I created a ViewComponent to display a List<Product>, the list is valorized taken data from a REST API service, this is my class implementation:
public class ProductsViewComponent : ViewComponent
{
private readonly HttpClient _client;
public ProductsViewComponent(HttpClient client)
{
_client = client ?? throw new ArgumentNullException(nameof(client));
}
public async Task<IViewComponentResult> InvokeAsync(string date)
{
using (var response = await _client.GetAsync($"/"product/get_products/{date}"))
{
response.EnsureSuccessStatusCode();
var products = await response.Content.ReadAsAsync<List<Product>>();
return View(products);
}
}
}
I load the List inside an html table which is available inside the Components folder: Views\Shared\Components\Products\Default.cshtml.
In each View that needs to display the Products I did:
#await Component.InvokeAsync("Products", new { date = myDate })
The REST API is called using the HttpClient configured in the Startup.cs as following:
services.AddHttpClient<ProductsViewComponent>(c =>
{
c.BaseAddress = new Uri('https://api.myservice.com');
});
This works well, but the main problem is each time the user reload the page or maybe go inside another View which require to display the list of products, then the app will make another API call.
Is possible store the list in something like a cache and prevent to call the API again if the date is equal than the previous date selected?
I'm learning ASP.NET Core so I'm not really expert on this argument.
Thanks in advance for any help.
As per microsoft documentation https://learn.microsoft.com/en-us/aspnet/core/performance/caching/memory?view=aspnetcore-2.1
you can use IMemoryCache to cache data
public class Startup
{
public void ConfigureServices(IServiceCollection services)
{
services.AddMemoryCache();
services.AddMvc().SetCompatibilityVersion(CompatibilityVersion.Version_2_1);
}
public void Configure(IApplicationBuilder app)
{
app.UseMvcWithDefaultRoute();
}
}
and create instance of IMemoryCache. This is an example from Microsoft documentation. You can Create another class to handle this all together and In below example this is just saving DateTime But, you can save any object in cache and when you try to read that value from cache just need to cast that object into a Type.
I will strongly recommend you go through the above documentation.
public class HomeController : Controller
{
private IMemoryCache _cache;
public HomeController(IMemoryCache memoryCache)
{
_cache = memoryCache;
}
public IActionResult CacheTryGetValueSet()
{
DateTime cacheEntry;
// Look for cache key.
if (!_cache.TryGetValue(CacheKeys.Entry, out cacheEntry))
{
// Key not in cache, so get data.
cacheEntry = DateTime.Now;
// Set cache options.
var cacheEntryOptions = new MemoryCacheEntryOptions()
// Keep in cache for this time, reset time if accessed.
.SetSlidingExpiration(TimeSpan.FromSeconds(3));
// Save data in cache.
_cache.Set(CacheKeys.Entry, cacheEntry, cacheEntryOptions);
}
return View("Cache", cacheEntry);
}
}
Update: CacheKeys.Entry is a static class where all keys are defined. (Just coding standards). Please check the above documentation link.
public static class CacheKeys
{
public static string Entry { get { return "_Entry"; } }
public static string CallbackEntry { get { return "_Callback"; } }
public static string CallbackMessage { get { return "_CallbackMessage"; } }
public static string Parent { get { return "_Parent"; } }
public static string Child { get { return "_Child"; } }
public static string DependentMessage { get { return "_DependentMessage";} }
public static string DependentCTS { get { return "_DependentCTS"; } }
public static string Ticks { get { return "_Ticks"; } }
public static string CancelMsg { get { return "_CancelMsg"; } }
public static string CancelTokenSource { get { return "_CancelTokenSource";} }
}
You can use a distributed cache and so use Redis for example with a ConnectionMultiplexer.
And so foreach call you can call your redis for the cache which is implement thanks to an interface call here 'IDistributedCache'
You can find a lot of documentation to implement cache and use it.
: .Net framework
DotNet Core
Your controller X :
[HttpGet]
[Route("{itemId}")]
public async Task<IHttpActionResult> GetItemById(int eventId, [FromUri]EventTabs tabId)
{
ServiceResponse<ItemDto> result = await _itemDispatcher.GetItemById(itemId);
return WrapResponse(result);
}
Your dispatcher to get the item by id which use redis cache (already implement)
public class ItemDispatcher : ItemDispatcher
{
private readonly IUnitOfWork _unitOfWork;
private readonly IDistributedCache _distributedCache; // use interface of your implementation of redis cache
private readonly int _cacheDuration;
private readonly bool _isCacheEnabled;
public EventDispatcher(IUnitOfWork unitOfWork, IDistributedCache distCache)
{
_unitOfWork = unitOfWork;
_distributedCache = distCache; // init cache in constructor
_cacheDuration = _configuration.Get<int>("cache.duration"); // duration of your cache
_isCacheEnabled = _configuration.Get<bool>("cache.isEnable"); // if the cache is enable or not
}
public async Task<ServiceResponse<ItemDto>> GetItemById(int id)
{
// Add this for each Task call
var cacheKey = string.Empty;
if (_isCacheEnabled)
{
cacheKey = CacheUtils.GetCacheKey(CacheKeys.Item, id);
itemDto cacheResult = await _distributedCache.Get<ItemDto>(cacheKey);
if (cacheResult != null)
return new ServiceResponse<Item>(cacheResult);
}
}
Try This
Cache["KeyName"] = VariableOrTable; Cache.Insert("Key", VariableOrTable, null,
Cache.NoAbsoluteExpiration, ts);
Related
I'm working on moving my API logic in my PATCH endpoint to a Mediatr Command. When applying my patch document, I usually check the model state like below. Normally, I'm doing this from a controller so there is no issue, but when moving this into a RequestHandler, I no longer have access to the model state property since I'm outside of the controller.
How would you recommend going about this?
Here is the model state logic I'd like to use outside of the controller:
updatePartialValueToReplaceCommand.PatchDoc.ApplyTo(valueToReplaceToPatch, ModelState); // apply patchdoc updates to the updatable valueToReplace
if (!TryValidateModel(valueToReplaceToPatch))
{
return ValidationProblem(ModelState);
}
The rest of the code for context:
Patch Endpoint
[HttpPatch("{valueToReplaceId}")]
public IActionResult PartiallyUpdateValueToReplace(int valueToReplaceId, JsonPatchDocument<ValueToReplaceForUpdateDto> patchDoc)
{
var query = new UpdatePartialValueToReplaceCommand(valueToReplaceId, patchDoc);
var result = _mediator.Send(query);
switch (result.Result.ToUpper())
{
case "NOTFOUND":
return NotFound();
case "NOCONTENT":
return NoContent();
default:
return BadRequest();
}
}
UpdatePartialValueToReplaceCommand
public class UpdatePartialValueToReplaceCommand : IRequest<string>
{
public int ValueToReplaceId { get; set; }
public JsonPatchDocument<ValueToReplaceForUpdateDto> PatchDoc { get; set; }
public UpdatePartialValueToReplaceCommand(int valueToReplaceId, JsonPatchDocument<ValueToReplaceForUpdateDto> patchDoc)
{
ValueToReplaceId = valueToReplaceId;
PatchDoc = patchDoc;
}
}
(BROKEN) UpdatePartialValueToReplaceHandler
public class UpdatePartialValueToReplaceHandler : IRequestHandler<UpdatePartialValueToReplaceCommand, string>
{
private readonly IValueToReplaceRepository _valueToReplaceRepository;
private readonly IMapper _mapper;
public UpdatePartialValueToReplaceHandler(IValueToReplaceRepository valueToReplaceRepository
, IMapper mapper)
{
_valueToReplaceRepository = valueToReplaceRepository ??
throw new ArgumentNullException(nameof(valueToReplaceRepository));
_mapper = mapper ??
throw new ArgumentNullException(nameof(mapper));
}
public async Task<string> Handle(UpdatePartialValueToReplaceCommand updatePartialValueToReplaceCommand, CancellationToken cancellationToken)
{
if (updatePartialValueToReplaceCommand.PatchDoc == null)
{
return "BadRequest";
}
var existingValueToReplace = _valueToReplaceRepository.GetValueToReplace(updatePartialValueToReplaceCommand.ValueToReplaceId);
if (existingValueToReplace == null)
{
return "NotFound";
}
var valueToReplaceToPatch = _mapper.Map<ValueToReplaceForUpdateDto>(existingValueToReplace); // map the valueToReplace we got from the database to an updatable valueToReplace model
updatePartialValueToReplaceCommand.PatchDoc.ApplyTo(valueToReplaceToPatch, ModelState); // apply patchdoc updates to the updatable valueToReplace -- THIS DOESN'T WORK IN A MEDIATR COMMAND BECAUSE I DON'T HAVE CONTROLLERBASE CONTEXT
if (!TryValidateModel(valueToReplaceToPatch))
{
return ValidationProblem(ModelState);
}
_mapper.Map(valueToReplaceToPatch, existingValueToReplace); // apply updates from the updatable valueToReplace to the db entity so we can apply the updates to the database
_valueToReplaceRepository.UpdateValueToReplace(existingValueToReplace); // apply business updates to data if needed
_valueToReplaceRepository.Save(); // save changes in the database
return "NoContent";
}
}
If your command handler is dependent on more information than it received in the command in order to act on it, then you're not providing enough information in the command; in this case, if you need to apply an operation based on ModelState, then you'd need to include and pass ModelState in the command.
Whether you can effectively do that or not, I'd more deeply call into question the need to use MediatR or some form of command bus here in the first place; you're synchronously performing an operation and waiting on the response, plus the handler is attempting to perform a lot of behavior (repository get, model validation, repository save), so although you've reduced the amount of code in the controller, you've really only shifted it to a new place that is still tightly coupled and now just obfuscates the dependencies of the controller.
The behavior you've packed into the Controller -> Command -> Handler and back seems like it would be just as well served by some form of provider (or likely multiple providers) injected into your controller via dependency injection; you can use an interface to keep your code flexible and your dependencies obvious while still reducing the grunt work being done within the controller code itself to help keep it clean by instead calling the (still abstracted) descriptive methods that express the intent.
Update 1
This is not a fully conceptualized example, but hopefully illustrative. If you wanted to use the command bus for cross-cutting concerns, there's still room to do so, but only after you've performed input validation, etc. No need to pass any of the controller state anymore.
public class YourController : Controller
{
private readonly ILogger<YourController> _logger;
private readonly IModelPatcher<SomeInput, SomeOutput> _modelPatcher;
private readonly IWriteRepository<SomeOutput> _writeRepository;
public YourController(ILogger<YourController> logger, IModelPatcher<SomeInput, SomeOutput> modelPatcher, IWriteRepository<SomeOutput> writeRepository)
{
_logger = logger;
_modelPatcher = modelPatcher;
_writeRepository = writeRepository;
}
[HttpPatch("{valueToReplaceId}")]
public IActionResult PartiallyUpdateValueToReplace(int valueToReplaceId, JsonPatchDocument<SomeInput> patchDoc)
{
if (patchDoc == null) return BadRequest();
var result = _modelPatcher.ApplyPatch(patchDoc, valueToReplaceId);
if (result == null) return NotFound();
if (!TryValidateModel(result)) return ValidationProblem(ModelState);
// var mapToDto = _mapper.Map(result); // maybe even here, before the repo...
_writeRepository.Update(result); // <-- This could be a command! Model is ready, validation is done.
return NoContent();
}
}
public class SomeInput { }
public class SomeOutput { }
public interface IModelPatcher<in TInput, out TResult>
{
TResult ApplyPatch(JsonPatchDocument<TInput> inputModel, int value);
}
public class SomeInputModelPatcher : IModelPatcher<SomeInput, SomeOutput>
{
private readonly IReadRepository<Something> _repository;
public SomeInputModelPatcher(IReadRepository<Something> repository)
{
_repository = repository;
}
public SomeOutput ApplyPatch(JsonPatchDocument<SomeInput> inputModel, int value)
{
// Do the patch related work
return new SomeOutput();
}
}
For those that are interested, here's what I ended up doing. Also got rid of those annoying magic strings!
[HttpPatch("{valueToReplaceId}")]
public IActionResult PartiallyUpdateValueToReplace(int valueToReplaceId, JsonPatchDocument<ValueToReplaceForUpdateDto> patchDoc)
{
var query = new UpdatePartialValueToReplaceCommand(valueToReplaceId, patchDoc, this);
var result = _mediator.Send(query);
return result.Result;
}
public class UpdatePartialValueToReplaceCommand : IRequest<IActionResult>
{
public int ValueToReplaceId { get; set; }
public JsonPatchDocument<ValueToReplaceForUpdateDto> PatchDoc { get; set; }
public Controller Controller { get; set; }
public UpdatePartialValueToReplaceCommand(int valueToReplaceId, JsonPatchDocument<ValueToReplaceForUpdateDto> patchDoc,
Controller controller)
{
ValueToReplaceId = valueToReplaceId;
PatchDoc = patchDoc;
Controller = controller;
}
}
public class UpdatePartialValueToReplaceHandler : IRequestHandler<UpdatePartialValueToReplaceCommand, IActionResult>
{
private readonly IValueToReplaceRepository _valueToReplaceRepository;
private readonly IMapper _mapper;
public UpdatePartialValueToReplaceHandler(IValueToReplaceRepository valueToReplaceRepository
, IMapper mapper)
{
_valueToReplaceRepository = valueToReplaceRepository ??
throw new ArgumentNullException(nameof(valueToReplaceRepository));
_mapper = mapper ??
throw new ArgumentNullException(nameof(mapper));
}
public async Task<IActionResult> Handle(UpdatePartialValueToReplaceCommand updatePartialValueToReplaceCommand, CancellationToken cancellationToken)
{
if (updatePartialValueToReplaceCommand.PatchDoc == null)
{
return updatePartialValueToReplaceCommand.Controller.BadRequest();
}
var existingValueToReplace = _valueToReplaceRepository.GetValueToReplace(updatePartialValueToReplaceCommand.ValueToReplaceId);
if (existingValueToReplace == null)
{
return updatePartialValueToReplaceCommand.Controller.NotFound();
}
var valueToReplaceToPatch = _mapper.Map<ValueToReplaceForUpdateDto>(existingValueToReplace); // map the valueToReplace we got from the database to an updatable valueToReplace model
updatePartialValueToReplaceCommand.PatchDoc.ApplyTo(valueToReplaceToPatch, updatePartialValueToReplaceCommand.Controller.ModelState); // apply patchdoc updates to the updatable valueToReplace
if (!updatePartialValueToReplaceCommand.Controller.TryValidateModel(valueToReplaceToPatch))
{
return updatePartialValueToReplaceCommand.Controller.ValidationProblem(updatePartialValueToReplaceCommand.Controller.ModelState);
}
_mapper.Map(valueToReplaceToPatch, existingValueToReplace); // apply updates from the updatable valueToReplace to the db entity so we can apply the updates to the database
_valueToReplaceRepository.UpdateValueToReplace(existingValueToReplace); // apply business updates to data if needed
_valueToReplaceRepository.Save(); // save changes in the database
return updatePartialValueToReplaceCommand.Controller.NoContent();
}
}
I have a API controller,and the scenario is:
I need to consume third party datasource(let's say the third party is provided as a dll file for simplicity, and the dll contain Student model and StudentDataSource that contain a lot of method to retrieve student ), and calling the third party data source is costly and data only gets updated every 6 hours.
so somehow I need to cache the output, below is some action method from my api controller:
// api controller that contain action methods below
[HttpGet]
public JsonResult GetAllStudentRecords()
{
var dataSource = new StudentDataSource();
return Json(dataSource.GetAllStudents());
}
[HttpGet("{id}")]
public JsonResult GetStudent(int id)
{
var dataSource = new StudentDataSource();
return Json(dataSource.getStudent(id));
}
then how should I cache the result especially for the second action method, it is dumb to cache every student result with different id
My team is implementing a similar caching strategy on an API controller using a custom Action filter attribute to handle the caching logic. See here for more info on Action filters.
The Action filter's OnActionExecuting method runs prior to your controller method, so you can check whether the data you're looking for is already cached and return it directly from here, bypassing the call to your third party datasource when cached data exists. We also use this method to check the type of request and reset the cache on updates and deletes, but it sounds like you won't be modifying data.
The Action filter's OnActionExecuted method runs immediately AFTER your controller method logic, giving you an opportunity to cache the response object before returning it to the client.
The specifics of how you implement the actual caching are harder to provide an answer for, but Microsoft provides some options for in-memory caching in .NET Core (see MemoryCache.Default not available in .NET Core?)
I used the solution with the cache strategy through the controller API as #chris-brenberg pointed out, it turned out like this
on controller class
[ServerResponseCache(false)]
[HttpGet]
[Route("cache")]
public ActionResult GetCache(string? dateFormat) {
Logger.LogInformation("Getting current datetime");
return Ok(new { date = DateTime.Now.ToString() });
}
on ServerResponseCacheAttribute.cs
namespace Site.Api.Filters {
using Microsoft.AspNetCore.Http.Extensions;
using Microsoft.AspNetCore.Mvc;
using Microsoft.AspNetCore.Mvc.Filters;
using Microsoft.Extensions.Caching.Memory;
using Microsoft.Extensions.Primitives;
using System.Globalization;
using System.Threading.Tasks;
public class ServerResponseCacheAttribute : TypeFilterAttribute {
public ServerResponseCacheAttribute(bool byUserContext = true) : base(typeof(ServerResponseCacheAttributeImplementation)) =>
Arguments = new object[] { new ServerResponseCacheProps { ByUserContext = byUserContext } };
public ServerResponseCacheAttribute(int secondsTimeout, bool byUserContext = true) : base(typeof(ServerResponseCacheAttributeImplementation)) =>
Arguments = new object[] { new ServerResponseCacheProps { SecondsTimeout = secondsTimeout, ByUserContext = byUserContext } };
public class ServerResponseCacheProps {
public int? SecondsTimeout { get; set; }
public bool ByUserContext { get; set; }
}
public class ServerResponseCacheConfig {
public bool Disabled { get; set; }
public int SecondsTimeout { get; set; } = 60;
public string[] HeadersOnCache { get; set; } = { "Accept-Language" };
}
private class ServerResponseCacheAttributeImplementation : IAsyncActionFilter {
private string _cacheKey = default;
readonly ILogger<ServerResponseCacheAttributeImplementation> _logger;
readonly IMemoryCache _memoryCache;
readonly ServerResponseCacheConfig _config;
readonly bool _byUserContext;
public ServerResponseCacheAttributeImplementation(ILogger<ServerResponseCacheAttributeImplementation> logger,
IMemoryCache memoryCache, ServerResponseCacheProps props) {
_logger = logger;
_memoryCache = memoryCache;
_byUserContext = props.ByUserContext;
_config = new ServerResponseCacheConfig {
SecondsTimeout = props.SecondsTimeout ?? 60,
HeadersOnCache = new[] { "Accept-Language" }
};
}
public async Task OnActionExecutionAsync(ActionExecutingContext context, ActionExecutionDelegate next) {
if (context == null) {
throw new ArgumentNullException(nameof(context));
}
if (next == null) {
throw new ArgumentNullException(nameof(next));
}
if (_config.Disabled) {
await next();
return;
}
OnActionExecutingAsync(context);
if (context.Result == null) {
OnActionExecuted(await next());
}
}
void OnActionExecutingAsync(ActionExecutingContext context) {
SetCacheKey(context.HttpContext.Request);
// Not use a stored response to satisfy the request. Will regenerates the response for the client, and updates the stored response in its cache.
bool noCache = context.HttpContext.Request.Headers.CacheControl.Contains("no-cache");
if (noCache) {
return;
}
TryLoadResultFromCache(context);
}
void SetCacheKey(HttpRequest request) {
if (request == null) {
throw new ArgumentException(nameof(request));
}
if (!string.Equals(request.Method, "GET", StringComparison.InvariantCultureIgnoreCase)) {
return;
}
List<string> cacheKeys = new List<string>();
if (_byUserContext && request.HttpContext.User.Identity.IsAuthenticated) {
cacheKeys.Add($"{request.HttpContext.User.Identity.Name}");
}
string uri = UriHelper.BuildAbsolute(request.Scheme, request.Host, request.PathBase, request.Path, request.QueryString);
cacheKeys.Add(uri);
foreach (string headerKey in _config.HeadersOnCache) {
StringValues headerValue;
if (request.Headers.TryGetValue(headerKey, out headerValue)) {
cacheKeys.Add($"{headerKey}:{headerValue}");
}
}
_cacheKey = string.Join('_', cacheKeys).ToLower();
}
void TryLoadResultFromCache(ActionExecutingContext context) {
ResultCache resultCache;
if (_cacheKey != null && _memoryCache.TryGetValue(_cacheKey, out resultCache)) {
_logger.LogInformation("ServerResponseCache: Response loaded from cache, cacheKey: {cacheKey}, expires at: {expiration}.", _cacheKey, resultCache.Expiration);
context.Result = resultCache.Result;
SetExpiresHeader(context.HttpContext.Response, resultCache.Expiration);
}
}
/// <summary>Add expires header (the time after which the response is considered stale).</summary>
void SetExpiresHeader(HttpResponse response, DateTimeOffset expiration) {
string expireHttpDate = expiration.UtcDateTime.ToString("ddd, dd MMM yyyy HH:mm:ss", CultureInfo.InvariantCulture);
response.Headers.Add("Expires", $"{expireHttpDate} GMT");
}
void OnActionExecuted(ActionExecutedContext context) {
if (_cacheKey == null) {
return;
}
if (context.Result != null) {
DateTimeOffset expiration = SetCache(context.Result);
SetExpiresHeader(context.HttpContext.Response, expiration);
} else {
RemoveCache();
}
}
DateTimeOffset SetCache(IActionResult result) {
DateTimeOffset absoluteExpiration = DateTimeOffset.Now.AddSeconds(_config.SecondsTimeout);
ResultCache resultCache = new ResultCache {
Result = result,
Expiration = absoluteExpiration
};
_memoryCache.Set(_cacheKey, resultCache, absoluteExpiration);
_logger.LogInformation("ServerResponseCache: Response set on cache, cacheKey: {cacheKey}, until: {expiration}.", _cacheKey, absoluteExpiration);
return absoluteExpiration;
}
void RemoveCache() {
_memoryCache.Remove(_cacheKey);
_logger.LogInformation("ServerResponseCache: Response removed from cache, cacheKey: {cacheKey}.", _cacheKey);
}
}
private class ResultCache {
public IActionResult Result { get; set; }
public DateTimeOffset Expiration { get; set; }
}
}}
I hope it helps someone, best regards
I am trying to add specific properties to telemetry request for every route.
After digging a bit, I've found that I can create my own custom TelemetryInitializer by implementing ITelemetryInitializer.
By doing this I've managed to add global properties to the request.
However, I still need to add specific properties at the controller level.
Do you have any idea how can I achieve this?
I've tried to inject TelemetryClient into the controller, but if I use it the properties are shared between requests.
This is how I've tried to log in the controller:
private TelemetryClient telemetryClient;
public ValueController(TelemetryClient telemetryClient)
{
this.telemetryClient = telemetryClient;
}
[HttpGet]
public async Task<IActionResult> RouteOne([FromQuery(Name = "param1")]string param1, [FromQuery(Name = "param2")]string param2)
{
telemetryClient.Context.GlobalProperties["param1"] = param1;
telemetryClient.Context.GlobalProperties["param2"] = param2;
}
[HttpGet]
public async Task<IActionResult> RouteTwo([FromQuery(Name = "param3")]string param3, [FromQuery(Name = "param4")]string param4)
{
telemetryClient.Context.GlobalProperties["param3"] = param3;
telemetryClient.Context.GlobalProperties["param4"] = param4;
}
And this is the implementation of ITelemetryInitializer:
public class CustomPropertiesTelemetryInitializer : ITelemetryInitializer
{
private readonly IHttpContextAccessor httpContextAccessor;
public CustomPropertiesTelemetryInitializer(IHttpContextAccessor httpContextAccessor)
{
this.httpContextAccessor = httpContextAccessor;
}
public void Initialize(ITelemetry telemetry)
{
telemetry.Context.GlobalProperties["RequestId"] = httpContextAccessor.HttpContext.GetProperty("requestId");
telemetry.Context.GlobalProperties["Ip"] = httpContextAccessor.HttpContext?.Connection.RemoteIpAddress.ToString();
telemetry.Context.GlobalProperties["RoutePath"] = httpContextAccessor.HttpContext?.Request.Path;
}
}
If the properties you added are always like "paramxxx", then there is a workaround(but it's really not very elegant).
In the controller constructor, check the GlobalProperties if it contains key like "paramxxx":
public ValueController(TelemetryClient telemetryClient)
{
this.telemetryClient = telemetryClient;
var props = this.telemetryClient.Context.GlobalProperties;
foreach (var p in props)
{
if (p.Key.Contains("param"))
{
props.Remove(p.Key);
}
}
}
The key here is to use the DI framework. You can use it to get request-scoped data or services into your ITelemetryInitializer.
(These examples are based on the standard ASP.Net Dependency Injection framework. This pattern should work with any DI framework, but will need to be adjusted slightly.)
First, create a class to represent your request-scoped telemetry. I've used a simple DTO, but this could also be a service that knows how to fetch/generate the data itself. Register it using AddScoped. "Scoped" means that a new instance will be created for each HTTP request, and then that instance will be re-used within that request.
Because I used a DTO, I didn't bother with an interface--you should use an interface if the class contains any logic you'll want to mock in unit tests.
public class RequestScopedTelemetry
{
public string MyCustomProperty { get; set; }
}
services.AddScoped<RequestScopedTelemetry>();
Now, create the ITelemetryInitializer and register it as a singleton. App Insights will discover and use it through the DI framework.
class RequestScopedTelemetryInitializer : ITelemetryInitializer
{
readonly IHttpContextAccessor httpContextAccessor;
public RequestScopedTelemetryInitializer(IHttpContextAccessor httpContextAccessor)
=> this.httpContextAccessor = httpContextAccessor;
public void Initialize(ITelemetry telemetry)
{
// Attempt to resolve the request-scoped telemetry from the DI container
var requestScopedTelemetry = httpContextAccessor
.HttpContext?
.RequestServices?
.GetService<RequestScopedTelemetry>();
// RequestScopedTelemetry is only available within an active request scope
// If no telemetry available, just move along...
if (requestScopedTelemetry == null)
return;
// If telemetry was available, add it to the App Insights telemetry collection
telemetry.Context.GlobalProperties[nameof(RequestScopedTelemetry.MyCustomProperty)]
= requestScopedTelemetry.MyCustomProperty;
}
}
services.AddSingleton<ITelemetryInitializer, RequestScopedTelemetryInitializer>();
Finally, in your controller method, set your per-request values. This part isn't necessary if your telemetry class is able to fetch or generate the data itself.
public class ExampleController : ControllerBase
{
readonly RequestScopedTelemetry telemetry;
public ValuesController(RequestScopedTelemetry telemetry)
=> this.telemetry = telemetry;
[HttpGet]
public ActionResult Get()
{
telemetry.MyCustomProperty = "MyCustomValue";
// Do what you want to
return Ok();
}
}
In order to add per request data into telemetry, you need to have a way to share data within the request. A reliable way is by using HttpContent.Items property, which is basically a Dictionary.
You can create a service to keep a Dictionary inside HttpContent.Items with all custom data you want in telemetry (key prefix is used to ensure we only read the things we want later in Initializer):
public class LogTelemetryRequest
{
private const string KEY_PREFIX = "CustomTelemetryData_";
private readonly IHttpContextAccessor _httpContextAccessor;
public LogTelemetryRequest(IHttpContextAccessor httpContextAccessor)
{
_httpContextAccessor = httpContextAccessor;
}
public void AddProperty(string key, string value)
{
_httpContextAccessor.HttpContext.Items[KEY_PREFIX + key] = value;
}
}
Register this as scoped in Startup.cs:
services.AddScoped<LogTelemetryRequest>();
Use it in your controller:
private LogTelemetryRequest logTelemetryRequest;
public ValueController(LogTelemetryRequest logTelemetryRequest)
{
this.logTelemetryRequest = logTelemetryRequest;
}
[HttpGet]
public async Task<IActionResult> RouteOne([FromQuery(Name = "param1")]string param1, [FromQuery(Name = "param2")]string param2)
{
// telemetryClient.Context.GlobalProperties["param1"] = param1;
// telemetryClient.Context.GlobalProperties["param2"] = param2;
logTelemetryRequest.AddProperty("param1", param1);
logTelemetryRequest.AddProperty("param2", param2);
}
Then read it within initializer:
public class AddCustomTelemetryInitializer : ITelemetryInitializer
{
private const string KEY_PREFIX = "CustomTelemetryData_";
private readonly IHttpContextAccessor _httpContextAccessor;
public AddCustomTelemetryInitializer(IHttpContextAccessor httpContextAccessor)
{
_httpContextAccessor = httpContextAccessor;
}
public void Initialize(ITelemetry telemetry)
{
var requestTelemetry = telemetry as RequestTelemetry;
if (requestTelemetry == null) return;
foreach (var item in _httpContextAccessor.HttpContext.Items)
{
if (item.Key is string key && key.StartsWith(KEY_PREFIX))
requestTelemetry.Properties.Add(key, item.Value.ToString());
}
}
}
Ideally LogTelemetryRequest should be registered using an interface, and the key prefix should be a single shared constant, didn't do for the sake of simplicity.
I have some problems with correctly working HttpContext.Current.User.Identity. From Controller constructor this doesn't work, I have to implement this to some method. Look at this example.
public class SomeControler : ApiController
{
private UserData userData;
// NOT WORKING
public ChartsController(
RegisteredUserData registeredUserData,
NotLoggedInUserData NotLoggedInUserData
{
var isAuthenticated = HttpContext.Current.User.Identity.IsAuthenticated;
this.userData = isAuthenticated
? (IUserData)registeredUserData
: (IUserData)NotLoggedInUserData;
}
// WORKING
public SomeMethod(
RegisteredUserData registeredUserData,
NotLoggedInUserData NotLoggedInUserData
{
var isAuthenticated = HttpContext.Current.User.Identity.IsAuthenticated;
this.userData = isAuthenticated
? (IUserData)registeredUserData
: (IUserData)NotLoggedInUserData;
}
}
How I can fix this? I spent a lot of time for answer in web but i didnt get this.
Regards.
edit
I found an answer. Is it good solution ?
public class SomeControler : ApiController
{
private RegisteredUserData registeredUserData;
private NotLoggedInUserData notLoggedInUserData;
private UserData userData
{
get
{
if (HttpContext.Current.User.Identity.IsAuthenticated)
{
return registeredUserData;
}
return notLoggedInUserData;
}
}
public ChartsController(
RegisteredUserData registeredUserData,
NotLoggedInUserData notLoggedInUserData
{
this.registeredUserData = registeredUserData;
this.notLoggedInUserData = notLoggedInUserData;
}
}
First, the request and HttpContext is not available yet in the construct of the controller because of where in the request flow the controller is initialized. You have to access it in an action where by then, the request and context would have been fully realized.
Next do not couple your controllers to HttpContext. It makes your code difficult to test and maintain.
Extract the desired information in a service abstraction.
public interface IUserDataAccessor {
IUserData UserData { get; }
}
public class UserDataAccessor : IUserDataAccessor {
private readonly RegisteredUserData registeredUserData;
private readonly NotLoggedInUserData notLoggedInUserData;
public UserDataAccessor(
RegisteredUserData registeredUserData,
NotLoggedInUserData notLoggedInUserData) {
this.registeredUserData = registeredUserData;
this.notLoggedInUserData = notLoggedInUserData;
}
public IUserData UserData {
get {
if (HttpContext.Current?.User?.Identity?.IsAuthenticated) {
return registeredUserData;
}
return notLoggedInUserData;
}
}
}
This allows the controller to remain lean with just the dependency on the abstraction.
public class ChartsController : ApiController {
private readonly IUserDataAccessor accessor;
public ChartsController(IUserDataAccessor accessor) {
this.accessor = accessor;
}
[HttpGet]
public IHttpActionResult SomeAction() {
var userData = accessor.UserData;
//...do something associated with user data
return OK();
}
}
Finally make sure that the abstraction and it's implementation a registered with your dependency container in your composition root.
So, there's a bug in some legacy code I'm maintaining. It causes some mild data corruption, so it's rather serious. I've found the root cause, and have made a sample application that reliable reproduces the bug. I would like to fix it with as little impact on existing applications as possible, but I'm struggling.
The bug lies in the data access layer. More specifically, in how an interceptor is injected into a new Nhibernate Session. The interceptor is used to set a specific entity property when saving or flushing. The property, LoggedInPersonID, is found on nearly all our entities. All entities are generated from CodeSmith templates using the database schema, so the LoggedInPersonID property corresponds to a column that is found on nearly all tables in the database. Together with a couple of other columns and triggers, it is used to keep track of which user created and modified a record in the database. Any transaction that inserts or updates data need to supply a LoggedInPersonID value, or else the transaction will fail.
Whenever a client requires a new session, a call is made to OpenSession in the SessionFactory (not Nhibernate's SessionFactory, but a wrapper). The code below shows the relevant parts of the SessionFactory wrapper class:
public class SessionFactory
{
private ISessionFactory sessionFactory;
private SessionFactory()
{
Init();
}
public static SessionFactory Instance
{
get
{
return Nested.SessionFactory;
}
}
private static readonly object _lock = new object();
public ISession OpenSession()
{
lock (_lock)
{
var beforeInitEventArgs = new SessionFactoryOpenSessionEventArgs(null);
if (BeforeInit != null)
{
BeforeInit(this, beforeInitEventArgs);
}
ISession session;
if (beforeInitEventArgs.Interceptor != null
&& beforeInitEventArgs.Interceptor is IInterceptor)
{
session = sessionFactory.OpenSession(beforeInitEventArgs.Interceptor);
}
else
{
session = sessionFactory.OpenSession();
}
return session;
}
}
private void Init()
{
try
{
var configuration = new Configuration().Configure();
OnSessionFactoryConfiguring(configuration);
sessionFactory = configuration.BuildSessionFactory();
}
catch (Exception ex)
{
Console.Error.WriteLine(ex.Message);
while (ex.InnerException != null)
{
Console.Error.WriteLine(ex.Message);
ex = ex.InnerException;
}
throw;
}
}
private void OnSessionFactoryConfiguring(Configuration configuration)
{
if(SessionFactoryConfiguring != null)
{
SessionFactoryConfiguring(this, new SessionFactoryConfiguringEventArgs(configuration));
}
}
public static event EventHandler<SessionFactoryOpenSessionEventArgs> BeforeInit;
public static event EventHandler<SessionFactoryOpenSessionEventArgs> AfterInit;
public static event EventHandler<SessionFactoryConfiguringEventArgs> SessionFactoryConfiguring;
public class SessionFactoryConfiguringEventArgs : EventArgs
{
public Configuration Configuration { get; private set; }
public SessionFactoryConfiguringEventArgs(Configuration configuration)
{
Configuration = configuration;
}
}
public class SessionFactoryOpenSessionEventArgs : EventArgs
{
private NHibernate.ISession session;
public SessionFactoryOpenSessionEventArgs(NHibernate.ISession session)
{
this.session = session;
}
public NHibernate.ISession Session
{
get
{
return this.session;
}
}
public NHibernate.IInterceptor Interceptor
{
get;
set;
}
}
/// <summary>
/// Assists with ensuring thread-safe, lazy singleton
/// </summary>
private class Nested
{
internal static readonly SessionFactory SessionFactory;
static Nested()
{
try
{
SessionFactory = new SessionFactory();
}
catch (Exception ex)
{
Console.Error.WriteLine(ex);
throw;
}
}
}
}
The interceptor is injected through the BeforeInit event. Below is the interceptor implementation:
public class LoggedInPersonIDInterceptor : NHibernate.EmptyInterceptor
{
private int? loggedInPersonID
{
get
{
return this.loggedInPersonIDProvider();
}
}
private Func<int?> loggedInPersonIDProvider;
public LoggedInPersonIDInterceptor(Func<int?> loggedInPersonIDProvider)
{
SetProvider(loggedInPersonIDProvider);
}
public void SetProvider(Func<int?> provider)
{
loggedInPersonIDProvider = provider;
}
public override bool OnFlushDirty(object entity, object id, object[] currentState, object[] previousState,
string[] propertyNames, NHibernate.Type.IType[] types)
{
return SetLoggedInPersonID(currentState, propertyNames);
}
public override bool OnSave(object entity, object id, object[] currentState,
string[] propertyNames, NHibernate.Type.IType[] types)
{
return SetLoggedInPersonID(currentState, propertyNames);
}
protected bool SetLoggedInPersonID(object[] currentState, string[] propertyNames)
{
int max = propertyNames.Length;
var lipid = loggedInPersonID;
for (int i = 0; i < max; i++)
{
if (propertyNames[i].ToLower() == "loggedinpersonid" && currentState[i] == null && lipid.HasValue)
{
currentState[i] = lipid;
return true;
}
}
return false;
}
}
Below is a helper class used by applications to register a BeforeInit event handler:
public static class LoggedInPersonIDInterceptorUtil
{
public static LoggedInPersonIDInterceptor Setup(Func<int?> loggedInPersonIDProvider)
{
var loggedInPersonIdInterceptor = new LoggedInPersonIDInterceptor(loggedInPersonIDProvider);
ShipRepDAL.ShipRepDAO.SessionFactory.BeforeInit += (s, args) =>
{
args.Interceptor = loggedInPersonIdInterceptor;
};
return loggedInPersonIdInterceptor;
}
}
}
The bug is especially prominent in our web services (WCF SOAP). The web services endpoint bindings are all basicHttpBinding. A new Nhibernate session is created for each client request. The LoggedInPersonIDInterceptorUtil.Setup method is called after a client is authenticated, with the authenticated client's ID captured in the closure. Then there's a race to reach code that triggers a call to SessionFactory.OpenSession before another client request registers an event handler to the BeforeInit event with a different closure - because, it's the last handler in the BeforeInit event's invocation list that "wins", potentially returning the wrong interceptor. The bug usually happens when two clients are making requests nearly simultaneously, but also when two clients are calling different web service methods with different execution times (one taking longer from authentication to OpenSession than another).
In addition to the data corruption, there's also a memory leak as the event handlers aren't de-registered? It might be the reason why our web service process is recycled at least once a day?
It really looks like the BeforeInit (and AfterInit) events need to go. I could alter the signature of the OpenSession method, and add an IInterceptor parameter. But this would break a lot of code, and I don't want to pass in an interceptor whenever a session is retrieved - I would like this to be transparent. Since the interceptor is a cross cutting concern in all applications using the DAL, would dependency injection be a viable solution? Unity is used in some other areas of our applications.
Any nudge in the right direction would be greatly appreciated :)
Instead of supplying the interceptor at each ISessionFactory.OpenSession call, I would use a single interceptor instance globally configured (Configuration.SetInterceptor()).
This instance would retrieve the data to use from an adequate context allowing to isolate this data per request/user/whatever suits the application.
(System.ServiceModel.OperationContext, System.Web.HttpContext, ..., depending on the application kind.)
The context data in your case would be set where LoggedInPersonIDInterceptorUtil.Setup is currently called.
If you need to use the same interceptor implementation for applications requiring different contextes, then you will need to choose the context to use according to some configuration parameter you would add (or inject it as a dependency in your interceptor).
Dependency Injection example:
DependencyInjectionInterceptor.cs:
using NHibernate;
using System;
using Microsoft.Extensions.DependencyInjection;
namespace MyAmazingApplication
{
public class DependencyInjectionInterceptor : EmptyInterceptor
{
private readonly IServiceProvider _serviceProvider;
public DependencyInjectionInterceptor(IServiceProvider serviceProvider)
{
_serviceProvider = serviceProvider;
}
public T GetService<T>() => _serviceProvider.GetService<T>();
public T GetRequiredService<T>() => _serviceProvider.GetRequiredService<T>();
}
}
Startup.cs:
public void ConfigureServices(IServiceCollection services)
{
...
var cfg = new Configuration();
... // your config setup
cfg.SetListeners(NHibernate.Event.ListenerType.PreInsert, new[] { new AuditEventListener() });
cfg.SetListeners(NHibernate.Event.ListenerType.PreUpdate, new[] { new AuditEventListener() });
services.AddSingleton(cfg);
services.AddSingleton(s => s.GetRequiredService<Configuration>().BuildSessionFactory());
services.AddScoped(s => s.GetRequiredService<ISessionFactory>().WithOptions().Interceptor(new DependencyInjectionInterceptor(s)).OpenSession());
... // you other services setup
}
AuditEventListener.cs:
public class AuditEventListener : IPreUpdateEventListener, IPreInsertEventListener
{
public bool OnPreUpdate(PreUpdateEvent e)
{
var user = ((DependencyInjectionInterceptor)e.Session.Interceptor).GetService<ICurrentUser>();
if (e.Entity is IEntity)
UpdateAuditTrail(user, e.State, e.Persister.PropertyNames, (IEntity)e.Entity, false);
return false;
}
}
So you use interceptor to get your scoped or any other service:
var myService = ((DependencyInjectionInterceptor)e.Session.Interceptor).GetService<IService>();
ICurrentUser in particular is a scoped service which uses HttpContext to get the current user.
I hope it might be helpful for everyone.