I'm reading the OWIN 1.0 spec at http://owin.org/spec/owin-1.0.0.html and just can't wrap my head around how it works. I've downloaded Katana source, but that's huge and didn't help any. I'm familiar with the somewhat standard way of having a project/assembly with interfaces only, which allows to integrate two projects without direct regencies. But I can't understand how the web server will call into the web app with only Func<> and Action<> definitions.
OWIN boils down to two things:
1) an "environment" dictionary
2) a method that processes requests and sends responses.
For #1, this is just a property bag that gives you access to the request headers, request stream, response headers, response stream and server data. Think of this as your HttpContext for ASP.NET or HttpListenerContext for System.Net.HttpListener. In fact, in more recent builds of Katana (https://katanaproject.codeplex.com/, which is an open source implementation from the ASP.NET team, there have been improvements (more to come) to simplify this down to an easier to use object model, including an OwinRequest, OwinResponse, and IOwinContext.
For #2, this is often called the "AppFunc" and the signature is:
using AppFunc = Func<IDictionary<string, object>, Task>;
This signature is used for "Middleware" that is in a pipeline of request handlers or it can be the end application which is generating HTML, is a WebAPI, etc.
In Katana, there is a class you can inherit from that simplifies this signature to consume the IOwinContext I mentioned previously. Take at look at OwinMiddlware
You can also read this article which gives an overview of the Katana/OWIN effort: http://www.asp.net/aspnet/overview/owin-and-katana/an-overview-of-project-katana
OWIN just defines how the web server and web application will talk to each other. Your application must implement one side of this contact, the other side which connects to the web server must be provided by installing a NuGet package specific to the web server. There is one for IIS, one for self hosting (stand alone application) etc.
Related
I have had a read on what's new in .NET4.6 and one of the things is ASP.NET 5 which I am quite excited about.
One of the new things is New modular HTTP request pipeline, however there is no more info on how exactly is it going to change.
The only reference in the article is
ASP.NET 5 introduces a new HTTP request pipeline that is lean and
fast. This pipeline is modular so you can add only the components that
you need. By reducing the overhead in the pipeline, your app will
experience better throughput. The new pipeline also supports OWIN.
What are major differences between ASP.NET4.5 and ASP.NET5 Http pipelines? How modularity will be controlled?
The biggest difference in my opinion is the modularity of the new request pipeline. In the past, the application lifecycle followed a relatively strict path that you could hook into via classes implementing IHttpModule. This would allow you to affect the request, but only at certain points along the way by subscribing to the different events that occur (e.g. BeginRequest, AuthenticateRequest, etc.).
The full descriptions of these can be found on MSDN: IIS 5 & 6 or IIS 7, and a walkthrough of creating such a module can be found here.
In the new ASP.NET 5 world, the request pipeline is decoupled from System.Web and IIS. Instead of a pre-defined path, it uses the concept of middleware. If you are familiar with OWIN, the idea is nearly identical, but the basic idea is that these Middleware Components are registered and then the request passes through them in the order that they are registered.
Each middleware component is provided a RequestDelegate (the next middleware component in the pipeline) and the current HttpContext per-request. On each request, the component is invoked, and then has the opportunity to pass the request along to the next in the chain if applicable. For example, an authentication component might opt not to pass the request along to the next component if authentication fails. Using this system, you can really handle a request any way you choose, and can be as light-weight or as feature-rich as you need it to be.
This example is a little bit dated now (e.g. IBuilder has been renamed to IApplicationBuilder), but it is still a great overview of how building and registering these components looks.
I'm having problems because of a poorly written third-party library which our system heavily depends on. This library is not thread-safe (because of some bugs and static variables) and I need to use it in a ASP.NET webservice, which handles each user request in a separate thread.
I've tried many solutions for this problem. The best solution for now is, in my opinion, let subprocesses handle the requests. One subprocess will listen and handle the requests for one user, so I can synchronize access to the library code in a per user fashion, which is much better than all that I can do when sharing static variables between requests.
How can I route requests received by IPC communication to the appropriate WebMethods without reinventing the wheel? If possible, I would like to use the classes from .Net that handle this in a normal ASP.NET webservice, but I'm having a hard time trying to find their names.
TL;DR: I have a class MyWebService (that inherits from System.Web.Services.WebService) with some methods marked with WebMethodAttribute and I want to pass a made-up HttpRequest (or HttpContext) to it and tell it "handle it like you're receiving this from a real HTTP server, despite the fact the current process is a console application".
First, you may want to consider using WCF instead of ASMX, which is a legacy technology, kept only for backwards compatibility.
Second, you have another option: ensure that only a single thread ever uses the third-party libarary at a time. Placing lock blocks around all access to the third-party library may solve the problem.
We have a Flex application which relies heavily on data driven content supplied via asp.net. Currently the majority of this data is provided via asp.net objects which are then XML serialised and sent via a simple ASHX handler. This is then parsed via e4x in singleton classes to populate either its self or arrays of sub classes which are then available to the rest of the application without making additional data calls.
This works but is it the best way? I've read quite a few articles discussing the subject but couldn't really find any consensus.
Should I look into converting these to Web Services? If so, how should I manage the bindings, automatically import them via Flex or build my own? What are the pro's and con's. An important factor in this decision is speed, lowest latency and highest throughput is essential
As a separate matter our application doesn't sit at the root of the domain, and when in local development makes data calls to our development servers. As a result we add flash vars to the application to specify the appRoot which is then appended to the service url as necessary.
MyService.url = GeneralData.ApplicationRootUrl + "Services/foobar.ashx";
Is this the best way? I have since discovered the rootURL property, should I be using this, how does it work in this context? If I were to convert the services to web services how would I go about implementing the same functionality to allow local development?
Many thanks
This works but is it the best way?
Best is very subjective based on your situation. If at all possible, I would recommend you use an AMF gateway. That way your objects can immediately convert from server side objects (.NET Classes) to client side objects (AS3 classes). This is a big time savings because you don't have to manually create your XML on the back end, nor manually process it in the front end. Also the binary format of AMF is going to give much smaller packets than XML or a SOAP WebService would.
For .NET AMF options, I'd look into WebORB or FlourineFX
Flex Application is always loaded in browser, and you can use relative URL, so that your application will connect to same server from where it is loaded.
MyService.url = "/Services/foobar.ashx";
"/" will certainly append host where it came from. And it is always good practice to connect to same host where the flash is loaded from.
Secondly, SOAP web services use xml serialization, so if you use your handler to do e4x serialization or you use SOAP web service generator of Flash Builder, speed will be almost same. SOAP web service will certainly be little slower, but the difference will be in micro seconds to milli seconds.
However, with Web services, your development will speed improve as you will not have to create proxy classes.
Since Flashbuilder does not support WCF over https, i am considering to use weborb remoting as alternative, but not really sure how flash is going to know weborb location, if they are sitting on different servers. Looked at destination, source fields, but not really find a field called url in remoteObject in Flex. Has anyone done similar things?
I know this is an old question, but thought I'd answer it anyway. You can expose your WCF services to remoting clients (Flash, Flex) via WebORB. WebORB supports both self-host and IIS-hosted WCF services. Here are links to instructions for both models.
Self-hosted: http://www.themidnightcoders.com/fileadmin/docs/dotnet/v4/guide/index.html?standalone_wcf_services.htm
IIS-hosted: http://www.themidnightcoders.com/fileadmin/docs/dotnet/v4/guide/index.html?iis_hosted_wcf_services.htm
Both documents address your questions. Here is an example of one approach:
Invoking Self-Hosted Service From Flex/AIR
Flex and AIR clients can use the RemoteObject API to invoke methods on self-hosted WCF services which use the AMF endpoint. There are two approaches for invoking self-hosted WCF service. The first approach requires less code, but creates a dependency on configuration files declaring destinations and channels (the files located in WEB-INF/flex). The second approach does not have any dependencies on the configuration files, but results in a few additional lines of code.Consider the examples of the API below:
Approach 1 (with dependency on configuration files):
var remoteObject:RemoteObject = new RemoteObject("GenericDestination");
remoteObject.endpoint = "http://localhost:8000/WCFAMFExample/amf"
remoteObject.GetQuote.addEventListener( ResultEvent.RESULT, gotResult );
remoteObject.GetQuote.addEventListener( FaultEvent.FAULT, gotError );
remoteObject.GetQuote( "name" );
The endpoint URL uniquely identifies the WCF service. Notice the /amf at the end of the URL, it is required for the AMF endpoint. With the approach demonstrated above, the destination name in the RemoteObject constructor is required however it is not used. As a result, for the code to work, the Flex/AIR application must be compiled with additional compile argument:
-services "C:\Program Files\WebORB for .NET\4.0\web-inf\flex\services-config.xml"
I hope this helps.
K
I am aware of Web Services and WCF but I have generic question with services.
I have a ASP.NET MVC Application which does some basic functionality. I just have a controller in which I am passing it the records and serializing the information to XML using XML Serializer. Then I return this information to the browser and it displays me the XML i got from the Controller Action. So I get the XML representation of my Class(Database Object) in XML and I am to give the URL of this application to the client and access and pull the information.
Is this a Service then?
I mean in the end all the Clients need is the Xml representation through services also right? I am not that experienced and probably being very silly but please help me out...if I provide xml this way to the client is that a Service? Or is there something I need to understand here?
Don't let all the buzz about "web services" fool you; the basic idea behind a web service is very, very simple. It is simply a matter of providing data in response to a request over standard web transport protocols (i.e., HTTP/HTTPS). Everything else (XML, SOAP, WSDL, etc.) is just layered technology to augment the basic functionality of a service. REST-based services are very basically the simplest services that you can build--they are built on the core HTTP/S transport protocol and not much else.
The main differentiator between a service and traditional web site is that a service is data-focussed rather than presentation-focussed; that is, services typically are not concerned with how data is formatted and displayed (that is up to the client), but rather what data is returned. So...you are delivering XML data over HTTP? Check. You have a service. Congratulations!
Yes, it is a service, returning an XML resource. Also it seems that it is accessible throught standard HTTP verbs such as GET, so one might assume it is RESTful. The difference with a standard SOAP XML service is that you don't have a WSDL that describes it, so you might need to provide a good documentation to clients wishing to consume your service.