stackoverflow in Maxima integral

[kcrisman]

sage: k=var('k')
sage: integral(sin(k*x)/x*erf(x^2), x, 0, oo)
RuntimeError: ECL says: C-STACK overflow at size 4259840. Stack can probably be resized.

This is related to this ask.sagemath question though I couldn't reproduce this behavior.

Upstream: https://sourceforge.net/p/maxima/bugs/3144/

Upstream: Completely fixed; Fix reported upstream

CC: @nbruin @sagetrac-tmonteil

Component: calculus

Author: Frédéric Chapoton

Branch/Commit: 22385d9

Reviewer: Nils Bruin

Issue created by migration from https://trac.sagemath.org/ticket/20467

[nbruin]

comment:1

As far as I can see, this happens in vanilla maxima on SBCL too:

Maxima 5.36.1 http://maxima.sourceforge.net
using Lisp SBCL 1.2.11-1.fc22
(%i1) integrate(sin(k*x)/x*erf(x^2),x,0,inf);
Maxima encountered a Lisp error:
Control stack exhausted (no more space for function call frames).

This is not the latest version, but it's the kind of bug that, unless it's been specifically addressed, it's probably still there. Might want to report it (why is there no online maxima calculator with an up-to-date maxima? maximacell or something like that?)

[kcrisman]

Upstream: Reported upstream. No feedback yet.

[kcrisman]

Changed upstream from Reported upstream. No feedback yet. to Fixed upstream, but not in a stable release.

[mkoeppe]

comment:4

Seems fixed in 9.2.beta10

[kcrisman]

comment:5

As always, verify with doctest :-)

[fchapoton]

Branch: u/chapoton/20467

[fchapoton]

Commit: 22385d9

[fchapoton]

comment:6

here is a doctest

---

New commits:

22385d9

adding one doctest for integral

[fchapoton]

Changed upstream from Fixed upstream, but not in a stable release. to Completely fixed; Fix reported upstream

[nbruin]

Reviewer: nbruin

[vbraun]

comment:9

Merge failure on top of:

cb51da1 Trac #17965: Uniformize the API to compute the inverse of an element

a833b0e Trac #34491: Upgrade igraph to 0.9.10

7747f43 Trac #34467: fix random doctest failure in EllipticCurveHom_velusqrt

cba244e Trac #34466: fix various linter errors

5d89d36 Trac #34228: tox -e docker-...-incremental

627b2bd Updated SageMath version to 9.7

author '' does not look right

[nbruin]

comment:10

Indeed. Since there's now a doctest submitted, there is actually something to be committed, so there needs to be an author.

[nbruin]

Author: Frédéric Chapoton

[vbraun]

comment:11

Merge failure on top of:

5278916 Trac #34203: Find and uninstall broken installed SPKGs and wheels

ed76311 Trac #34118: Update sympy to 1.11.1

97beeae Trac #33981: latte_int: ExponentialSubst.h:62:3: error: ISO C++17 does not allow dynamic exception specifications

787b4be Trac #33640: sage fails to factor some easy expressions

700a1f5 Trac #33613: Add pip packages to conda environments, add missing conda.txt files

92b1db1 Trac #33431: Category of submodules of sage.modules.free_module.FreeModule_ambient_pid

d238396 Trac #33093: 5 doctests failing related to external program octave

509ed92 Trac #33002: Method tikz of polyhedron class can now return an object of type TikzPicture

2a41c6e Trac #32669: Adding upper and lower Bruhat cones of M. Dyer to sage/combinat/root_system/reflection_group_real.py

12756f6 Trac #29619: Matrix and Components should have a sparse iterator

cb51da1 Trac #17965: Uniformize the API to compute the inverse of an element

a833b0e Trac #34491: Upgrade igraph to 0.9.10

7747f43 Trac #34467: fix random doctest failure in EllipticCurveHom_velusqrt

cba244e Trac #34466: fix various linter errors

5d89d36 Trac #34228: tox -e docker-...-incremental

627b2bd Updated SageMath version to 9.7

reviewer 'nbruin' does not look right

[nbruin]

Changed reviewer from nbruin to Nils Bruin

[vbraun]

Changed branch from u/chapoton/20467 to 22385d9