On Dec 18, 8:17 am, Jeroen Demeyer <jdeme...@cage.ugent.be> wrote: > On 2010-12-18 17:10, John H Palmieri wrote: > > > > > > > I took a vanilla 4.6 and typed > > > ./sage > > -upgradehttp://sage.math.washington.edu/home/release/sage-4.6.1.rc0/sage-4.6.... > > > It said > > > The following packages will be upgraded: > > > cvxopt-1.1.3 ecl-10.4.1 examples-4.6.1.rc0 extcode-4.6.1.rc0 > > jinja2-2.5.5 lapack-20071123.p2 matplotlib-1.0.0.p0 maxima-5.22.1 > > mercurial-1.6.4.p0 mpir-1.2.2.p2 numpy-1.5.0 pari-2.4.3.alpha.p0 > > patch-2.5.9 pil-1.1.6.p4 pygments-1.3.1.p0 readline-6.1 > > sage-4.6.1.rc0 sage_scripts-4.6.1.rc0 sagenb-0.8.10 scipy-0.8 > > sphinx-1.0.4.p5 > > > Mercurial is included. > > Thanks John! I turns out I was not using a vanilla 4.6, but a 4.6 with > Mercurial already upgraded "by hand" (./sage -i mercurial...). This > Sage has a spkg/optional/mercurial-1.6.4.p0.spkg but no > spkg/standard/mercurial-* > > From this setup, it seems upgrading is doomed to fail because mercurial > is built again but it fails because it's not in spkg/standard. Do we > consider this a bug or is upgrading from a non-vanilla Sage something > one simply shouldn't do?
As far as I know, upgrading from a non-vanilla Sage should be avoided. The upgrade process does print out a big warning message at the beginning, after all. -- John -- To post to this group, send an email to sage-devel@googlegroups.com To unsubscribe from this group, send an email to sage-devel+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/sage-devel URL: http://www.sagemath.org
