The fix splits client validation and model validation into two separate hierarchies.
Introduced ClientModelValidatorProvider in MvcOptions, which can be iterated to produce IClientModelValidators.
As a result of this, HtmlGenerator code can be free of ActionBindingContext and directly consumes options.
This also means that we do not modify the client validations during resource filters.
- set correct `MarkAsHtmlEncodedMethodName` value in `MvcRazorHost`
- handle `HtmlString` values in `TagHelperOutput.Attributes` in `RazorPage`
- special-case double-quotes in `HtmlString` values
- add `static WriteTo()` method for use in tag helpers
- handle non-`string` `output.Attributes` values in tag helpers
- make `TagHelperContentWrapperTextWriter` a `public` class
- provide a `TagHelperContent.Append(object, ...)` extension method
- add `LinkTagHelper.Href` and `ScriptTagHelper.Src` properties
- avoid Razor HTML-encoding these attribute values before their use
- add `JavaScriptEncoder` properties in `LinkTagHelper` and `ScriptTagHelper`
- allow encoding testing without unit testing the default encoder
- handle MVC and Razor changes for this bug in existing tests
- add functional tests of encodings
- add test encoders to TestCommon project
nits:
- correct `InputTagHelper` to pass `type=""` through unchanged
- set correct `TagHelperContentTypeName` value in `MvcRazorHost`
- remove unnecessary `FormTagHelper.Method` and `OptionTagHelper.Selected` properties
- remove complex ternaries and `ShouldAddFileVersion()` methods
- add a few debug assertions
- fix some odd wrapping
- remove or `#if`-out unused `using`s
- remove trailing whitespace
The DataAnnotationsMetadataProvider was setting the bool? IsRequired, all of the
time instead of only setting it to true when we found a RequiredAttribute.
So we never actually executed the fallback logic here. Found
this while working on removing some reflection code from the validator,
and wanted to split it out because it's simple.
This change removes reflection from validator providers, and instead
relies on cached metadata in in the modelmetadata.
In general this means that our MVPs don't need to cache anything, they
just look at the metadata and create what they need.
In the case of data-annotations, we update the model details provider to
add validation attributes to the modelmetadata. This would allow someone
to replace the DataAnnotationsValidatorProvider, but still use the
metadata in these attributes.
The change to the IModelValidatorProvider api (to use a context) is
intended to minimize allocations. Currently each validator provider needs
to return a list so you end up with N+1 lists (N validators + a final list
to compine them all). This change will let us just create the final list
(and a small context object). This is a very very high traffic API so it
seemed worth doing.
There's also some general massaging of namespaces and file locations.