I noticed that a handful of people are still subscribed to this little ol' weblog. Well, I've decided to start fresh at a new url http://bluefenix.net

Hope to see ya there. ;) Ciao!

[note: this started as a comment but started getting a little long-winded. what follows is a little gentle ribbing of Paul for a recent post he wrote about being a lazy cache programmer :P]

Mr. Stovell, you've officially lost your marbles :P j/k

okay, just to play devil's advocate...>:) I think the asp.net cache framework fits very nicely for its purpose - what you propose seems like another abstraction on top of it, for the benefit of thinking *less* about what items you should cache to *optimize* code execution time? So, just throw everything possible at it. Cache can handle that, right? hehe. nope. That's why you think about it, apply it where appropriate, and don't waste cycles calculating if you should even cache or not as part of the caching process of a production site. I dunno, seems like a wasted effort to me, especially when you start to go down the rabbit hole of deciding what should/shouldn't be cached as part of the process.

I'm not posting this to be a jerk...I'm putting your feet to the fire a little! Think it thru a little more, maybe (since I seem to be concentrating on the calculations as part of the process) move that logic outside of the [Deterministic] control flow to an external "cache minder" process. have your cake and cache it too. but the main point of cache in asp.net is to improve performance by taking anything that takes longer than O(1) and returning it nearly that fast (a guess considering that Cache exposes itself as a Dictionary to store its items) I guess that means my case (while I'm advocating;) is for using the cache as is, and make conscious decisions on what should go into it instead of letting some attribute-fu decide for me. What happens when your attribute logic is wrong? DOOM! :P hehe, not really. What do you think?

p.s. my server crapped before I had a chance to post this. Is there some new law that states that technology starts to break down the minute you start to be a smart-ass?

posted up watchin Shrek 2 with Ethan.:)

...and so dark in here.

I would like to know who thought it was a good idea to remove "import data..." and "export data..." from this particular flavor of the worst front end to an otherwise badass rdbms. I shouldn't have to fight with a program to get my databases in order. What a joke.

I recently ran into a neat little nugget of functionality in C# with events. Normally in C# when we define events we stop at something like this:

public event EventHandler<MyEventArgs> MyEvent;

The thing is, you can explicitly implement the add and remove accessors if you throw some curly braces into the mix. Why does this matter? Imagine that you have a MainForm, and a usercontrol named ControlPanel. ControlPanel contains another usercontrol called hiddenControl that exposes an event that you want to handle in MainForm, but all MainForm has access to is ControlPanel...

public event EventHandler<MyEventArgs> MyEvent{
	add{
		this.hiddenControl.MyEvent += value;
	}
	remove{
		this.hiddenControl.MyEvent -= value;
	}
}

Now you can subscribe to the event in MainForm without making the usercontrol member public in ControlPanel.

[ Currently Playing : Burn Away - Foo Fighters - One by One (4:57) ]

 hehe...

Rather surprised by the new 'updates are ready' message from Microsoft. Hmmm.

Perhaps they know just how much frustration we had on the home pc recently due to their dodgy updates.

(Fixed thanks to Scott Swigart)

Source: Windows Updates Make Me Nervous

This is just a re-statement of a forum thread that discusses the fix, but since gotdotnet is not going to be around very much longer I thought I'd post this little tidbit here as well. Basically, when I upgraded one of my sites to .net 2.0, skmMenu got upgraded right along with it. The only issue was that all of my menus would show up at the far left corner of the screen, and when you try to navigate to them over there they disappear thanks to the menu items between the cursor and the target. I think I only saw this behavior in firefox, but the code fix works in both firefox and IE. I just went thru the .js file and the javascript in the .resx for embedded javascript goodness and placed a 'px' after any integer value. The reason for this is that firefox requires measurement properties be set with appropriate identifiers when it's displaying a structured document. Anyways, here's a reprint of the code:

function skm_mousedOverMenu(menuID,elem,parent,displayedVertically,imageSource){
	skm_stopTick();
	skm_closeSubMenus(elem);
	var childID=elem.id+"-subMenu";  // Display child menu if needed
	if (document.getElementById(childID)!=null){  // make the child menu visible and specify that its position is specified in absolute coordinates
		document.getElementById(childID).style.display='block';
		document.getElementById(childID).style.position='absolute';
		skm_OpenMenuItems = skm_OpenMenuItems.concat(childID);
		if (displayedVertically){ // Set the child menu's left and top attributes according to the menu's offsets
			document.getElementById(childID).style.left=skm_getAscendingLefts(parent)+parent.offsetWidth+'px';
			document.getElementById(childID).style.top=skm_getAscendingTops(elem)+'px';
			var visibleWidth=parseInt(window.outerWidth?window.outerWidth-9:document.body.clientWidth,10+'px');
			if ((parseInt(document.getElementById(childID).offsetLeft,10+'px')+parseInt(document.getElementById(childID).offsetWidth,10+'px'))>visibleWidth) {
				document.getElementById(childID).style.left=visibleWidth-parseInt(document.getElementById(childID).offsetWidth,10+'px');
			}
		}else{  // Set the child menu's left and top attributes according to the menu's offsets
			document.getElementById(childID).style.left=skm_getAscendingLefts(elem)+'px';
			document.getElementById(childID).style.top=skm_getAscendingTops(parent)+parent.offsetHeight+'px';
			if (document.getElementById(childID).offsetWidth<elem.offsetWidth)
				document.getElementById(childID).style.width=elem.offsetWidth+'px';
		}
	}
	if (skm_SelectedMenuStyleInfos[menuID] != null) skm_SelectedMenuStyleInfos[menuID].applyToElement(elem);
	if (skm_highlightTopMenus[menuID]){
		var eId=elem.id+'';
		while (eId.indexOf('-subMenu')>=0){
			eId=eId.substring(0, eId.lastIndexOf('-subMenu'));
			skm_SelectedMenuStyleInfos[menuID].applyToElement(document.getElementById(eId));
		}
	}	
	if (imageSource!=''){
		setimage(elem,imageSource)
	}
}

function skm_shimSetVisibility(makevisible, tableid){
	var tblRef=document.getElementById(tableid);
	var IfrRef=document.getElementById('shim'+tableid);
	if (tblRef!=null && IfrRef!=null){
		if(makevisible){
			IfrRef.style.width=tblRef.offsetWidth+'px';
			IfrRef.style.height=tblRef.offsetHeight+'px';
			IfrRef.style.top=tblRef.style.top+'px';
			IfrRef.style.left=tblRef.style.left+'px';
			IfrRef.style.zIndex=tblRef.style.zIndex-1;
			IfrRef.style.display="block";
		}else{
			IfrRef.style.display="none";
		}
	}
}

These were the only two functions that needed code adjustment. Enjoy.

[ Currently Playing : Mannequin Republic - At the Drive-In - Relationship of Command [Japan (3:02) ]

I read this excellent post from ScottGu and decided to use it with a page that implemented a masterpage. I didn't have to use postbacks in my scenario, but there were links included from the masterpage. The problem is that if you use app-relative paths for your href attributes (ex: <a href="/default.aspx">) the browser (FF 2.0.0.2 and IE 7 anyways) interperets the url with pathinfo differently than a url without. The base url includes the original page (ex: http://localhost:3333/rewriter.aspx/default.aspx). Guess what? The webserver picks up the last .aspx extension, default.aspx, that bad boy doesn't exist, and you get a 404 instead of going to http://localhost:3333/default.aspx.

In the comments, Ian Oxley suggested that you can re-base links in your page/css/other static files using the <base> element in the head of your page. I expanded on it a little, since this behavior is only on one page of my site currently, and added the following code into the Page_Load event of the offending page:

this.Header.Controls.Add(
new LiteralControl("<base href=\"" +
 Request.Url.ToString().Replace(Request.RawUrl, "").Replace(Request.PathInfo, "") + 
"\">"));

Now the base tag works on both the live site and the one that WebDev.WebServer spins up as well.

[ Currently Playing : Them Bones - Alice in Chains - Nothing Safe: Best of the Box (2:29) ]