What was Adobe thinking? Not even Java requires jumping through this many hoops!
A pity, because ActionScript 2.0 was actually quite a nice language.

I experimented with it a bit for the new site I was developing for my music. I waded through all of the tedious conversions, getting rid of things like SetInterval in favor of the Timer class, loading movies using the Loader class instead of a direct method, finding lots of things taken out, lots of names changed, but basically the same capabilities packaged in a more structured yet klutzier form.

Then I tried to unload a movieclip.

You can’t do it. Flash WILL NOT LET GO of the object, even after you call one of the removeChild methods. And since the object in question is an ActionScript 2 music player and (surprise surprise) you can’t interoperate with ActionScript 2 movies, I’m stuck in a situation where I can’t change the file that is playing and I can’t remove and reload the component.

Alone, this is a bad issue, but after hours of struggling with the language’s weird idiosyncrasies, I’m ready to go back to AS2 and forget about learning AS3 until some of the issues are fixed.

Update: I finally solved this problem using LocalConnection. Yes, I needed to use what is essentially a light form of IPC to solve this problem.

Apparently Microsoft has no easily accessible place to report bugs in their programming languages, so I’m posting it here and hopefully someone who can fix it will eventually read it (and maybe even put an easy-to-find link up to report bugs on MSDN?)

If it’s not a bug with .NET itself, perhaps someone can tell me what I’m doing wrote here, C# newbie that I still am.

I have a DataGridView bound to an ArrayList of objects of the following class:

public class SeriesPoint : IComparable {
    private double _price;
    public DateTime? Timestamp;

    public double Price {
        get { return _price; }
        set { _price = value; }
    }

    public string Date {
        get { return Timestamp.ToString(); }
        set { Timestamp = DateTime.Parse(value); }
    }

    public SeriesPoint() { _price = 0; Timestamp = null; }
    public SeriesPoint(double p, DateTime? d) { _price = p; Timestamp = d; }

    //These are compared only based on date.
    public static bool operator<(SeriesPoint lhs, SeriesPoint rhs) { return lhs.Timestamp < rhs.Timestamp; }
    public static bool operator>(SeriesPoint lhs, SeriesPoint rhs) { return lhs.Timestamp > rhs.Timestamp; }

    //Equivalent in C++: return rhs.timestamp == timestamp;
    public override bool Equals(object obj) {
        if (obj == null || !(obj is SeriesPoint)) return false;
        if (obj == (Object) this) return true;

        return ((SeriesPoint) obj).Timestamp.Equals(Timestamp);
    }

    //Nothing special about how it's hashed.
    public override int GetHashCode() {
        return base.GetHashCode();
    }

    public int CompareTo(object other) {
        //Nulls to the beginning.
        if (this == null && other != null) return -1;
        if (this != null && other == null) return 1;

        if (this < (SeriesPoint) other) return -1;
        if (this > (SeriesPoint) other) return 1;
        return 0;
    }
}

And I run the following to attempt to find the last null value in a sorted list:

int lastnull = sortedpoints.LastIndexOf(new SeriesPoint(0.00, null));

This actually works, but when the function exits and control presumably returns to the main loop, the program suddenly crashes with the following exception in Application.Run() (not even running on the same thread as my code!):

“An unhandled exception of type ‘System.InvalidOperationException’ occurred in System.Windows.Forms.dll

Additional information: Operation is not valid due to the current state of the object.”

I checked all of my code, put try blocks around things, made sure nothing was throwing when it shouldn’t be, removed functions, rewrote functions, reordered functions, checked boundary conditions… nothing. My code is running fine and returning correct results. .NET is then crashing. I have no idea why, but the data binding is definitely a possibility.

I suppose my best course of action is to remove the Equals() method (this fixes the problem) and rewrite my code without it.

Am I the only one who finds newer programming languages, such as Java and C#, to take strong typing to a very undesireable extreme? It seems like a reaction to most programmers not being able to get type safety right in C++, but it ends up making the code look awful, obscuring its function, and requiring a lot of extra work to explicitly cast things around. Especially when working with arrays, it oftentimes requires creating entirely new loops dedicated solely to the purpose of casting. This is stupid in many different ways. ConvertAll sort of helps, but it’s clunky and doesn’t work for collections. Also, why does the cast operator have lower precedence than the dot operator? I would think the other usage would be more common.

In Perl, which I now consider my favorite language (because it doesn’t presume to impose seemingly arbitrary restrictions on the programmer), you can join an array of doubles into a string in one line of code. You can do it in two in C++, which isn’t bad, since the stream operators have overloads for numeric types. Try it in C# or Java and you’ll need at least four or five lines.

I think I prefer the flexibility of weak typing despite the added risk. It’s a pity that “hot” languages are all following the “programming with mittens” paradigm right now.

Most people who use the phrase “Best Practices” probably don’t know what they are. I haven’t really found anyone using these terms who can explain why these practices are truly “best”, and in the realm of software engineering (not so much for web dev.), the code of people who endorse them tends to be awful. They can sometimes spit out why one model is better than another particular model, because they’ve memorized the dogma, but present them with a completely new paradigm and ask them to compare it and they cannot. It seems like they’re substituting process for talent.