So we’re trying a new procedure to combat my restless leg syndrome (RLS). It has been encouraging thus far, but unfortunately my usual dose (300mg of Madopar) isn’t working for some reason and I’m up, again, wondering if I’m stuck with this condition..
What better a time than to look ahead to future blog posts? I’ve had some interesting wait times trying to contact Telstra. Our home phone has a fault and I’m going to need it since I’m working from home for a few days stating today. I feel a rant coming on!
This week I rolled out some code I wrote early last year which generates a simplistic CRUD (Create/Read/Update/Delete) data access layer (complete with serializable data transfer objects!) and I’m seriously considering putting it up (probably at CodePlex) under some form of GPL license for free. It uses an interesting mix of database libraries to do schema analysis as the basis for the code generation.
I’ve used a semi-template driven initiative to form the outputted code (written in C#). It’s kind of like a poor man’s Enterprise Library Data Application Block, though it has only been tested against SQL Server 2005 and Access. It generates SQL statements (e.g. "SELECT [Column] FROM [Table] WHERE [PK] = <value>") but doesn’t out-of-the-box support stored procs (though can be modified to support them).
What I’m keen to learn is – would anyone have a use for something like this? It’s a heck-of-a-lot lighter weight than NetTiers or CodeSmith and has about 10% of the setup and configuration time. It means you can spend far more time tailoring out outputted code (by editing the templates) and it’s great for use with Unit Tests!
Give me your feedback and it’ll help me decide whether to polish the code and release it!!