I wonder what I was supposed to see.
I wonder what I was supposed to see.
I have been reading up on naming standards today. It has been quite a while since I challenged by naming convention habits.
This has all come about because I was using GhostDoc today to quickly put in the bulk of comments in an assembly for me to then go through and tweak.What I found was that GhostDoc didn’t like some of my parameter naming. This was the catalyst for my reading about naming standards after thinking about naming standards for quite a while.
I read Microsoft’s naming guidelines, but it seemed to not address the naming standards of member level variables. The closest discussion it seemed to get on the issue was guidelines for static fields. What standards are people using for member level variables? I read a recent post where it seems that everyone has a different opinion.
I have recently fixed up some code from another developer. There were a few things in the code that I got a little chuckle about. The weird thing is that I have seen these coding behaviors in a few jobs that I have had.
First one that I have to laugh about (to avoid crying) is when developers comment out code blocks, then put a modified version of that code block below the commented version. Maybe this isn’t a bad idea, but seriously, if you are using a source control system then what is the point? Isn’t that what version control is all about? You can look at older versions of the same file, and usually, you can even compare the files to see what was changed.
The second one is when I come across code that basically says if true, do something, if false, then do something, else do something else. How is the else statement ever going to get hit? I’m all for defensive coding, but is this going too far?
I came across a third one last week, but I think I’ll keep that gem to myself.
I just saw Dave saying his bit about representation of geeks in movies and TV. Overall, I think geeks get displayed in a way that is very unlike the majority of the geek population. While I agree with Dave, I think that some of the ideas in movies can also be the kind of things that geeks will want to use or build, no matter how impractical they are or difficult they would be to get.
Two things spring immediately to mind. The cool data searching tool used in Minority Report and the GUI compiler in Swordfish. While the GUI compiler offers no benefit at all, I just can’t go past a great UI no matter what it does or doesn’t do.
What’s your favorite piece of fictional movie/TV software?
Scott Guthrie posted some ASP.Net tips several weeks ago. There were heaps of great ideas that he put into his presentation, one of which was about registering controls for aspx pages.
User controls and custom controls that are used on a page need to be registered. The control registration allows a tag prefix to be defined and identifies the location where the control can be found. These control registrations are normally placed at the top of the aspx markup along with the page directive. If you drag and drop an unregistered control onto the page, the registration will be added for you (with the exception of dragging user controls onto the markup view).
If you happen to come across a situation where controls change location, assembly or namespace, then every registration for the controls affected will need to be changed. This means that, when using the typical control registration method, each aspx page that uses those controls needs to be changed. This maintenance problem is mostly solved by Scott’s tip of putting the control registrations into the web.config file by adding add elements under system.web/pages/controls.
Edit - Removed comment about this not working for master pages. Something must have gone wrong with my build as it was throwing compile errors with registrations missing from the master page. After emailing Scott and retesting this, I have found that it is fine and works with master pages as expected.
Last week, a colleague told me how to get intellisense support in xml documents in the VS2005 IDE. It is really easy, although there are some interesting quirks in the process.
To pull this off, you first need to get an xsd file that defines your xml file. The IDE uses the xsd file to know what elements and attributes are available for a specified location in the document.
The next step is to get at the property grid for your xml file. Just selecting the file in Solution Explorer and then displaying the Properties window doesn’t give you the property grid contents that you need. This property grid view just displays the generic properties for a file, rather than being more specific to the xml file type. To get at the property that needs to be changed, open the xml file and show the Properties window. You should now see that a Schema property is available.
Set the Schemas property value to the file path to your xsd file. I have found that the editor for the Schemas property is quite broken. The Add button doesn’t work and also has a few other issues. It’s easiest to just copy and paste the file path from explorer to the property grid rather than fighting with the editor dialog.
Just a couple of days ago, I went to the Perisher Blue site to see how the snow was going. There was nothing but grass apart from a light dusting at the top of the high mountains.
It’s a little different now.
Yesterday I installed the new Consolas font for Visual Studio 2005. It is a very nice font to code with. Grab it here.
I have been developing some control adapters over the last month. Control adapters are really cool stuff for ASP.Net. They allow you to modify the rendering behavior of a control in a way that is independent from the control and the web site code. Scott Guthrie posted about control adapters late last year and again more recently.
One of the greatest uses I have come across for control adapters is for developers to use controls that they are familiar with, such as the intrinsic ASP.Net controls, but modify the html they produce to be more suitable to the application. I am involved in a web project that requires CSS to be used for document layout rather than tables. By using adapters, I can allow the developers to use a FormView control which they are familiar with and the adapter will take over the rendering of that control to ensure that tables are not rendered.
I was going to put out an example of creating adapters, but it looks like the ASP.Net crew might beat me to it. I kind of suspect their solutions might be different to mine so it will be interesting to see what they come out with.
I had a great time at CodeCampOz this year. Geoff (Gringo) Appleby and I drove up together and had a really good chat on the trip.
Of the sessions that I really enjoyed, first up was Windows Workflow Foundation by Dan Green. This was a really good presentation and Dan obviously knows his stuff. Unfortunately there was only so much that could be fit into the hour and WF is a big topic. I’m keen to learn a lot more about this one.
Introduction to WinFS by Chris Hewitt was an interesting glimpse at the next file system to be released. I am looking forward to seeing how this new system goes when it is released. I think the greatest thing that will come out of this database driven file system will be some amazing file searching capabilities. Being database driven, this will no doubt mean many 3rd party searching programs will be developed.