Movies Screensaver

So, in short, I’ve created a screensaver that randomly chooses a movie and plays that instead of the default Windows screensavers.

I’ve got a few movies converted to DIVX (.avi) format and have them all saved in a particular folder on my server at home and thought it would be much more fun to have a movie-based screensaver and so set about trying to make one.

I stumbled into MSDN’s Coding4Fun website and found a WPF project on there that someone has done well in writing, which displays various images across the screen. Great. 

Then, it all got much better unexpectedly, as when I ran their screensaver code on my PC, I found that they’re also catering for the multi-monitor-folk, meaning that the images are different on each monitor. Fantastic! (Hopefully, you can see where I’m going with this…)

A few deletions here and there in their code and added a call direct to a movie to see if it would work. And it did. Dirty Harry played well on my two monitors, but I wanted to see how far I could go.

I then made a quick-and-dirty .Net app to skim through the folders on my server and recorded each movie within a quick SQL Server Express database.

A little Linq-to-SQL here and there and within minutes, I had the WPF screensaver application running, randomly choosing a movie for each monitor available (in my case two) and displaying them on separate screens.  And it did this when the system was idle for a few minutes.

It was originally 10 minutes, but patience was out of the window… I needed to play!

The next problem was something I didn’t expect at first, but soon found the remedy; when the movies started playing, they mostly displayed the title credits and some took ages before seeing any real action.  So, in the code, I made each movie jump exactly 37 minutes in, as after half-an-hour, you’d expect to see something more interesting, right? 

The next problem, was that there are then two movies (or however many screens you have) playing simultaneously and so you get two sets of sound effects coming through the speakers! (Dur!) This would manifest itself as having the shooting scenes from one movie overlaid over another, which wasn’t too bad until Die Hard and Atonement would come up at the same time…  So, I’ve taken the fairly drastic step at the moment of just muting the lot.

So, the result is a screensaver that randomly chooses the movies each time, meaning that coming back to your PC takes slightly longer to do, as each time I seem mesmerised and simply have to watch a few minutes of each before returning to whatever I was doing.

I haven’t done any more to the screensaver as yet, but hope to, as the problem is that I haven’t seen some of the movies that it throws up for me, so I end up watching 5-10 minutes of the movie, part-way through and in mute.  I was thinking that I could put a label in the corner of the screen, showing the name of the movie to help give me a clue!

And, maybe I could trap the event firing for a certain key that could toggle the sound for a given monitor…

Also, ideally, the movie will show for 5 minutes and then the power management for the PC should kick-in and power-down the monitors?

Thoughts and opinions are welcome.

ASP .Net Win-Forms or MVC?

For those that haven’t yet tried out the world of MVC, this post was intended to give you my take on it, comparing and contrasting where possible, and to the best of my knowledge. (Don’t shoot me if I’m wrong, I’m still learning about it too!)

If you look at the documentation from Microsoft, ASP .Net MVC is built upon the ASP .Net stack, meaning that it isn’t meant to reside alongside the Win-Forms route, but in addition to it, so don’t think that this *has* to replace the Win-Forms approach, as there’s probably a reason to use one over the other.

Some key points:

• ASP .Net MVC is still a platform to make web-based applications
• HTML pages are still produced, that use CSS for styling
• You can still work in either VB .Net or C#
• It still compiles in the same manner
• and you can achieve the same results (almost), but the way that it happens is different.

Having said that, there are a few differences between them, and it’s these differences, for me, that make me want to learn to use MVC a little more-so than.

• MVC doesn’t use ViewState and therefore is stateless
• MVC encourages cleaner URLs
• MVC encourages better separation of concerns
• MVC attempts cleaner output HTML markup
• MVC encourages the ‘Convention Over Configuration’ theory
• MVC discourages the code-behind element

In terms of the lack of ViewState, this isn’t a bad thing; the output HTML markup is much cleaner as the ViewState doesn’t get added to the page.

The cleaner URLs are almost like a bi-product of the separation of the code out between the Models and the Controllers; what I mean here, is that by grouping pages (now called Views) into subfolders and having a dedicated controller for each subfolder, the URLs within the address bar of the browser look more like:

http://whatever.com/Accounts/Create

which means that search engines, such as Bing, Google, etc can spider and categorise the pages better, if so desired.

Also, notice that the Create page doesn’t have an extension on it by default. This is due to the routing engine of MVC, as page-requests from the browser no longer go direct to the .aspx pages as before in ASP .Net. Instead, page-requests are sent to the associated controller, which then produces the requested page as a result and returns it back to the browser.  So, in actual fact, in the example link above, the ‘Create’ method, or ActionResult as they’re now titled, is called within the AccountsController object.  Being a regular piece of C#/VB.Net code, this method can call-out to the data layer and bring back the necessary information to produce that page alone, resulting in a return statement, producing the page.

Why is this better? This approach helps to make the Controller handle any sort of business logic, and help keep the View page dumb, displaying only what it’s passed to, by it’s controller. 

The way that the data is passed to the page is significant also, as objects from the Model can be used, meaning that you can create and populate fully-fledged Person object and pass it to the View like this:

return View(myPerson);

and within the markup of that page, you can access any property of the Person object like this:

<div>
     <p>Your name is <%=Html.Encode(Model.Firstname) %></p>
</div>

And yes, I know what you’re going to say about the <% %> syntax, and it’s similarity to the good ol’ Classic ASP days, but you have to see past this and reap the benefits that it brings.  It does mean that the HTML markup has lots of <% %> markers in it, but this will only get as messy as you make it.  I forget who said about the fact that “developers make spaghetti code”, meaning that your code is your fault when it looks a state. 

Busy pages can be refactored out into Partial Views, allowing their use by multiple other pages, as well as promoting adherence to the DRY principle (“Don’t Repeat Yourself”).

In terms of further separating out the business and/or data layers to their own projects, this is still acceptable, but it’s worth pointing out that the vast majority of MVC examples that I’ve seen so far, throw all layers of a site into the same Visual Studio project, probably for ease of display.

Getting started with MVC is easy once you have downloaded and installed the small Visual Studio plug-in, as it’s literally:

File > New > ASP .Net MVC Project

By default, the new MVC project has a few pages as well as a fully operational log-in/registration area for a members area of the site.  So, right after creating the project, you can run it and click between the dummy pages that exist to demo it’s early stage. 

This gives you, as a developer, a starting point to learn from as in order for this [albeit basic] functionality to work, it must have the necessary Controllers, Views, etc in place.  Also, when creating a new project, having a visual starting point is also good, as the default CSS file ensures that the environment is already there to start work from, but can be totally changed later.

I mentioned about the Convention over Configuration that MVC has taken on, and to start with, I was fully aware of how wide this reaches within the project. What Convention over Configuration means, is that, for example, the controllers all have to end with ‘Controller’, so you’d have AccountsController, HomeController, SomethingElseController, etc.  Then, when the Views are generated by Visual Studio for you, they use whatever is before the …Controller part as the folder names, such as Accounts, Home, SomethingElse, etc.  Because the whole process assumes that everything works to the same convention, things ‘just work’ on that basis.  Like almost everything else in MVC, you can override this behaviour, but once you’ve made a cock=up or two doing it, you probably won’t forget it, and even come to liking this way of working, rather than having to specify details of things in .config files.

Personally, I’m a fan of this whole MVC thing.  I’ve been playing around with MVC 1.0 which came out a while ago, but also been using MVC 2.0 which is not released until 2010.  Some of the improvements to 2.0, such as the Data Annotation Validators, are great and

I learn new things like this best when I’m actually working on something constructive, rather than just playing around with dummy (pointless) apps, so I’ve been making a CRM application in MVC (1.0) and, bearing in mind I knew nothing about MVC before I started, it’s not too bad.  One thing I will admit, looking back at the work I’ve done earlier on in the project, I see big areas that I now want to refactor after learning the better way of doing things in ‘MVC-land’.

That’s not to say that I’ve done it particularly wrong or badly, but there is a cleaner path to the same outcome that can be achieved, if I want to go back an re-work it.  I think that at the time, there was quite a bit of difference to what I was used to, and probably more that I was a little too eager to get something down in code.  Either way, I’m happy that what I’ve done so far is ‘ok’, but I know the next project that I’m already working on is already better because of the first.

But as to what technology to use for your next project, that’s entirely up to you, but I know what I’d like to use.

What’s this all about?

Well, good point.

Basically, I thought I needed another blogging outlet in my life and hence Brett’s Blog has started.  This is to be my ‘geeky chuntering’ site, somewhere that I can talk ‘geek’ without putting anyone off.

I started blogging almost a year ago, creating Brett’s Baby Blog, which was to be my musings on being a dad for the first time.  It’s still going, I still enjoy writing posts for it although admittedly there’s been a few gaps in the blog-post-timeline, but nonetheless, it’s still there. 

The Baby Blog site uses DasBlog, which is an ASP .Net 2.0 application that I downloaded from the CodePlex.com site.  DasBlog is great, as out of the box there’s a range of templates available to choose from, as well as customise accordingly. Back then, I didn’t have a great deal of time, as I think when I started it, my wife was about due to pop, so I wanted to get it started whilst I had the opportunity and haven’t really seen a need to revamp it!

The problem with the Baby Blog site, was the title instantly restricted the content that I could put up on there, which helped me to focus the content specifically on baby-stuff, but when something geeky came up that was directly related to my baby, I thought twice about writing a post about it, for fear of some people tuning out and not reading any of it. For example, one post that did make it through, was about backing up photos; I have been mentioning how many photos that we’ve taken of the little lad so far, and how I really didn’t want to lose them, so I blogged about ensuring that photos were backed up, and how I did it, etc. The problem was, a few people were a bit lost by that post, as it wasn’t anything to do with Jack, nor babies as a whole, but I felt it was important to write something about backing up data, especially photos, as I have been the victim of a failed hard drive once before and lost all of our holiday photos up to that point.

I have also created a spin-off mini-site for my son, Jack, which is piggy-backed on the ASP website I created for our wedding photos about 5 years ago almost.  The whole site was designed to be private, restricting it to friends and family, etc. This meant that I can put up more personal details about what’s going on, access to the photo and video library of him, etc. Obviously stuff that I’d like to share but have a little control as to who sees it really.

So, back to the point about this blog space. Well, I’ve fancied a geeky blog a little while ago, but didn’t know if I could upkeep it.  But since we’ve created an internal blog at work, specifically for in-house geeky topics, I think it’s fair to say that I’ve given that a good go; there’s lots more stuff I’d like to add but can’t as it’s not really work-related.  So, that’s where this blog steps in.

I’m not sure who I expect to follow this blog, as I don’t know as yet how this will all pan-out, but in a way, it’s not about who reads it I suppose. 

Just to be clear, I’m not saying I’m hoping to contribute anything exciting to the world with this blog, but I’ve been involved with computers for quite a while now, and would like to think that I can offer something of interest at some point in the future, but it depends on how much fluff you have to sift through to get it!