Doctest fix for: series() yields wrong result depending on precision

[behackl]

In the current 8.0.beta5 I get the following:

sage: f = x/(1-x^2)
sage: f.series(x==0, 10)
1*x^2 + 1*x^4 + 1*x^6 + 1*x^8 + Order(x^10)
sage: f.series(x==0, 11)
1*x + 1*x^3 + 1*x^5 + 1*x^7 + 1*x^9 + Order(x^11)

This seems to be a regression from somewhere between sage-7.5 (where the result is still correct) and sage-7.6.

Depends on #22969

CC: @rwst

Component: calculus

Author: Ralf Stephan

Branch/Commit: b121061

Reviewer: Benjamin Hackl

Issue created by migration from https://trac.sagemath.org/ticket/22959

[behackl]

comment:1

As this means that series delivers a mathematically wrong result, I've set the priority to blocker (and this should really be fixed before 8.0).

[rwst]

comment:2

pynac/pynac#246

[rwst]

Upstream: Reported upstream. Developers acknowledge bug.

[rwst]

Changed upstream from Reported upstream. Developers acknowledge bug. to Fixed upstream, in a later stable release.

[rwst]

Dependencies: pynac-0.7.7

[kcrisman]

comment:4

Nice work!

[rwst]

Branch: u/rws/series___yields_wrong_result_depending_on_precision

[rwst]

New commits:

b121061

22959: Doctest fix for: series() yields wrong result depending on precision

[rwst]

Author: Ralf Stephan

[rwst]

Changed upstream from Fixed upstream, in a later stable release. to none

[rwst]

Changed dependencies from pynac-0.7.7 to #22969

[rwst]

Commit: b121061

[behackl]

comment:7

Thanks a lot for your work, Ralf! Patch LGTM, I'm happy to set this to positive_review as soon as the patchbot tested it.

[behackl]

Reviewer: Benjamin Hackl

[fchapoton]

comment:8

For your information, the best way to have a patchbot run immediatly on your prefered ticket is to launch your own patchbot yourself on this ticket...

[behackl]

comment:9

Replying to @fchapoton:

For your information, the best way to have a patchbot run immediatly on your prefered ticket is to launch your own patchbot yourself on this ticket...

Usually I'd do that, thanks. :-)

However, I'm at a conference and have a very unstable internet connection, so I don't want to build the latest beta on my laptop. And as of a rather unlucky coincidence, my VPN connection to our university servers does not work either, so I also can't start the patchbot we have there.

[behackl]

comment:10

Running make ptestlong on my laptop lets the test run through, except for being stuck on the has_mathematica()-call in src/sage/doctest/external.py. I don't think that this problem is related to the changes on this ticket, so this is positive_review from me.

[vbraun]

Changed branch from u/rws/series___yields_wrong_result_depending_on_precision to b121061