On Wed, Feb 25, 2009 at 2:33 PM, Maurizio <maurizio.gran...@gmail.com> wrote: > > Carl, > this is certainly reasonable, but maybe a little unfeasible. ... > Can you derive some information from this description, or has this > been useless? Maybe I can provide something like the last part of the > notebook, placing some intermediate expression by hand.
The description doesn't help me. I'm willing to take a brief look if I have a simple reproduction case, but your description doesn't give me much hope of reproducing the problem. (It might be enough to give a hint to somebody familiar with the code, but I'm not.) You could try: test = loads(dumps(bad_expr)) where bad_expr is the slow one. If test acts the same as bad_expr (is similarly slow), then that means that you can just do: save(bad_expr, 'slow_substitution.sobj') and post slow_substitution.sobj. > Could you also show me where should I look at for better understanding > this? Is it a good idea to verify the types of the expressions used? > Is it a problem interfacing to any symbolic package different from the > ordinary maxima one? I don't exactly recall if I took advantage of > others. I see that you're substituting in a symbolic value. Does it change anything if you change Vgr_pk to a numeric value instead? Looking at types is good. Probably where I would start is by running the slow command, pressing Control-C when it's about halfway through, and looking at the backtrace to see if I could figure out what was going on. The problem would almost certainly be related to something on the backtrace, so that's a reasonable place to start. My first guess would be that you'll discover that the slowness is because it's interacting with maxima somehow; if that's the case, then you need to figure out why. Using other symbolic packages might or might not be a problem. (It's not supposed to be a problem, but we're assuming there's a bug somewhere. Interfaces between separate packages are always a fruitful place for bugs.) Carl --~--~---------~--~----~------------~-------~--~----~ To post to this group, send email to sage-devel@googlegroups.com To unsubscribe from this group, send email to sage-devel-unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/sage-devel URLs: http://www.sagemath.org -~----------~----~----~----~------~----~------~--~---
