On Thu, 2004-07-08 at 18:51 -0500, wrote: > Here's an early peek at the Edd code that I've rewritten almost > completely based on your recommendations..Based on my limited testing, > the code works fine but once we agree on the general scheme of the code > I will refine it further... Very nice. Actually, I'm going ahead and going a step further to commit it and cleaned it up a little style-wise (mostly just changing to using sprintf in places instead of strcats, some spacing stuff that could well be mailer artifact, making methods static, and removing some old/debug code). The only real functional change in what I committed is that instead of having to explicitly call probeBiosDisks(), it figures it out when it needs to be called. > ii) If unique signatures don't exist, it returns with a Warning > message that attempt to use "EDD Scheme" will fail.Otherwise proceed to > (iii). > iii)Maps BIOS disk num to real device names. > iv) If (iii) fails for some reason, it returns with a Warning > Message that attempt to use "Edd Scheme" will fail, otherwise returns > with success. Right now the warnings are just printed, but some better handling of this will (obviously) have to happen. > Let me know what you think... Like I said, looked nice so I've gone ahead and committed what's currently there to the trunk of CVS. `cvs -d:pserver: anonymous@xxxxxxxxxxxxxxxxxx:/usr/local/CVS co anaconda` should get you what's currently there. I haven't tested it beyond "it builds", but plan to do so either over the weekend or on Monday. Jeremy
