[sage-devel] Re: Proposal: Remove GNUTLS from Sage.

My vote: [X] Yes, remove them! Somewhat coincidentally, Jason and I were looking at this today. The express summary is that Jason and I have put together a patch today that solves the FlaskNB+HTTPS problem (caused by an attempt to use TLS with the latest Twisted 2), and to do so it rips out G

Re: [sage-devel] crash on source build of 4.7 on Linux (CentOS 5.6) at MPIR

I reported this problem back in April as well, and I thought I had found a fix for it, but after some digging it appears the problem was entirely on my end: a corrupted /usr/bin/nm. I was looking in the wrong place for the problem! Anyway, "yum reinstall binutils" has fixed this problem and th

[sage-devel] Re: MPIR seg fault when building Sage from source on Linux

OK, I tracked down the source of the problem: /bin/usr/nm was corrupted. After a re-installation of binutils, things seem to be progressing better. -- 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...@goo

[sage-devel] .hg and build directories in binary tarball?

Are ".hg" and "build" directories supposed to be in the binary tarballs? I'm looking, in particular, at * sage-4.6.2-linux-64bit-red_hat_enterprise_linux_server_release_5.6_tikanga-x86_64-Linux.tar.gz.

Re: [sage-devel] MPIR seg fault when building Sage from source on Linux

Short answer: gcc 4.1.2 running on a fairly stock CentOS 5.5 with kernel 2.6.18-194.26.1.el5 (x86_64). It is running on a VM (kvm), with source files sitting on an NFS served partition (I mention this because I know NFS+VM+ntp/time+make can cause problems). I've compiled sage in the past on ot

[sage-devel] MPIR seg fault when building Sage from source on Linux

I've just downloaded Sage 4.6.2 and have spent the past hour trying to sort out an error with the MPIR 1.2.2.p2 compilation. Here are some extracts which could help someone tell me if I've done something strange. Ian #./configure output: ###

[sage-devel] Re: Maxima upgrade problems and hard-coded string "/Users/robert/..."

> No, this is a binary build by Robert Bradshaw and has nothing to do > with Robert Dodier. Ooops, sorry. > > It seems this problem extends to clisp.bin: > > > [ijsto...@moose /Developer/sage]$ ./local/bin/clisp.bin > > ./local/bin/clisp.bin: /Users/robert/sage/current/local/lib/clisp-2.46/ > >

[sage-devel] Re: Maxima upgrade problems and hard-coded string "/Users/robert/..."

On Jan 12, 11:18 am, "William Stein" wrote: > > The main problem seems to be that "/Users/robert/sage/current/local" > > is scattered throughout my Sage install. > > That's not a problem -- it's just where the Sage binary you have was built. Unfortunately, that is not true. We have no one here

[sage-devel] Maxima upgrade problems and hard-coded string "/Users/robert/..."

I just did a sage -upgrade (I was at 3.2.1, I believe) and the Maxima update failed. It said to email this list with details, so here goes: The main problem seems to be that "/Users/robert/sage/current/local" is scattered throughout my Sage install. There is nothing obvious in the install.log