Okay, I promised the last one would be it for the night. I lied.
My plan of attack on the CGI:
Well, I could just forge right ahead and build the cart-viewing code and stuff now...but really, there's a little bit of research I should be doing.
The product (a trademark-infringement-lawsuit-happy company's card game elements) does not have a set value for each type of thing. Each thing could cost one thing one day, and on another update cost something else entirely.
One notion that occurred to me was that, since prices for these things are mass-produced (I won't say what program is used for it, but its initials are Cl**ster), I could build code that would pull a bunch of compiled Cl**ster lists, work all the values together, and generate a price file for use until the *next* Cloister update. When someone looks at a page of singles, the names and stock values are pulled from the inventory file, the names and prices are pulled from the price file, and the web page is generated from the two.
Well, I can build a little utility to convert downloaded Cl**ster lists with small difficulty (famous last words) so I just need to find out how to get reliable Cl**ster lists, and how to make sure they're commercially viable.
If they generally aren't, I need to figure out how to set up my own Cl**ster and if *that* is commercially viable.
And if THAT isn't, I need to build my own damn price generator, whole cloth.
On a side note, I've always wanted to build a little program that you feed a trade into and it tells you how good or bad that trade would be for you. You enter what you're offering, what the other person is offering, and it would take a list of cards you want (along with values indicating how badly you want them), a Cl**ster list, and produce a rating for it (probably a decimal value, that is, if what they offer is effectively worth 11 dollars and what you offer is effectively worth 10, the value of the trade would come up "1.1". While it wouldn't be all that handy for desktops ("What do you mean, you only trade at your house?") it might be handy for laptops, and just imagine having it on a PDA. :)
Building a price generator isn't a good idea. For one, I haven't kept up with the price environment...I've been working on a program. ;) Second, the prices are so unpredictable in many cases that basing prices on rarity or on number of words on the card just won't produce realistic results.
Another idea I had once (and never wrote down) was a power index. It would take a list of cards - presumably an entire new set - and go through, calculating an abstract 'power rating' for each card based on cost, how much "power" each word on the card has, and how many words are on the card. (Plus, if the card contains something to the effect that you and your opponent do something, it notices and reduces the power of the effect significantly.)I guess I need to research Cl**ster indices. Hopefully I'll do that later, assuming my SO didn't decide to use the computer tonight.(I spent a lot of time playing M*gic, so I imagined a lot of handy little tools for it. As I started going through _Learning Perl_, it occurred to me that Perl has a LOT of facilities for this kind of thing, and would probably make writing these tools easier.)
