Hi Martin, On Feb 23, 2:41 pm, Martin Albrecht <m...@informatik.uni-bremen.de> wrote: > Simon, can you trace the commands that go in and out of Singular for this? I > think you can set some logfile attribute for the Singular pexpect interface.
Yes, actually it results in two log-files. Both are at http://sage.math.washington.edu/home/SimonKing/tests/ > Simon reported this doctest failure with his custom Singular 3-1-0 and > sage-3.3: > > > > File > > > "/home/king/SAGE/devel/sage-3.2.3/devel/sage/sage/interfaces/singular.py" > > >, line 116: > > > sage: I2 > > > Expected: > > > x1^2*x2^2, > > > x0*x2^3-x1^2*x2^2+x1*x2^3, > > > x0*x1-x0*x2-x1*x2, > > > x0^2*x2-x0*x1*x2 > > > Got: > > > x1^2*x2^2, > > > x0*x2^3-x1^2*x2^2+x1*x2^3, > > > x0*x1-x0*x2-x1*x2, > > > x0^2*x2-x0*x2^2-x1*x2^2 Martin, apologies to you since I wrote off list that the expected output of the doc test does not fit to the Singular output. It does, if you take std or groebner in Singular-3-0-3/4 and if you do *not* use option(redSB). I tested the corresponding example (line 113 ff in singular.py) in various versions of Singular: Singular 3-0-4 or Singular 3-0-3: > std(I1); _[1]=x1^2*x2^2 _[2]=x0*x2^3-x1^2*x2^2+x1*x2^3 _[3]=x0*x1-x0*x2-x1*x2 _[4]=x0^2*x2-x0*x1*x2 > slimgb(I1); _[1]=x0*x1-x0*x2-x1*x2 _[2]=x0^2*x2-x0*x2^2-x1*x2^2 _[3]=x1^2*x2^2 _[4]=x0*x2^3+x1*x2^3 > groebner(I1); _[1]=x1^2*x2^2 _[2]=x0*x2^3-x1^2*x2^2+x1*x2^3 _[3]=x0*x1-x0*x2-x1*x2 _[4]=x0^2*x2-x0*x1*x2 Singular-3-1-0-Beta does > std(I1); _[1]=x1^2*x2^2 _[2]=x0*x2^3-x1^2*x2^2+x1*x2^3 _[3]=x0*x1-x0*x2-x1*x2 _[4]=x0^2*x2-x0*x1*x2 > slimgb(I1); _[1]=x0*x1-x0*x2-x1*x2 _[2]=x0^2*x2-x0*x2^2-x1*x2^2 _[3]=x1^2*x2^2 _[4]=x0*x2^3+x1*x2^3 > groebner(I1); _[1]=x1^2*x2^2 _[2]=x0*x2^3-x1^2*x2^2+x1*x2^3 _[3]=x0*x1-x0*x2-x1*x2 _[4]=x0^2*x2-x0*x2^2-x1*x2^2 Interestingly, groebner() in Singular-3-1-0-Beta yields a result that differs from both slimgb and std. However, everything is alright. If you use option(redSB), you get _[1]=x1^2*x2^2 _[2]=x0*x2^3+x1*x2^3 _[3]=x0*x1-x0*x2-x1*x2 _[4]=x0^2*x2-x0*x2^2-x1*x2^2 in all Singular versions and with all three commands (std, slimgb, groebner). Conclusion: Wouldn't it make sense to use option(redSB) for the doc tests, since otherwise the result is mathematically not well defined? Cheers, Simon --~--~---------~--~----~------------~-------~--~----~ To post to this group, send email to sage-devel@googlegroups.com To unsubscribe from this group, send email to sage-devel-unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/sage-devel URLs: http://www.sagemath.org -~----------~----~----~----~------~----~------~--~---
