I have a Seam application that kicks off business processes. At the moment there is no other application or component that needs to co-ordinate, there is just the one web app.
I'd like to use asynchronous continuations in the business processes, but at the moment if I do the process just sits there inactive and the relevant actions are not executed.
How do "switch on" asynchronous continuations in a Seam environment?
I've seen this article, but cannot locate any of the classes mentioned and am a bit stuck!
I eventually stumbled into a work-around attached to JBSEAM-3837, but only after giving up Googling, attempting to write a Seam component like that one myself, then giving up on that and opting for a standalone version.
Talk about irritating!
Related
I prefer to keep my handlers free from ASP.NET infrastructure that is very hard to test (yes, even in ASP.NET Core). But sometimes it happens and you have a dependency like UserManager (I'd like to know one day why it's not an interface), HttpContext, etc. and unit-tests are turned into a mocking-hell.
I tried using integration testing to deal with it by creating a TestServer and having all the ASP.NET infrastructure initialized for every api call. It works quite well but sometimes seems like an overkill if I want to test simple logic of my handler. And while it solves technical problem of mocking ASP.NET infrastructure, it keeps architectural problem (if you consider it so) of having ASP.NET infrastructure into your handlers.
I'd like to know what are the recommended approaches to deal with it?
I feel your pain. I stumbled across a fantastic blog post from Jimmy Bogard that handles this problem by using what Martin Fowler calls Subcutaneous Tests. I will leave the deep explanation to those experts but in a nutshell subcutaneous tests simply avoid all the difficult to test aspects of the UI.
Shameless plug: I am currently in the process of writing up a wiki that demonstrates these patterns in a sample end-to-end project on github. It's not difficult to follow but is probably too much code to post for a SO answer.
To Summarize:
If you are using MediatR correctly your controllers should be very thin which makes testing them pointless.
What you want to test are your handlers.
However, you want to test your handlers as part of your real world pipeline.
To Solve:
Wrap the http request in a transaction.
Build a test fixture that mimics the applications Startup.cs
Setup a test db server to execute queries and commands against but also is reset after each test.
That's basically it. Everytime you run an integration test against one of your handlers:
The hosting environment is mocked but your application is started up in a real world test.
Your query or command is wrapped in a transaction mimicking your DbContext.
The handler is executed against a real database and then reset.
I would add more code examples to my answer but between the blog post and the wiki I provided, it is much easier to follow the code examples there.
Edit 8/2021:
Stick with the source. Jimmy Bogard keeps the contoso university project current on his github page. Another great and a little more advanced example is the modular monolith project by Kamil Grzybek. That also is updated regularly on his github page.
Mediatr or no, you should always try to have only very basic pass this along logic in your controllers and call injected business logic classes from there to do the actual work. As you inject them with interfaces to this business logic, your controllers' dependencies are easily mocked in your unit tests, and your tests can focus on if they implement those interfaces properly and do only the basic work of routing input/output. And your actual business logic can be tested even easier.
For those classes that are static, for instance for reading the web.config settings, one strategy that I like a lot is make an interfaced wrapper class around them. While ConfigurationManager is static, I can still just write a regular class with an interface that I put methods or properties on to read a specific setting (preferably semantically named) from the Configuration Manager. Now I can easily mock any configured setting (or absence of it) in my test by just mocking the interface and setting up different return values.
I'd say it depends on the level of confidence you want to get in the end. If you want to make sure the whole system works as expected, then integration tests using a TestServer are probably the way to go.
One advantage of MediatR, though, is it allows you to decouple your business logic from the application using it, which is why at the very top level, let's say in controllers, there's no logic but just a delegation to the mediator.
That being said, you're right that sometimes your logic needs information from the hosting application. An example would be the user making the request, which is accessible in the HTTP context.
In that case, if you want to avoid having to set up a test HTTP server to test your logic works, you could represent that information in an abstraction and your handler would then take a dependency on that abstraction. Your tests could then mock that dependency while using the real system for everything else.
Does that make sense?
I have to see if I can work around a known thread safety issue on third party component. The plan was to let an ASP.Net app to talk to the third party component via a WCF service, this was based on the assumption that I will be able to assign the WCF service to it's own application pool, restrict the pool to one working process and the working process to one thread. Requests to the service will have to wait for their turn, but that's OK because we expect them to require very little time and to be rare as well.
Problem is, I can't find anything that suggests how to achieve this part: "restrict the working process to one thread".
I have found a few useful pages, but no solution:
On IISforums, the discussion seem to suggest that I can achieve this in IIS 7+ but does not mention IIS6.
On MSDN Blogs and linked MSDN books chapter (can't put the link because I'm new here!) the discussion is mostly about what you can set via the machine.config file, which, if I'm understanding it correctly, is going to apply to all AppPools/Worker processes and is not what I'm trying to do, I would like to control what one single app can do, leaving the other applications untouched.
Questions:
Is it possible to achieve what I'm trying to do? (assign the WCF service to it's own application pool, restrict the pool to one working process and the working process to one thread" via IIS 6 configurations)
If not, can this be achieved programmatically somehow? For example, using locks or other threading-related tricks within my WCF service implementation.
It doesn't make much sense to me.. A web server have to be multithreaded by definition, because it must handle different income messages at the same time, if there is only one thread once in use any new request will fail.
What about to wrap the component in a class with a SynchonizationAttribute, so only one thread can access the component? Even if this will make your solution less scalable, at least it may work
http://msdn.microsoft.com/en-us/library/system.runtime.remoting.contexts.synchronizationattribute(v=vs.110).aspx
My team has been tasked with designing a web application that is workflow driven. I need some advice regarding the design.
The workflows need to be dynamic. Meaning, users can define the workflows through some interface and apply those workflows to a given scenario (The definitions will live in a SQL 2008 Database). The scenarios are defined by the business and will never change. So there may be only 2 types of scenarios a workflow can be defined for. The workflows are not necessarily linear. Some sort of state will drive the workflow. States will also be dynamic, but only exist in a workflow.
I have been looking at examples of workflows and state machines and my head is spinning. I am not sure I want o leverage Workflow Foundation or something we develop. I have seen this and think it may work, but I am not sure the state full implementation will work for us.
You can do this using WF4. I have never used Objectflow so I can't really comment on that but it appears to be an in memory solution and with an ASP.NET web site hosted in IIS that means you will occasionally lose state as IIS recycles and AppDomain. Usually not a big problem as it doesn't happen often but a WF4 InstanceStore will take care of that. It will also allow you to run on a web farm without sticky sessions and have the workflow migrate from machine to machine.
Another nice thing is the workflow designer. Its a WPF control you can rehost in your own app. Not in am ASP.NET or Silverlight app but you can provide a smart client to have users update the workflow definition using the sane designer as you use in VS2010.
The biggest problem with WF4 is the asynchronous execution nature. You will need to use a SynchronizationContext to execute the activities and wait for the workflow to go idle in the new state before you return the resulting HTML to the browser.
I am writing a web application in ASP.NET 3.5 that takes care of some basic data entry scenarios. There is also a component to the application that needs to continuously poll some data and perform actions based on business logic.
What is the best way to implement the "polling" component? It needs to run and check the data every couple of minutes or so.
I have seen a couple of different options in the past:
The web application starts a background thread that will always run while the web application does. (The implementation I saw started the thread in the Application_Start event.)
Create a windows service that is always running
What are the benefits to either of these options? Are there additional options?
I am leaning toward a windows service because it is separated and can run on a different server (more scalable) as well as there is more control over when it is started/stopped, etc. However, I feel like the compactness of having the "background" logic running in the process of the web application might make the entire solution more understandable.
I'd go for the separate Windows service primarily for the reasons you give:
You can run it on a different server if necessary.
You can start and stop it independently of the web site.
I'd also add that it could well have some impact on the performance of the web site itself - something you want to avoid.
The buzz-word here is "separation of concerns". The web site is concerned with presenting the data to the user, the service with checking the integrity of the data.
You can also update the web site and service independently of each other should you need to.
I was going to suggest that you look at a scheduled task and let Windows control when the process runs, but I re-read your question and noted that you wanted the checks to run every couple of minutes. The overhead of starting the process might be too great in this case - though some experimentation would probably prove this one way or the other.
If you use a scheduled task there's also the possibility that you could start the next check before the current one has finished - something you can code for if you're in complete control.
Why not just use a console app that has no ui? Can do all that the windows service can and is much easier to debug and maintain. I would not do a windows service unless you absolutely have to.
You might find that the SQL Server job scheduler sufficient for what you want.
Console application does not do well in this case. I wrote a TAPI application which has to stay in the background and intercept incoming calls. But it did it only once because the tapi manager got GCed and was never available for the second incoming call.
I've been using WWF for a while as part of an internal call center application (ASP.NET), and while learning it was a good practice in understanding how a state machine based workflow system should work, I am definitely not in love with WWF itself. In my opinion it is:
Overly complex, especially for use within web apps (all that threaded runtime stuff)
Immature (ever worked with that horrible designer?)
Anemic in its current feature set
Does anyone have a suggestion for a better .NET based workflow framework? Specifically, I am looking for the following features:
State machine based (mapping states to available actions)
A focus on user permissions (controlling who has access to what actions)
The ability to run workflows as timed background tasks (for example, to send out reminders for items that have been sitting in a certain state for x days)
That's really all I need. I don't need to be able to "drag and drop" any activities or visually design the flow. I am perfectly comfortable writing actual code once a particular action is triggered.
You could try Simple State Machine. You would have to implement access control and background timers yourself, but that shouldn't be a big deal. SSM was also built out of frustration with WF. There are some other state machine implementations on Codeplex as well. If one of them doesn't fit he bill out of the box, they are open source and should get you close enough.
I wholeheartedly agree with you about state machines in WF - they aren't testable, are too complicated, the threading model is peculiar and hard to follow, and I'm not sure a visual designer could have been more poorly conceived for designing state machines graphically. I think this may be because the state machine concept feels tacked onto the WF runtime, which was designed for sequential state machines, something WF does a much better job with, in my opinion. The problem is that state machines are really not the same animal as a sequential work flow, and should have been given a first class implementation of their own, because the warping of WF to make it seem to support them turned out to be more or less unsupportable, if not actually unusable.
I would stay away from Drools.Net since it's last SVN commit was in September 2007. Looks nice but it seems a bit too risky to make such a big library part of your project when you know it doesn't get any attention anymore.
Try Drools.NET
Have a look at Workflow Engine. It is a lightweight workflow framework for .NET and Java solutions. It has an HTML5 visual designer, version control, a decent UI and supports a wide range of databases.
Do you have the option to consider BizTalk Server?
I quite enjoyed working with Oracle BPEL Process Manager. It's part of JDeveloper.
http://www.oracle.com/technology/bpel/index.html
http://gemsres.com/story/dec06/313602/jellema-fig1.jpg
You may want to take a look at Jazz - http://jazz.codeplex.com/
Try WF4.5. It was completely redesigned since .NET4.0.
First of all you should look for a engine supporting BPMN. BPMN is a standard in Workflow and Process management and well supported from a lot of projects.
Second you should think about the requirements to thus an engine.
When you look for a BPMN Engine, there are two different approaches:
Task-Orientated
These engines (e.g. JBoss BPM - jbpm) are designed to process an input data by a well defined process model. Each task in the model gives the control to a piece of code - either a standard or an individual implementation. The process ends when the process-token reaches the end of the process model (End-Event). This kind of processing takes milliseconds. The engine can be used for batch jobs or processing data with a complex process orientated flow.
Event-Driven
Human-Centric workflow engines are event driven (e.g. Imixs-Workflow). This is a kind of state machine but offers typically much more functionality. You can start a new processinstance by assigning your business object with the initial task (defined by the start event). Than the workflow engine allows you to trigger events assigned to each task, defined in your model. Each event (Intermediate CatchEvent) triggers the workflow engine to transfer the running processinstance to the next task (state). Until no new event is triggered, the processinstance 'waits' in the current task (state). An approval process is an typical example for this kind of human-centric workflow.
You can find a list of engines here.