jfw.id.au

This is just a mini rant about Visual Studio .Net 2003 and it’s inability to be concise and clean when creating HTML and CSS.

It understands the clean shorthand notation as follows

padding:6px 0;

but insists on turning that into

PADDING-RIGHT: 0px; PADDING-LEFT: 0px; PADDING-BOTTOM: 6px; PADDING-TOP: 10px;

I can see no good reason for this, and no way to stop it.  At half passed midnight on a Monday morning I’m at work finalising some content for a project that is supposed to go live in less than 24 hours and I’m fed up with the cruft that VS.Net is producing.

I’m over it!

[censored]

This is just a quick rant to say:

The svn add command exists for a reason.
Learn it, use it, love it!

I went into work for a few hours today as we have a rather pressing deadline for the current project and it would be nice to get stuff finished on time to allow for some testing. After making a bunch of changes to the code that I was working on I was ready to compile and give it a test. I decided to do the prudent thing and update my local copy so that I coud test what I had done with the latest version of the project.

Unfortunately someone(s) else had incompletely checked in some work since my last update. I say incomplete because they had obviuosly created new files and the VS.NET project file had references to where they should be, but they didn’t exist in the repository.

After some carefully thought out exclusions in the solution and a quick informative email later I was able to get my changes to compile. Even though I have not completed what I was working on, and hence not checked it back in I emailed the other developers to let them know what I had done (the exclusions) and why I had done it. Obviously excluding someone elses work from a project solution is not the best thing to do when there are tight deadlines. I didn’t want the things that I had excluded to become overlooked simply because the original developer(s) thought that they were now complete.

If we were running test driven development that would not be a problem as it would be caught in the next build and test phase, however we have not yet reached that point of restructure of our methodologies.

At work we run a Cisco PIX that is also a VPN Concentrator. For the last month or so I have been using the Cisco Windows client software for connceting into work to do any after hours tasks. A couple of recent events have forced me to re-evaluate the method I use to connect in. Primarily, the windows service that is installed with the client has been randomly refusing to startup correctly with no apparent reason. Secondly, I’ve recently purchased a laptop and with it dropping into hibernation or standby whenever I walk away for more then a couple of minutes (something that happens often with a small child) I keep getting disconnected.

I have a small LAN with 2 servers and 2 client machines. The 2 servers are Wallace, a Centos4 installation that basically runs the network, and Gromit, a windows 2003 server for basic development testing and the beginnings of my HTPC. The 2 clients are my norbert, the desktop in the office, and puddlejumper, my reently acquired laptop.

I got the official Cisco client installed and working on Wallace, with 1 small exception. I could not for the life of me manage to get the rest of the LAN able to make use of the connection. I started to google for options and quickly found an article on Linux Resources on the Dartmouth College math faculty website.

This page in turn lead me to the vpnc information for Dartmouth, and I eventually ended up find a couple more helpful articles on configuration of vpnc.

__(’Read the rest of this entry »’)

At work we are all pretty busy with the project known as ‘Phase 2′. A project that the sales and marketing team had decided on a deadline before we even knew that it existed. This should be the last, or at the worst next to last time that something like this happens on such a massive project.

I have no illusions about it not happening again, however we can dream can we not?

Any way, the point of this post is simply to marvel at a lovely way to do some PL/SQL to build a bit of a data layer for the extensions that we are adding. I’ve not done a lot of PL/SQL in the past, and the stuff that I have done is fairly basic so I was quite impressed with the simplicity and functionality of this little block:

PROCEDURE getMenu
   (pDetails    OUT grcDetails,
    pMenuID   IN Menu.MENUID%TYPE DEFAULT NULL) IS
BEGIN
   OPEN pDetails FOR
   SELECT *
   FROM MENU
   WHERE MENUID = DECODE(pMenuID, 0, MENUID, NULL, MENUID, pMenuID);
END getMenu;

All the smarts of the statement are in the DECODE. For those that don’t recognise it, this is a small function written in PL/SQL for Oracle. Basically this function allows us to load all of the records in the MENU table, or the one specified by the pMenuID if it is supplied.

I was just impressed with it’s power and simplicity and had to share.

During our recent Oracle upgrade we encountered a lovely little bug.  This was a particularly annoying bug as it was not publicly acknowledged, nor was a patch publicly available.

The bug is apparently specific to deployments of Clusterware from release 2 of 10G database server on EMC storage devices, or devices that use the EMC powerpath software for multi-channel communication with the disk device.

Clusterware is very touchy during installation, and if anything is not 100% right it complains and dies.  On a brand new storage device, using brand new disks with brand new partitioning on them we were unable to complete the installation without help from Oracle support.

__(’Read the rest of this entry »’)