On Wed, Jun 16, 1999 at 11:45:40PM -0400, Richard Gordon wrote: > At 01:11 +0000 6/17/1999, Richard K. Moore wrote: > >I'm curious as to why people haven't been advising the use of a hash (or > >two). Alan's application seems to be one with only a small amount of data. > >Is the Schwartzian Transform incredibly faster as a solution for large > >amounts of data? > > I had essentially the same thought since hashes are very fast and, > until earlier today, I had never even heard of a Schwartzian > Transform. Efficiency is a splendid thing, but short of making time > back up, I don't know how my perl stuff could run much faster than it > does now using extremely ordinary hash and array techniques. Okay, let's see some Perl code that uses hashes to sort an array. I am very intrigued by this idea. > In Solaris, I can execute 21 successive web file maintenance routines > on about 180 linked files in 3 directories in less than 40 seconds. > The text is maybe 5-6 million characters in the aggregate and many of > the routines are not trivial. When cgi's are run later, the return of > results appears to be limited by dialup bandwidth more than anything > else. We're not worthy. We're not worthy. > For several years, I was MIS director or a payphone company that used > FoxPro- one of our programmers made some really nice, elegant stuff > that I admired, but her production work was always overcoded and > undertested. Another one who was a retired air force fighter pilot > took a much more functional approach and made things that worked and > showed up on time and roared even if they were largely based on brute > force. Who do you guess the guys who sign the checks told me to get > rid of? Does this have a point? Are you suggesting that you can either write brute force code that works well, or elegant code that doesn't, but no other combination? Hmm... > Maybe Schwartzian Transform is cool and maybe it isn't, but it > doesn't seem likely to put any more money in your pocket in either > case. Maybe you still don't get it. ;) Ronald ===== Want to unsubscribe from this list? ===== Send mail with body "unsubscribe" to macperl-request@macperl.org