From: Scott W. <swa...@my...> - 2001-11-05 17:47:06
|
>matching entry field. It would be nice to at the very least have >"headers" which would group items. For example, if I'm converting a >paper form that has a layout like this: >Foo > Number ___ > Size ___ > Quality ___ > >Bar > Number ___ > Size ___ > Quality ___ I like this idea, and had been thinking about something fairly similar. I have a few ideas for you: Instead of defining all the headers int the csv file, or when the database schema is created, it would be cool (and I don't think too hard) to do this dynamically. So if you have some records: <name> <number> <size> <quality> Foo 10 big high Foo 12 small low Bar 11 big low Bar 13 big low You could select a field a do some clustering of the records based on the values in the field. So, if I wanted to cluster on <name>, the program would scan its values, and make 2 headers, "Foo" and "Bar". This would work great for popup lists, but you could relatively easily extend it to other fields as well. The advantage of this approach is that it doesn't require thinking about how the data will be displayed when you decide what should be in each record. There is an experimental plugin interface in db, that I have been looking at a bit lately. It seems like it supports plugins for new list views. I think this would be a great plugin capability, I would definately use it. -Scott |