- remove tests already thoroughly covered in unit tests
- #3139 part 2 of 3
- make checks of `<form>` tag helper more consistent with others
- use a tag helper initializer but testing every variant; covered in unit tests
- remove `MvcStartupTest` and associated web site
- scenario covered elsewhere
- remove `PrecompilationTest.PrecompiledView_RendersCorrectly()`
- scenario covered elsewhere and shouldn't do I/O on checked-in files
- use `MvcTestFixture` instead of `TestHelper`
- part of #3139
- continue to test e2e interaction of `FlushAsync()` with sections, partials, et cetera
- remove `FlushPointTest.FlushBeforeCallingLayout()`
- `RazorPageTest.FlushAsync_ThrowsIfTheLayoutHasBeenSet()` unit test covers the scenario
- Added functional tests to validate data created from a `SelectTagHelper` does not impact following `<select>` tags.
- Also moved the new `SelectTagHelper` communication flow into `TagHelper.Init`.
#3347
- #3227
- much of change is to tests, creating and passing `ModelMetadata`
- updated `InputFormatterContext` to make `ModelMetadata` available to `JsonInputFormatter`
- walk `ModelMetadata` tree to get information about property with an issue
- add missing `null` checks in `ModelStateDictionaryExtensions`
- #2994
- Affects both HtmlHelper and TagHelper scenarios
- Checkboxes not enclosed in a form will generate inline hidden tags
- Added necessary properties to FormContext
- Added some functional and unit tests
- aspnet/Mvc#3138 part 2/2
- request's Content-Type header must be a subset of what an `IInputFormatter` can consume
- `[Consumes]` is similar
- what an `IOutputFormatter` produces must be a subset of the request's Accept header
- `FormatFilter` and `ObjectResult` are similar
- `ObjectResult` no longer falls back to `Content-Type` header if no `Accept` value is acceptable
- left `WebApiCompatShim` code alone for consistency with down-level `System.Net.Http.Formatting`
- correct tests to match new behaviour
- do not test `Accept` values containing a `charset` parameter; that case is not valid
WIP:
- four test failures; something about comparing media types w/ charset included
- why do some localization tests fail in VS?
nits:
- add `InputFormatterTests`
- add / update comments and doc comments
- correct xUnit attributes in `ActionResultTest`; odd it doesn't show up in command-line runs
Abstractions - Core MVC extensibility
Controllers - MVC implementations of .Abstractions and supporting
contracts
Infrastructure - General purpose support APIs. Metadata APIs that don't
fit clearly with a feature or with .Abstraction
- #3066
- reduce `TestServer` -> `PhysicalFileProvider` -> `FileSystemWatcher` count enough to run with Core CLR on Linux
- remove use of `HttpClient.DefaultRequestHeaders`; any client change affects other tests
- remove use of `RequestBuilder` class; creates a per-test `HttpClient` and requires the `TestServer`
- updated a few expectations because `CommonTestEncoder` does JavaScript a bit differently
- "JavaScriptEncode[[...]]" -> "JavaScriptStringEncode[[...]]"
- side benefit: xUnit reports functional tests execute for only ~12.4s; was >30s before this change
Infrastructure: Enhance `MvcTestFixture`
- handle `ConfigureServices()` methods that are not `void`
- handle `Configure(IApplicationBuilder, ILoggerFactory)`
- ensure server is initialized with consistent `CurrentCulture` and `CurrentUICulture`
- add `FilteredDefaultAssemblyProviderFixture<TStartup>` and `MvcEncodedTestFixture<TStartup>`
- add `MvcTextFixture.AddAdditionalServices()` extension point supporting these
- do not expose the `TestServer`; an anti-pattern for tests to manipulate the server
- update class names to match containing files
- use existing `TestApplicationEnvironment`
- apply some `MvcTestFixture` improvements to the shared `TestApplicationEnvironment` class
- remove unused methods from `TestHelper`
nits:
- touched-up some leftover `_app` &c declarations to be more explicit and minimize `using`s
- moved statements into correct sections of methods in `RoutingTests`
- removed `TestLoggerFactory` and related classes from `TagHelperSampleTest`
- use `[ConditionalFact]` or `[ConditionalTheory]` instead
- in some cases test was already conditional
- move other `IsMono` checks into data sets where possible
- not reported as skipped but at least not counted as successful
nit: switch from `[ConditionalTheory]` to `[ConditionalFact]` as appropriate
- slightly more efficient since xUnit won't check for data
- use consistent line endings on all platforms so that hashes and file versions match
- add another `.gitattribute` file
- move files to avoid required index recreation in all Linux repos
- The corresponding Razor change results in `HelperResult`s being rendered with async lambdas.
- This change enables `TagHelper`s and other async code to exist inside of `HelperResult` blocks.
- Added test to validate Templates (they generate `HelperResult`s) can utilize `TagHelper`s correctly.
- Rename `RazorPage`s `RenderBodyDelegate` to `RenderBodyDelegateAsync`.
aspnet/Razor#494
- `ScriptTagHelper`, `LinkTagHelper` and `ImageTagHelper` now default to using `output.Attributes["href|src"]` if it's present when they run. This enables other `TagHelper`s to run prior and add those attributes.
- Added unit tests to validate this behavior.
- Updated `ImageTagHelper` functional test resources. Now that we're always defaulting to `output.Attributes["src"]` for `ImageTagHelper.Src` we're properly copying attributes back into the `output.Attributes` collection in the correct order (isntead of appending to the end).
#2902
- Updated implementation to do a best guess at copying an attribute back into `TagHelperOutput.Attributes`.
- Updated existing functional and unit tests to account for maintained attribute order.
- Added a set of complex order based unit tests to validate `CopyHtmlAttribute` properly maintains order.
#2639
Doug Bunting
Pavel Krymets
N. Taylor Mullen
Chris R
Pranav K
Kiran Challa
Ajay Bhargav Baaskaran
Ryan Nowak
Kirthi Krishnamraju
Gerardo Melendrez
Stefán Jökull Sigurðarson
Hao Kung
Daniel Lo Nigro
sornaks
Sornakumar Sundararajan