If I may ba a curmudgeon for a while and point out some things about DBIx::HTMLTable

• It takes it's data in an array of arrays. Nothing wrong with that, but it's not really specific to DBI, so why is it in DBIx?
• Considering that it takes it's DBIx, I'd like to see a mode where it takes a statement handle and does it's own fetches - thereby reducing memory consumption.
• I see your point about the attributes. The documentation describes the attributes the function supposedly takes, but does not give any incling on how to describe which attribute goes where (like what to do if you want alternating rows in different colors, or different onclick handlers on different cells). Note that that could simply be a deficiency in the documentation - possibly the author considers his methods obvious.

Personaly, I'd contact the author to ask him how to do the alternating rows trick. Then if he doesn't have an answer, or if he doesn't answer, or if you don't like his answer, you can extend DBIx::HTMLTable with your own code. Perhaps with something where you can declare per-column and per-row callbacks that return a list of attributes, or something like that.

Oh, and an auto-fetch mode. That would probably be the first thing I'd add if I were messing with this module.

Having said that, thanks for mentioning CGI::AJAX. I haven't heard of it before, and now that I have I'm itching to try it in some project...