes.h Outdated Show resolved Hide resolved

asklar reviewed May 31, 2020

View reviewed changes

vnext/Microsoft.ReactNative.Cxx/UI.Xaml.impl.override.h Outdated Show resolved Hide resolved

Disable non-runable Desktop unit tests

0729055

vmoroz added 2 commits June 1, 2020 07:54

Format code

78d6d15

Remove extra space in XML comments

5d8f4cd

vmoroz mentioned this pull request Jun 1, 2020

Group developer settings to simplify their use #5078

Open

NickGerleman approved these changes Jun 1, 2020

View reviewed changes

Remove extra definitions from Desktop pch.h

3a9ccad

vmoroz added 2 commits June 1, 2020 10:39

Format and add comments

f3bda0c

Use xaml alias in sample app

3043c89

NikoAri reviewed Jun 1, 2020

View reviewed changes

vmoroz commented Jun 1, 2020

View reviewed changes

vmoroz added the AutoMerge Causes a PR to be automatically merged once all requirements are passed (label drives bot activity) label Jun 1, 2020

vmoroz added the Backport to 0.62 label Jun 1, 2020

aeulitz reviewed Jun 1, 2020

View reviewed changes

ghost merged commit 80d1a40 into microsoft:master Jun 1, 2020

aeulitz reviewed Jun 1, 2020

View reviewed changes

Khalef1 added the Request Backport label Jun 2, 2020

JunielKatarn reviewed Jun 2, 2020

View reviewed changes

NickGerleman removed the Backport to 0.62 label Jun 3, 2020

Khalef1 added Backport Approved and removed Request Backport labels Jun 8, 2020

This was referenced Jun 12, 2020

[0.62] Cherry pick PR #5071 for CallJSFunction queuing #5189

Closed

[0.62] Cherry pick PR #5071 for CallJsFunction queuing #5193

Merged

acoates-ms added a commit that referenced this pull request Jun 12, 2020

[0.62] Cherry pick PR #5071 for CallJsFunction queuing (#5193)

d3bcc6d

* [0.62] Cherry pick PR #5071 for CallJsFunction queuing * Change files Co-authored-by: Andrew Coates <30809111+acoates-ms@users.noreply.github.com>

Khalef1 added Backported to 0.62 and removed Backport Approved labels Jun 12, 2020

This pull request was closed.

		@@ -1,3 +1,5 @@
		<!-- Copyright (c) Microsoft Corporation. All rights reserved.
		Licensed under the MIT License. -->


		namespace Mso {

		struct AsyncActionFutureAdapter : winrt::implements<

Enable ReactNativeHost integration test #5071

Enable ReactNativeHost integration test #5071

Uh oh!

Conversation

vmoroz commented May 30, 2020 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Microsoft Reviewers: Open in CodeFlow

Uh oh!

dannyvv May 31, 2020 • edited by vmoroz Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

asklar May 31, 2020 • edited by vmoroz Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

dannyvv May 31, 2020 • edited by vmoroz Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

dannyvv May 31, 2020 • edited by vmoroz Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

dannyvv May 31, 2020 • edited by vmoroz Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

dannyvv May 31, 2020 • edited by vmoroz Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

dannyvv left a comment

Choose a reason for hiding this comment

Uh oh!

NickGerleman May 31, 2020 • edited by vmoroz Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

NickGerleman May 31, 2020 • edited by vmoroz Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

NickGerleman May 31, 2020 • edited by vmoroz Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

NickGerleman May 31, 2020 • edited by vmoroz Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

NickGerleman May 31, 2020 • edited by vmoroz Loading Uh oh! There was an error while loading. Please reload this page.

vmoroz commented May 30, 2020 •

edited

Loading

dannyvv May 31, 2020 •

edited by vmoroz

Loading

asklar May 31, 2020 •

edited by vmoroz

Loading

dannyvv May 31, 2020 •

edited by vmoroz

Loading

dannyvv May 31, 2020 •

edited by vmoroz

Loading

dannyvv May 31, 2020 •

edited by vmoroz

Loading

dannyvv May 31, 2020 •

edited by vmoroz

Loading

NickGerleman May 31, 2020 •

edited by vmoroz

Loading

NickGerleman May 31, 2020 •

edited by vmoroz

Loading

NickGerleman May 31, 2020 •

edited by vmoroz

Loading

NickGerleman May 31, 2020 •

edited by vmoroz

Loading

NickGerleman May 31, 2020 •

edited by vmoroz

Loading

NickGerleman May 31, 2020 •

edited by vmoroz

Loading

asklar commented May 31, 2020 •

edited by vmoroz

Loading

NikoAri Jun 1, 2020 •

edited by vmoroz

Loading

vmoroz Jun 1, 2020 •

edited

Loading

p.s. you can customize the way I help with merging this pull request, such as holding this pull request until a specific person approves. Simply @mention me (`@msftbot`) and give me an instruction to get started! Learn more here.

aeulitz Jun 1, 2020 •

edited by vmoroz

Loading

aeulitz Jun 1, 2020 •

edited by vmoroz

Loading