You can change the fontsize per tag with CSS, maybe someone needs to give that one a try. Sounds like it could work. We'll then have a perfectly good way to get those tag wrapped images to display in textfields.

<img id="dragImage" ….
<img id="mainImage" ….

then use getImageReference("dragImage") and getImageReference("mainImage") etc.

It looks like a lot of these problems stem from the fact that TextField is, as you might suspect from the name, just for displaying text. Adobe have obviously bodged some sort of vague html support into it but it remains a bodge. For instance: tags create a Loader instance, attach it to the TextField and pass it a URL – but as TextField doesn't inherit from DisplayObjectContainer it shouldn't be able to do that. You also don't get any of the DisplayObjectContainer methods, so while you can reference the Loader object that's a child of textField, you can't remove it with textField.removeChild or loader.parent.removeChild. TextFields can't do that. I'll use AIR next time.

I then attach event listeners to them like this:

i = 1;

while (displayBox.getImageReference(i.toString()) != null) {
displayBox.getImageReference(i.toString()).addEventListener(MouseEvent.MOUSE_DOWN, displayImage);

i++;
}

That works fine for me and allows me to bring up full size images. My problem is that as soon as I alter htmlText even a TINY bit (eg htmlText += ".") the formatting goes all over the shop. Stumped.

var reImages:RegExp = //sigm;
var oIMGMatches:Object = reImages.exec(sContent);

What you do is put a MouseEvent.CLICK listener on the TextField then you check to see if the mouse clicked within the rectangle area of the embedded image.

//t is the TextField
//emb is the embedded image

t.addEventListener(MouseEvent.CLICK, onTextClick);
emb = t.textField.getImageReference("icon") as Loader;

function onTextClick(e:MouseEvent):void
{
var p:Point = new Point(e.localX, e.localY);
var r:Rectangle = btnRuleBook.getRect(e.target as TextField);
if (r.containsPoint(p))
{
//launch URL or whatever
}
}