解决C#全屏幕截图的实现方法

今天一位同事想写一个全屏幕截图的代码。当然要实现的第一步是能够获取整个屏幕的位图，记得win32 api的createdc, bitblt等函数可以使用。于是上网查了下，果然屏幕截图用这些函数。但winform已经可以把api都忘记了，所以得寻找一个无win32 api的实现方式。综合了网上的实现，以及自己的一些设计，实现思路如下：
1. 开始截图时，创建一个与屏幕大小一样的位图，然后用graphics.copyfromscreen()把屏幕位图拷贝到该位图上。这是很关键的一步，这样所有的操作就都可以在该位图上进行了，而无实际屏幕无关了。 

code
int width = screen.primaryscreen.bounds.width;
int height = screen.primaryscreen.bounds.height;
bitmap bmp = new bitmap(width, height);
using (graphics g = graphics.fromimage(bmp)) {
    g.copyfromscreen(0, 0, 0, 0, new size(width, height));
}

2. 接下来为了方便在这之上进行截图，有一个很重要的设计实现方式：用全屏幕窗体代替现有真实屏幕，这样就可以把截图过程的所有操作都在那个窗体上实现（该窗体设置成无边框，高宽等于屏幕大小即可），另外为了显示掩蔽效果（只能正常显示选择的部分屏幕内容，而其实部分用一个如半透明层覆盖），就添加一层半透明位置位图。具体代码如下：

code
public partial class fullscreenform : form {
    private rectangle rectselected = rectangle.empty;
    private bool isclipping = false;
    private bitmap screen;
    private bitmap coverlayer = null;
    private color covercolor;
    private brush rectbrush = null;
    private bitmap resultbmp = null;
    public fullscreenform(bitmap screen) {
        initializecomponent();
        int width = screen.primaryscreen.bounds.width;
        int height = screen.primaryscreen.bounds.height;
        coverlayer = new bitmap(width, height);
        covercolor = color.fromargb(50, 200, 0, 0);
        rectbrush = new solidbrush(covercolor);
        using (graphics g = graphics.fromimage(coverlayer)) {
            g.clear(covercolor);
        }
        this.bounds = new rectangle(0, 0, width, height);
        this.screen = screen;
        this.doublebuffered = true;
    }
    protected override void onmousedown(mouseeventargs e) {
        if (e.button == mousebuttons.left) {
            isclipping = true;
            rectselected.location = e.location;
        }
        else if (e.button == mousebuttons.right) {
            this.dialogresult = dialogresult.ok;
        }
    }
    protected override void onmousemove(mouseeventargs e) {
        if (e.button == mousebuttons.left && isclipping) {
            rectselected.width = e.x - rectselected.x;
            rectselected.height = e.y - rectselected.y;

            this.invalidate();
        }
    }
    protected override void onmouseup(mouseeventargs e) {
        if (e.button == mousebuttons.left && isclipping) {
            rectselected.width = e.x - rectselected.x;
            rectselected.height = e.y - rectselected.y;
            this.invalidate();
            resultbmp = new bitmap(rectselected.width, rectselected.height);
            using (graphics g = graphics.fromimage(resultbmp)) {
                g.drawimage(screen,new rectangle(0, 0, rectselected.width, rectselected.height), rectselected, graphicsunit.pixel);
            }
            this.dialogresult = dialogresult.ok;
        }
    }
    protected override void onpaint(painteventargs e) {
        graphics g = e.graphics;
        g.drawimage(screen, 0, 0);
        g.drawimage(coverlayer, 0, 0);
        paintrectangle();
    }
    protected override void onpaintbackground(painteventargs e) {

    }
    protected override void onkeydown(keyeventargs e) {
        if (e.keycode == keys.escape) {
            this.dialogresult = dialogresult.cancel;
        }
    }
    private void paintrectangle() {
        using (graphics g = graphics.fromimage(coverlayer)) {
            g.clear(covercolor);
            graphicspath path = new graphicspath();
            path.addrectangle(this.bounds);
            path.addrectangle(rectselected);
            g.fillpath(rectbrush, path);
            g.drawrectangle(pens.blue, rectselected);
        }
    }
    public bitmap resultbitmap {
        get { return resultbmp; }
    }
}

上面的代码都很容易看明白，这里有一个技巧就是graphicspath，它自动会形成一个中空的区域。上面的实现很容易扩展：多区域截图，多裁判截图等都很容易实现。