integrate(sqrt(a+%i*a*tan(c+d*x))/tan(c+d*x)^(3/2),x) and integrate((a+%i*a*tan(e+f*x))^(7/2)*(A+B*tan(e+f*x))/(c-%i*c*tan(e+f*x))^(13/2),x) take several orders of magnitude more time after commits [c68475] and [0ca8a1].
Something strange is going on there. With GCL (and probably also ECL), the calculation stalls, but when I press Ctrl+C to interrupt it, it immediately prints the result. Will investigate.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Fixed by commit [317711].
Maxima's GCD algorithm used sort, whose behavior is unspecified when two items compare equal. GCL decides randomly, while SBCL's deterministic choice was "lucky". The code now uses stable-sort with a better sorting predicate.
The first integral is back to normal with commit [5f89f7]. The other one will also be fixed soon.
Related
Commit: [5f89f7]
Fixed by commit [386344].
Related
Commit: [386344]
Last edit: David Scherfgen 2026-10-08
The first integral seems to take forever in ecl and gcl, even with these commits.
Something strange is going on there. With GCL (and probably also ECL), the calculation stalls, but when I press Ctrl+C to interrupt it, it immediately prints the result. Will investigate.
Fixed by commit [317711].
Maxima's GCD algorithm used
sort, whose behavior is unspecified when two items compare equal. GCL decides randomly, while SBCL's deterministic choice was "lucky". The code now usesstable-sortwith a better sorting predicate.Related
Commit: [317711]