- Make sure we always await the last flush task before calling FlushAsync
again instead of preemptively calling FlushAsync and checking to see
if the ValueTask is incomplete before bothering to acquire the _flushLock
- This now acquires the _flushLock fore every call to Response.Body.Write
whereas this only happened for truly async writes before.
- I don't think this is a big concern since this should normally be uncontested,
and DefaultPipeWriter.FlushAsync/GetResult already acquire a lock.
`GetRuntimeInterfaceMap` is a rather expensive API and it doesn't look like the use here is warranted.
The replacement should have equivalent behavior.
This is a massive set of changes to Kestrel to remove the existing pubternal transport layer and implement a public facing API for listeners and clients, see the details here #10308.
This change only has the server side pieces of the story as I don't want to add the client APIs without having ported SignalR to use them. Here are the highlights:
- Transport.Abstractions is empty (will be removed in a separate PR as it requires removing it from a ton of places)
- TransportConnection has been moved to Connection.Abstractions (we can decide if we need to consolidate with DefaultConnectionContext in a later PR)
- Added FileHandleEndPoint which allows binding to a file handle (could be a pipe or tcp handle)
ListenOptions has been gutted for most pubternal API and returns various types of binding information . The source of truth is the EndPoint instance.
- Cleaned up a bunch of libuv tests decoupling them from Kestrel.Core
## Breaking changes
- Removing pubternal API is itself a breaking change but one that we already planned to do.
- We've removed the ability to set the scheduling mode on Kestrel
- DisposeAsync was added to ConnectionContext
- NoDelay was removed from ListenOptions. This has been moved to each of the transports. One major difference though is that it's no longer localized per endpoint but is global. We'd need a derived EndPoint type (or maybe extend IPEndPoint) to store both the socket options and the binding information.
Attempt to improve reliability of running yarn commands. Our project structure might cause yarn to be launched multiple times. According to yarn docs, this option should avoid conflicts between multiple instances of yarn
https://yarnpkg.com/en/docs/cli/#toc-concurrency-and-mutex
* Updates Identity server dependency for preview4
* Infrastructure improvements.
* Updates react versions.
* Updates migrations.
* Makes templates show up in Visual Studio
* Improved selenium start and tear down
* Selenium is set up and torn down in an assembly fixture.
* Selenium is initialized lazily and in a non-blocking way.
* Selenium processes are tracked as part of the build and their pids
written to a file on disk for cleanup in the event of unexpected
termination of the test process.
* Browser fixture retries with linear backoff to create a remote
driver. Under heavy load (like when we are doing a simultaneous NPM
restore) the selenium server can become unresponsive so we retry
three times, with a longer comand timeout allowance each time up to
a max of 3 minutes.
* Moved test project setup to build time instead of runtime.
* Added target PrepareForTest to create the required files for testing
* The template creation folder.
* The template props file to use our built packages.
* The folder for the custom hive.
* Added assembly metadata attributes to find all the data we need to
run the tests.
* Path to the artifacts shipping packages folder.
* Path to the artifacts non-shipping packages folder.
* Path to the test templates creation folder.
* Path to use for the custom templating hive used in tests.
* Proper cleanup as part of the build
* Remove the test templates creation folder.
* Remove the test packages restore path.
* Recreate the test templates creation folder.
* Recreate the test packages restore path.
* Generated Directory.Build.Props and Directory.Build.Targets in the
test templates creation folder.
* Cleaned up potentially stale templatetestsprops.
* Improved test flows
* Initialization is done lazily and asynchronously.
* Selenium
* Browser fixture
* Template initialization.
* Flattened test flows to avoid assertions inside deep callstacks.
* All assertions happen at the test level with improved error messages.
* With the exception of the migrations assertions.
* Assertions contain information about which step failed, for what
project and what failure details.
* Broke down tests to perform individual steps instead of mixing build
and publish.
* Publish project.
* Build project. (Debug)
* Run built project.
* Run published project.
* Concentrated build logic into the Project class.
* Context between the different steps of a test is maintained in
this class.
* All operations that require coordination are performed within this
class.
* There is a lock for dotnet and a lock for nodejs. When building
SPAs we acquire the nodejs lock to correctly prevent multiple
runs of nodejs in parallel.
[ApiAuthorization template cleanups]
* Fix preview3 issues with breaking changes on Entity framework by
manually configuring the model in ApiAuthorizationDbContext.
* Add app.db to the project file when using local db.
* Fix linting errors on angular template.
* Fix react tests
* Add tests to cover new auth options in the SPA templates.
* Unify the Templating and Components testing infrastructure.
* Enable test project discovery in the components E2E tests.
* Enable selectively disabling Selenium tests through build properties.
When a `<Reference>` is named `*.Sources`, set PrivateAssets=All (exclude from generated nuspec) and IncludeAssets=ContentFiles (only consume content files, not .dll's)
Changes:
* Remove obsolete targets which are unnecessary now that this repo no longer builds git submodules in a separate build process
* Remove the need for static analysis of 'ArtifactInfo' items
* Simplify how the code signing task is configured
* Remove unused repo tasks
* Remove duplicate lists of external dependencies and packages to be produced
* Remove obsolete build definition
* Remove obsolete build script parameters
* Add VisualStudioSetupOutputPath
Part of #4246
Changes:
* Update source code layout to follow the new conventions for this repo
* Update project files to use `<Reference>`
* Update targets to build NPM packages
* Update BuildTools to support custom 'restore' and 'test' targets
Stephen Halter
Brennan
Günther Foidl
John Luo
YPTopMan
Justin Kotalik
Michal Strehovský
Hao Kung
David Fowler
Dylan Dmitri Gray
Nate McMaster
Ben Adams
Javier Calvarro Nelson
Pavel Krymets
Ryan Nowak
Chris Ross
James Newton-King
Nate McMaster