Hi all, Here's a philosophical question for you all :)
I have a batch credit card decryption/payment processing circuit. The various stages all happen over a few fuseactions. I've tried to make each fuse do just one thing as much as possible, it makes it easier to debug. The payment data is stored in a session struct. I would judge the re-usability of these fuses as quite low. Perhaps they could be recycled quite well. But this is a fairly specific system. At the beginning of each fuse that requires it, I copy the session struct to the local scope, and back again at the end if required. Now that I look at it (it's at the testing stage) there are a few places where the two or more fuses which make up a fuseaction are each copying the session struct to the local scope and then copying it back again when they have finished, when in reality these actions only need to happen once per fuseaction. Is it bad style to only copy them over in the first fuse, and back in the last? It feels like it would be breaking some kind of modularity rules. Or should I put the copy over/back into the switch? Or (and I just thought of this while I was typing) should I create a separate fuse to copy the struct over, and another to copy it back, and resuse these? Maybe this could be done in a UDF? I guess what I'm really trying to decide is, what's more important: minimizing the cflock/session overhead, or maximizing the reusability of the fuses? Any thoughts? Apart from "you're the only one who knows what's best for your app, work it out yourself??" Kay. ==^================================================================ This email was sent to: [email protected] EASY UNSUBSCRIBE click here: http://topica.com/u/?bUrFMa.bV0Kx9 Or send an email to: [EMAIL PROTECTED] T O P I C A -- Register now to manage your mail! http://www.topica.com/partner/tag02/register ==^================================================================
