#10682: sum fails with lower bound != 0 or 1 (upgrade maxima to 5.26)
---------------------------------------------------------+------------------
Reporter: fmaltey | Owner:
burcin
Type: defect | Status:
needs_work
Priority: critical | Milestone:
sage-5.0
Component: symbolics | Keywords:
maxima 5.26.0 binomial sum
Work_issues: domain issue, error handling in nintegral | Upstream: N/A
Reviewer: Jean-Pierre Flori, | Author:
Dima Pasechnick
Merged: | Dependencies:
---------------------------------------------------------+------------------
Changes (by newvalueoldvalue):
* keywords: => maxima 5.26.0 binomial sum
* reviewer: => Jean-Pierre Flori,
* status: needs_review => needs_work
* work_issues: several doctests need to be patched due to changes in
output format/term order => domain
issue, error handling in nintegral
* author: => Dima Pasechnick
Comment:
I would rather follow Nils approach and at least include a try/finally
statement in cas an error is raised.
Doing so, we're sure not to let the domain set to real after leaving the
function because of an error.
Then, you'd better use "P" already present in the code rather than
sage.calculus.calculs.maxima so that you're sure to use the same maxima as
the one that will be used for solving the equation. Currently it's
equivalent, but if something else changes in the future, such a choice
will make everything easier.
Finally, on my side, I'd really prefer first to use the option approach
you already implemented, and then to bakup the previous value of domain
and set it back in the finally statement (rather than setting it to
complex all the time).*
If you want to do it more cleanly, you can use
{{{
def desolve(..., domain='real')
...
olddomain=P.get('domain')
P.set('domain',domain)
...
try:
...
finally:
P.set('domain',olddomain)
...
}}}
rather than P('domain :'+str)
And the error handling at the end of nintegral should definitely be
reworked.
For example the test "if 'quad_qags' in str(v)" is useless now that Maxima
behavior has changed.
Moreover, I'm not sure the "ERROR in srt(err)" can ever be reached now
(was it before???).
I could not locate the commit corresponding to this change although I
tried a little hard. Maxima source tree is a little obscure for me. This
commit should be mentionned in the doc and would help to answer the two
above remarks about error handling. Some expert of Maxima is needed for
that.
A last question is to know what we do when nintegral returns an error
code.
Do we let the situation as is? Or do we raise an error. I would agrre that
this should be treated in another ticket, but I also don't think there is
a better choice to make. Some discussion should happen.
--
Ticket URL: <http://trac.sagemath.org/sage_trac/ticket/10682#comment:43>
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.
