Tuesday 23 November 2010

How to Register Two Interfaces as a Singleton in Unity

Every now and then I find myself wanting to configure a single implementor for multiple interfaces in Unity. Say I have the following interfaces and I only want one instance of Foo, yet I want to be able to request either IFoo or IFoo2 from the container:

public interface IFoo
{
    ...
}

interface IFoo2 : IFoo
{
    ...
}

public class Foo : IFoo2
{
    ...
}

The way I would previously go about this is:

var foo = container.Resolve<Foo>();
container.RegisterInstance<IFoo>(foo);
container.RegisterInstance<IFoo2>(foo);

This approach, works, but is less than ideal. First, it requites you to resolve something in the container, possibly before you have completely finished configuring the container (e.g. the dependencies of Foo might not be fully set up yet). Second, the container will dispose foo twice.

However, thanks to an answer from Sven Künzler to a question on Stack Overflow, there is a much better way:

container.RegisterType<Foo>(new ContainerControlledLifetimeManager());
container.RegisterType<IFoo, Foo>();
container.RegisterType<IFoo2, Foo>();
Assert.AreSame(container.Resolve<IFoo>(), container.Resolve<IFoo2>());

You simply register the concrete type as a singleton, and then point as many interfaces at that type as you like.

Wednesday 10 November 2010

How to Invoke a Command on the ViewModel by Pressing the Enter Key in a TextBox with Silverlight and MVVM

I recently attempted to upgrade a WPF application to compile for Silverlight as well. One of the many issues I ran into was that in the WPF version, pressing the Enter key while I was in a TextBox caused the OK button I had on the form to be clicked by virtue of the fact that I could set the IsDefault property on the button. However, after porting to Silverlight, that no longer worked, since the IsDefault property is missing..

A web-search revealed that someone had made a “behavior” that allows a specified button to be clicked when you press Enter within a TextBox (available for download here). However, it had one big problem: at the point that the command fired in my ViewModel, the value bound to the textbox contents had not been updated, since the textbox had not lost focus.

The original WPF binding I had an UpdateSourceTrigger ensuring that the ViewModel was always kept up to date with what was in the TextBox:

<TextBox Text="{Binding Answer, Mode=TwoWay, UpdateSourceTrigger=PropertyChanged}" />

but in Silverlight, the PropertyChanged UpdateSourceTrigger is not available, so  I was left with the following:

<TextBox Text="{Binding Answer, Mode=TwoWay}" />

I decided to make my own EnterKeyCommand binding that would allow you to specify for a TextBox which command on the ViewModel should be run. Here’s the code:

public static class EnterKeyHelpers
{
    public static ICommand GetEnterKeyCommand(DependencyObject target)
    {
        return (ICommand)target.GetValue(EnterKeyCommandProperty);
    }

    public static void SetEnterKeyCommand(DependencyObject target, ICommand value)
    {
        target.SetValue(EnterKeyCommandProperty, value);
    }

    public static readonly DependencyProperty EnterKeyCommandProperty =
        DependencyProperty.RegisterAttached(
            "EnterKeyCommand",
            typeof(ICommand),
            typeof(EnterKeyHelpers),
            new PropertyMetadata(null, OnEnterKeyCommandChanged));

    static void OnEnterKeyCommandChanged(DependencyObject target, DependencyPropertyChangedEventArgs e)
    {
        ICommand command = (ICommand)e.NewValue;
        FrameworkElement fe = (FrameworkElement)target;
        Control control = (Control)target;
        control.KeyDown += (s, args) =>
        {
            if (args.Key == Key.Enter)
            {
                // make sure the textbox binding updates its source first
                BindingExpression b = control.GetBindingExpression(TextBox.TextProperty);
                if (b != null)
                {
                    b.UpdateSource();
                }
                command.Execute(null);
            }
        };
    }
}

Most of it is pretty simple, and it will allow an Enter key command to be specified for any control, not just textboxes. However, if you have bound to a textbox, it will call UpdateSource on any Text binding you have made, to ensure your ViewModel operates on the latest data.

Here’s how you use it in XAML:

<TextBox 
    Text="{Binding Answer, Mode=TwoWay}" 
    my:EnterKeyHelpers.EnterKeyCommand="{Binding SubmitAnswerCommand}"/>

It also has the advantage of being considerably more succinct than the equivalent XAML for using the behavior I linked to earlier.

Monday 8 November 2010

Merging MP3 Files with NAudio in C# and IronPython

If you would like to concatenate MP3 files using NAudio, it is quite simple to do. I recommend getting the very latest source code and building your own copy of NAudio, as this will work best with some of the changes that are in preparation for NAudio 1.4.

Here’s the C# code for a function that takes MP3 filenames, and writes a combined MP3 to the output stream:

public static void Combine(string[] inputFiles, Stream output)
{
    foreach (string file in inputFiles)
    {
        Mp3FileReader reader = new Mp3FileReader(file);
        if ((output.Position == 0) && (reader.Id3v2Tag != null))
        {
            output.Write(reader.Id3v2Tag.RawData, 0, reader.Id3v2Tag.RawData.Length);
        }
        Mp3Frame frame;
        while ((frame = reader.ReadNextFrame()) != null)
        {
            output.Write(frame.RawData, 0, frame.RawData.Length);
        }
    }
}

And here’s an IronPython script (just put NAudio.dll in the same folder as the mp3merge.py script):

import clr
clr.AddReference('NAudio.dll')

import sys
from NAudio.Wave import Mp3FileReader
from System.IO import File

def GetAllFrames(reader):
    while True:
        frame = reader.ReadNextFrame()
        if frame:
            yield frame
        else:
            return

def Merge(files, outputStream):
    for file in files:
        with Mp3FileReader(file) as reader:
            if reader.XingHeader:
                print 'discarding a Xing header'
            if not outputStream.Position and reader.Id3v2Tag:
                outputStream.Write(reader.Id3v2Tag.RawData, 0, reader.Id3v2Tag.RawData.Length)                
            for frame in GetAllFrames(reader):
                outputStream.Write(frame.RawData, 0, frame.RawData.Length);
            
if __name__ == '__main__':
    if len(sys.argv) < 3:
        print "Usage: ipy mp3merge.py output.mp3 File1.mp3 File2.mp3"
    else:
        with File.OpenWrite(sys.argv[1]) as outStream:
            Merge(sys.argv[2:],outStream)

Notes:

I simply copy across the ID3v2 tag from the first MP3 file if present. All other ID3v2 tags are discarded (as are ID3v1 tags). Also, I discard the Xing frame from VBR files. It could easily be re-included if desired, although it’s information will not necessarily be valid about the combined MP3 file. One final thing, I wouldn’t recommend merging MP3 files of different sample rates, or mixing mono with stereo, as it could cause various players issues.

Sunday 7 November 2010

State of MP3 Playback Support in NAudio

The MP3 playback support in NAudio was always rather experimental. The ACM conversion code I had written assumed CBR (constant bit rate) and constant block sizes. With MP3s this is not always the case, since there are VBR files with variable block sizes, and even in CBR MP3 files, padding means you can get frames of different sizes. However, despite these issues I did manage to get MP3 more or less playing back, which was cool, but not 100% reliable. People ran into issues like the occasional error while repositioning a stream, or the more irritating fact that Mp3FileReader was not good at calculating the duration of a file.

In this post I will go over some of the key challenges to getting good MP3 playback support, with details of some recent changes I have checked in, along with some ideas for the future.

1. Correctly parsing MP3 frame headers

To work effectively with MP3 files you need to be able to parse frame headers correctly and determine their exact size. If this cannot be done, we have no choice but to pass blocks of MP3 file directly to the decoder without knowing whether we are giving it whole frames or not.

Finding out how to properly parse MP3 frame headers was a much harder challenge than it seemed. Googling for info on MP3 frames reveals some articles that look authoritative but simply failed to correctly parse everything I threw at them. Mono or low sample rate MP3s got their frame size calculation wrong. However, eventually I found this article, whose source code had the final missing piece of information that allowed me to get it reliable.

2. A single frame decoder

Once we can calculate frame sizes reliably, we are able to decode them one by one. The WaveFormatConversionStream assumes everywhere it is working with CBR, so I have removed it from the equation. Now a simple MP3 Frame Decoder class (final name and interface to be decided) is used to decode MP3 frames one at a time using ACM. Alternative frame decoders could easily be plugged in if required in the future (e.g. using DMO or NLayer).

The really big change is that this means that the MP3FileReader no longer returns MP3 data in its Read method, but emits PCM. This makes life so much easier downstream and simplifies the playback graph considerably (no more BlockAlignReductionStream). I’ve made the ReadFrame method public so if you have a pressing need to get the compressed data out instead there is nothing stopping you.

3. Accurate length reporting

Accurate length reporting was never possible before, since it relied on an estimate of the bitrate. But now we can parse MP3 frames, we have accurate knowledge of exactly how many samples each frame will decompress into, and the TotalTime property (and CurrentTime property) of MP3FileReader should be entirely accurate. n.b. I think that it may be possible that the first frame in a VBR MP3 file decompresses to zero samples (although I already exclude the Xing frame so maybe there is another similar meta-data type frame), so we might actually very slightly over-report the length – I’ll need to look into that.

4. Repositioning to frame granularity

When you reposition with the MP3FileReader.Position property, there is of course every possibility that you will ask for it to reposition to a place midway through a frame. We now automatically move you to the start of the frame that contains the position you asked for.

An earlier NAudio contributor had done some cool stuff with a BinarySearch to speed this up. I needed to drop this temporarily as I was making changes to the table of contents generation. However, there is no reason why this could not be reinstated now, to speed up performance further (although repositioning perf doesn’t seem to be a major issue with the tests I have done on 1 hour long MP3s).

5. Repositioning with sample granularity

Obviously, it would be even nicer to support MP3 repositioning with sample granularity. This would involve us decompressing a frame during the seek process, so when the next Read occurs we can read from part-way through that frame. The framework to do that is already in place (we keep track of “leftovers”, so this could be a feature I add in the not too distant future).

6. Forward only

One thing I haven’t got round to doing yet, is making it possible to use MP3FileReader from an input stream that doesn’t support repositioning. Obviously, it would not be able to work out its Length, and there would be no real need for a TOC. Proper support for forward only streams would be useful to people wanting to do network streamed playback, which seems to be one of the most common queries I get.

7. Changes of Sample Rate and Number of Channels

It is theoretically possible within an MP3 file for the sample rate and number of channels to change from frame to frame. However, I have no immediate plans to support this since I’ve never seen an MP3 file that does this. The only scenario in which I could imaging this would be if someone was attempting to concatenate two MP3 files by simply copying frames from one into the other.

8. ID3 Tag Support

I have no plans to introduce ID3 tag reading or writing, since there are other open source libraries out there that do this perfectly well.

Give me feedback

Please grab the latest NAudio code from Codeplex and let me know how you get on with it. As always, the best way to give it a run-through is to use the NAudioDemo app that is included in the solution. Load it up, select WAV playback and try it with whatever MP3s you have lying around on your hard disk. It would be great to have robust MP3 playback as a headline feature for NAudio 1.4.

Tuesday 2 November 2010

Why you should use DVCS for Personal Projects

For a long time it bugged me that there wasn’t an easy way for me to use version control for my personal projects. I have over 100 small applications sitting around on my computer. Most of them are just test apps that will probably never be visited again. But others are more useful utilities, or perhaps future open source projects currently in incubation. Some of them are work related, others purely for personal enjoyment or learning. Often I found myself wishing that I could have the benefits of source control, so I could back out of changes that broke something.

Pre Distributed Version Control Systems


In the days before DVCS (or, to be more accurate, before I knew about DVCS), at different times, I tried all the following approaches:
  
Store projects on my company’s VCS. This usually involved asking permission to have some space on SourceSafe or TFS for my personal projects. Whilst this has the benefit of meaning that I can share my work with others easily (and it is guaranteed to be backed up), there is the hassle of getting this set up in the first place, plus the fact that some of these projects are very shortlived, while others are “skunkworks” ideas which you don’t want to give publicity to until they are ready for it.

Run a private VCS server on my dev machine. It is possible to install Subversion, SourceSafe, TFS etc servers on your local machine, and use them for VCS. However, as well as using up valuable resources on your personal machine, it has a very poor migration story. If you rebuild your PC, need to quickly copy code onto a USB stick and work on it from home, this option ends up being more hassle than it is worth.

Subversion file-based repository. A few years ago I discovered that you could get Subversion to back up to a file:// path. I thought this would be the answer to all my issues. The reality is, that it was quite fragile, especially since I was storing code on USB sticks, so the drive letter might change. I ended up corrupting my repositories so regularly I gave up on this option pretty quickly.

Make it open source. When CodePlex showed up, I immediately moved several of my projects there. This meant I had access to a free central repository, enabling me to work from different computers if I needed to. The downside is that most of my projects weren’t appropriate for making into open source applications.

Don’t bother with version control and make backup zip files. This ended up being my most common approach. Every now and then I would backup to a zip file. Of course, those backups are few and far between and don’t even exist for most projects. And I almost never had a backup available on those few occasions when I genuinely needed one.

Advantages of DVCS


But all that has changed after I decided to find out what all the fuss was about with Distributed Version Control systems. Whilst the idea seemed a little crazy to me at first (everyone gets a copy of the entire repostory?), the obvious advantages for personal projects won me over pretty quickly. I decided to try out Mercurial, since it seemed to have slightly better support on Windows, although I’m sure Git is just as good. Here’s some of the top advantages to using it on your personal projects:

No Server Required. This is a huge benefit. I don’t need a central server on the internet, or on my company network. If I want to move to another PC, I can just copy the code folder over (or Sync folders using something like DropBox) and it just works.

Version Control Everything. It’s now a no-brainer to put a new test project under version control. It takes only a few seconds to do. If for some reason you decide you don’t want version control anymore, just delete the .hg folder and its gone.

Migrate to a Central Repository Later. When I added NAudio to Codeplex, it already had been in development for several years. However, I have no checkin history up to that point, just a bunch of backup zip files. With Mercurial, you can move to using a centralised repository at any point in the future (whether public or private) and all your checkin history comes along for the ride.

Unconstrained branching strategy. Admittedly, for small personal projects, branching is not often that important. But it can come in handy when are half way through implementing one feature, and then want to work on a different task. Without version control, you have to decide whether to bin the half-finished changes, or to copy them somewhere else and manually merge them back in later. With Mercurial, it becomes trivial to create as many branches as you need. And you can merge directly between any two branches, irrespective of how many intermediate branches were created between them.

Merging divergent copies. Sometimes I have a copy of a personal project at home and at work and have no idea which one is the latest and greatest. Or maybe after backing it up to a few places, I have inadvertently made changes to two separate copies. One membership application I wrote for a youth group 8 years ago turns out to be still in use and they asked me for new features recently. I had to work out which of several copies was the one I should be using. With Mercurial, it is trivial to ensure that one copy is not missing any changes in another.

Little and often checkins. One really nice feature for my open source projects is that I can check in little changes without needing to immediately push them to the central server. This means I can check in little and often, and only do a push to the server once I have tested and made sure my feature is robust.

What was I doing?. Another advantage is that by using a DVCS with my personal projects, if I need to come back to one after a couple of years I can quickly examine the log, looking at diffs to see what I was up to last time I worked on it. This can be handy if I had left it in a state where there were some half-finished new features in progress, and actually I want to discard them and resume from an earlier point.

Trivial rollback – one of the things that scared me about DVCS was the idea that once I had checked in a file, it lives on in the repository forever. So accidentally checkin several megabytes of compiled binaries and you have an unnecessarily bloated repository. But the reality is that issue only exists if you push that to a central repository (and even then there are usually ways of working round the issue). If you haven’t you can just clone to the prior revision and your mistake is gone. I’ll perhaps do a post later on my thoughts about DVCS in the enterprise, where matters like this are quite important.

If you take anything away from this post, it is learn a DVCS and use it wherever you can. You will be glad you did.