On Thu, Nov 13, 2003 at 02:23:51PM +0000, Carwyn Edwards wrote: > > <quote> > > I know you were just joking (the smiley gave it away) but there is a > tidy solution to the "endless if" problem. > > Make a dict[] of names->functions. This makes the whole thing rather > modular and tidy. > > </quote> > > I don't see why you need a huge "if block" or a dict map to do this? I never said "need". In fact, I said quite explicitly that there were many ways to do it. Including... > The best way to do this is with a plugin system that uses a dynamic > class/module loader. See: Quoting myself: "Define the functions wherever you want, including 'plugins' if you like." As it turns out, I have a little experience with python plugins and dynamic command lists. I like the concept. It's good for some things. > <quote> > > GIVEN A LONG LIST of name->function pairs, how do we more cleanly > deal with them? > > </quote> > > Answer: you don't have a long list. You use introspection/reflection. > Why build a mapping when you can derive it at runtime? Perhaps you should read again the text you quoted. I write "GIVEN A LONG LIST" (in caps, I might add), and you immediately say "you don't have a long list". I agree, there are many other solutions. I tried (and apparently failed) to express that I was addressing the situation where you DO have a long list. Now, the answer to YOUR question "Why build a mapping when you can derive it at runtime?": it's a design tradeoff. Do you want to pull in hundreds of lines of code to manage "plugins" when you only have 3 options and you will never have more? No. You hardcode it in 6 lines and be done. Hundreds of commands that may change? Absolutely, it's a no-brainer. In the middle, it's a harder decision. Now, back to yum, as my little side comment has dragged us far far far away.... the limiting issue here is not internal implementation, exciting as I may find it. The limiting issue is human interface and Seth's (laudable) desire to keep it clean and simple. -Michael -- Michael Stenner Office Phone: 919-660-2513 Duke University, Dept. of Physics mstenner@xxxxxxxxxxxx Box 90305, Durham N.C. 27708-0305
