Hi William, On 28 Jul., 05:01, William Stein <wst...@gmail.com> wrote: > Hi, > > I took the sage-4.1.1.alpha1 release build I had, then did "./sage > -bdist", took the result, extracted it, and did "make test". > > 1) It sits there and builds the documentation again, which takes a > *long* time. It shouldn't do this for a binary. > > 2) Worse, every single test failed, with errors like this: > > sage -t > "/home/wstein/build/sage-4.1.1.alpha1/dist/sage-4.1.1.alpha1-x86_64-Linux/devel/sage/doc/common/buil > der.py" > File "./builder.py", line 18 > from > /home/wstein/build/sage-4.1.1.alpha1/dist/sage-4.1.1.alpha1-x86_64-Linux/devel/sage/doc/common/build > er import * > ^ > > 3) I tried do "./sage" to run Sage, then typed "make test" again > about 10 minutes ago. For some reason, the docs are building again... > and I expect the same behavior as above after that finally finishes. > > Buiding Sage, doing "./sage -bdist", then extracting the result and > having "make test" 100% is a blocker for making the sage-4.1.1 > release. I've made this: > http://trac.sagemath.org/sage_trac/ticket/6645 > > (I'm not sure sage-4.1 was any better, by the way.) > > William
just take a Sage source tree, and do "make && make". This will build all of the documentation *twice* in a row. So the "make" problem you reported is not restricted to binaries, and I think this problem exists since the ReSTification of the docs (Sage 3.4? Sage 4.0?). IMHO, it would be far more important to solve the problem that "sage - clone" rebuilds all of the documentation in a better way than using these hard links recently introduced, an issue that was discussed here before (by David Loeffler and Minh IIRC). Maybe it's possible to solve both problems at the same time. The first time I looked into how Sphinx checks whether it should rebuild which docs I gave up, when I couldn't figure out if Sphinx pickles the doc files together with their absolute path names (which would be pretty annoying in light of "sage -clone"), or with relative path names only, or however. Cheers, Georg --~--~---------~--~----~------------~-------~--~----~ 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 -~----------~----~----~----~------~----~------~--~---
