This is a fix for the specific case reported in #3743 and also a few
related cases where the object being validated does not come from user
input.
Because the CancellationToken is bound using the 'empty prefix',
suppressing validation causes the validation system to suppress ALL keys
in the MSD. This will wipe out validation info for other, unrelated model
data. The same can occur for [FromServices], IFormCollection, IFormFile,
and HttpRequestMessage (anythere when MSD key == null and
SuppressValidation == true).
The other change here is related to the [FromBody]. We don't want to
create an entry in the model state for a 'valid' model with the empty
prefix. This can cause the system to miss validation errors. If you end up
in a situation where there's an entry with the empty prefix for an invalid
state, you won't accidentally miss validation errors.
- #2969
- `RemoteAttribute` did not support `IStringLocalizer` overrides
- use same `MvcDataAnnotationsLocalizationOptions` property as for other `ValidationAttribute`s
- error message `NumericClientModelValidator` added could not be overridden
- not related to `IStringLocalizer` because users have no way to set the resource lookup key
- extend `IModelBindingMessageProvider` to add the necessary `Func<,>`
- also correct problem using resources with `RemoteAttribute` and add lots of tests
* Add an UnsupportedContentType to the ModelState dictionary when no formatter can read the body.
* Add a filter to the pipeline that searches for that specific exception and transforms the response into 415.
This change removes a layer of abstraction. These validators now just do
what they do now without creating a bunch of intermediate objects.
ModelClientValidationRule has been removed, and client validations
manipulate the attributes collection of a tag directly.
* Moved instantiation of view components into DefaultViewComponentActivator
* Introduced IViewComponentFactory to handling setup of new view component instances.
* Added a release method on IViewComponentActivator for handling tear down of view
component instances.
* Added a Release method to IControllerActivator
* Changed Create in IControllerActivator to take in a ControllerActionContext
* Move the check to determine if a controller can be instantiated into the controller activator.
* Move logic for disposing controllers into the controller activator and make release on the
controller factory delegate into the activator.
* Changed release methods to take in a controller context.
- have one test muted at the moment in our CI
- add `Reset()` action to the service, enabling `// Arrange` to bring `Index()` content to a known state
- add a couple more tests to `TagHelperSampleTest` covering the new action and empty `Index()` list
nits:
- use `ViewData`, not `ViewBag`
These changes are aimed at significantly improving the performance of
MVC/Razor when a large amount of content is in play or a large number of
TagHelpers are used.
A few issues addressed:
- Buffer sync writes after a flush has occurred so that we can write them
asyncronously. The issue is that an IHtmlContent can only do sync
writes. This is very bad for Kestrel in general. Doing these writes
async is better for our overall perf, and the buffer that we use for it
is from the pool.
- 'Flatten' ViewBuffers when possible. A page with lots of TagHelpers can
end up renting a ViewBuffer and only write 2-3 things into it. When a
ViewBuffer sees another ViewBuffer we can either steal its pages, or
copy data out and 'return' its pages. This lets us use 3-4 buffers for a
large Razor page instead of hundreds.
Doug Bunting
Ryan Nowak
jacalvar
Ajay Bhargav Baaskaran
Cesar Blum Silveira
Derek Gray
ryanbrandenburg
Brennan
N. Taylor Mullen
Pavel Krymets
Pranav K