#10341: make MIP backend interface more Python-ic
----------------------------------------------+-----------------------------
Reporter: malb | Owner: ncohen
Type: enhancement | Status: needs_work
Priority: major | Milestone: sage-4.6.1
Component: linear programming | Keywords: LP, MIP
Author: Martin Albrecht, Nathann Cohen | Upstream: N/A
Reviewer: | Merged:
Work_issues: |
----------------------------------------------+-----------------------------
Comment(by ncohen):
Replying to [comment:22 malb]:
> Did you check my patch? For some reason p.show() still works, or is the
doctest not enough to ensure this?
Well, if your doctests pass then I guess the docstring is not enough.
Anyway, for the moment the variable's names are just "cached" inside of
MixedIntegerLinearProgram, because I thought stupid to call routines to
updates the names unless someone wants to show() it or to export it to a
file (which one never does when actually trying to solve them). There is a
method named update_variable_name, which you commented in your patch,
which does just that : before print show() or exporting the result, it
updates the variable's names. This is the only way for the variables to
have a name, and as it is usually done after the LP has been generated, it
has to use the set_col_name commands. I remember having written somewhere
that it was broken at some point, but I am never too eager to deal with
this name mess.
Oh, of course everything works fine if you comment #update_variable_names.
The .show() method will just print x_1, x_2, etc and the .lp and .mps
files will have the same kind of names. I do not think it is that awful,
and removing all this name-related stuff from the class will just make it
easier to work with.
I mean, it would have been nice to be able to deal with names, but :
* The solvers all have a different way to deal with them. If we want
to be able to deal with them all, we'll just end up with an awkward
interface
* You don't use them, I don't use them, they play no part in the
actual solving of LP which stil;l has to be considered the main point here
* The LP file formats deal with names, the solver's libraries too, but
then again people write LP in languages like AMPL or Mathprog (for GLPK),
which have a very basic set of commands available as they are only meant
to deal with LP. In Sage, we have Python which is infinitely more
expressive...
It looks like a lot of work for something we're not even sure we can
use...
Nathann
--
Ticket URL: <http://trac.sagemath.org/sage_trac/ticket/10341#comment:23>
Sage <http://www.sagemath.org>
Sage: Creating a Viable Open Source Alternative to Magma, Maple, Mathematica,
and MATLAB
--
You received this message because you are subscribed to the Google Groups
"sage-trac" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to
[email protected].
For more options, visit this group at
http://groups.google.com/group/sage-trac?hl=en.
