Vertical Text in Internet Explorer

 

Vertical Text

CSS seems to lack an real control over text direction – it should, but of course the reality is different.

What I found was a way of making vertical in IE using some of those pointless filters, and the support IE has for top to bottom, right to left languages. It’s a bit sneaky in doing this – but it works in IE, at least. I haven’t found a way of doing it in Mozilla.
<div style="{ writing-mode:tb-rl;
filter: flipv fliph;
white-space: nowrap;}">

Vertical Text in Internet Explorer

Alpha channel PNGs in IE

Gifs suck. PNGs (Portable Network Graphics) are much better – more colours, freedom from intellectual property rights, and (best of all) an alpha channel for transparency. Some good articles:

PNG opacity (with some nice examples)

Fuzzy Drop shadows (like on Google Maps)

Transparency with HTC files

And again

– Choosing what image to serve in PHP on the server

Related: Opacity in CSS, being able to see through DIVS – CSS Transparency for IE, CSS Opacity

Alpha channel PNGs in IE

Finding the offset of a page item

A couple of functions from my friend Jonathan Beckett for finding the x and y offsets of an item on a page. Basically, it works it’s way back up the DOM finding the offset of each parent item. There may be better ways; I haven’t checked that hard.
function get_x(obj){
var xpos = 0;
if (obj.offsetParent){
while (obj.offsetParent){
xpos += obj.offsetLeft;
obj = obj.offsetParent;
}
}
else if (obj.x) xpos += obj.x;
return xpos;
}

function get_y(obj){
var ypos = 0;
if (obj.offsetParent){
while (obj.offsetParent){
ypos += obj.offsetTop;
obj = obj.offsetParent;
}
}
else if (obj.y) ypos += obj.y;
return ypos;
}

Finding the offset of a page item

Thoughts on Coding

When writing web applications, ultimately, you probably want to look at performing load testing. This is most easily done by an application, running a script, and pretending to be a number of users.

Of course, if you’re doing something like, say, creating nodes in some sort of a hierarchy (think messages, or files/folders), then you might need to do some stuff before performing other operations – like see if your node was created, get its ID, that sort of thing.

So here’s my thought – make that easy. Put things like ‘New Node ID = ‘ into HTML comments near the top of the page. That way, it should be possible to find and read them with many of these tools.

At the moment, we’re using OpenSTA, which works. That’s about the best that can be said for it. I like the look of Webload but it’s probably too expensive.

Thoughts on Coding