desolve_system is broken for a system of one equation

[rhinton]

desolve_system fails for a system with only one equation:

sage: t = var('t')
sage: x = function('x', t)
sage: de1 = diff(x,t) + 1 == 0
sage: desolve_system([de1], [x]) 

CC: @robert-marik

Component: calculus

Keywords: maxima, symbolics

Author: Robert Mařík

Reviewer: David Joyner, Ryan Hinton

Merged: sage-4.6.alpha2

Issue created by migration from https://trac.sagemath.org/ticket/9823

[robert-marik]

comment:1

The patch solves the problem by passing to desolve_laplace. Both desoove_laplace and desolve_system use function desolve from Maxima and perhaps bouth could be merged into one function. Perhaps when solving #9824 ?

Also removes unnecessary spawned Maxima processes.

Install after the patch for #9835.

[robert-marik]

Depends on Ticket #9835

[robert-marik]

comment:2

Attachment: trac_9823.patch.gz

Updated the patch - removed plotting picture from testing, since Sage complains

verbose 0 (3495: plot.py, generate_plot_points) WARNING: When plotting, failed to evaluate function at 200 points.

on one of my computers (Debian, AMD 64 bit).

[wdjoyner]

comment:3

Does this really depend on 9835? It seemed to apply and test fine for me without it.

[robert-marik]

comment:4

You are right, thanks. Both patches touch different part of the same file and I expected numbers from the patch utility. But both patches are independent and #9835 can be installed on the top of this patch. Thanks for mentioning this.

[wdjoyner]

comment:5

I have tested this without 9835 and it passes fine. Also, the patch is very simple and does as it says and also adds a doctest illustrating the new fix.

Positive review from me, without 9835. I am currently also testing it with 9835.

[wdjoyner]

comment:6

All tests passes even with 9835.

Positive review from me but maybe rhinton should look at it?

[rhinton]

comment:7

Looks great to me!

[loefflerd]

Reviewer: David Joyner, Ryan Hinton

[loefflerd]

Author: Robert Mařík

[qed777]

Merged: sage-4.6.alpha2