[sage-devel] Re: sage-2.8.7

2007-10-16 Thread William Stein
> > Well, could you just for fun rebuild pari? I think the compiler should > > blow up in that case, too, and I am curious why pari's build system > > doesn't catch it. > > The command > sage -f pari-2.3.2.p3 > worked fine on the suse machine. The problem isn't with building pari but with some n

[sage-devel] Re: sage-2.8.7

2007-10-16 Thread William Stein
On 10/16/07, David Joyner <[EMAIL PROTECTED]> wrote: > On 10/16/07, mabshoff <[EMAIL PROTECTED]> wrote: > -fPIC -I/home/wdj/sagefiles/sage-2.8.7/local//include > -I/home/wdj/sagefiles/sage-2.8.7/local//include/csage > -I/home/wdj/sagefiles/sage-2.8.7/devel//sage/sage/ext > -I/home/wdj/sagefiles/s

[sage-devel] Re: sage-2.8.7

2007-10-16 Thread David Joyner
On 10/16/07, mabshoff <[EMAIL PROTECTED]> wrote: > > > > On Oct 16, 6:09 pm, "David Joyner" <[EMAIL PROTECTED]> wrote: > > On 10/16/07, mabshoff <[EMAIL PROTECTED]> wrote: > > > > Hi David, > > > > > > gcc blows up, there is little we can do about that. Please check if > > > there are updates com

[sage-devel] Re: sage-2.8.7

2007-10-16 Thread William Stein
Sage-2.8.5 compiled fine on SUSE, so this problem only occurs with a very recent version of gen.pyx. I recently modified the factor function in gen.pyx so that it would do provably-correct factorization of integers. Since that's the function causing the compiler error, probably something I did cau

[sage-devel] Re: sage-2.8.7

2007-10-16 Thread mabshoff
On Oct 16, 6:09 pm, "David Joyner" <[EMAIL PROTECTED]> wrote: > On 10/16/07, mabshoff <[EMAIL PROTECTED]> wrote: > Hi David, > > gcc blows up, there is little we can do about that. Please check if > > there are updates compilers available for SuSE 10.2. If not you can > > either install your

[sage-devel] Re: sage-2.8.7

2007-10-16 Thread David Joyner
On 10/16/07, mabshoff <[EMAIL PROTECTED]> wrote: > > > > On Oct 16, 5:46 pm, "David Joyner" <[EMAIL PROTECTED]> wrote: > > On 10/16/07, mabshoff <[EMAIL PROTECTED]> wrote: > > > > > > > > > On Oct 16, 4:26 pm, "David Joyner" <[EMAIL PROTECTED]> wrote: > > > > > Hi David, > > > > Hi David, > > > >

[sage-devel] Re: sage-2.8.7

2007-10-16 Thread mabshoff
On Oct 16, 5:46 pm, "David Joyner" <[EMAIL PROTECTED]> wrote: > On 10/16/07, mabshoff <[EMAIL PROTECTED]> wrote: > > > > > On Oct 16, 4:26 pm, "David Joyner" <[EMAIL PROTECTED]> wrote: > > > Hi David, > Hi David, > > > This version does not upgrade or install on suse 9.1 amd64. > > > As far as

[sage-devel] Re: sage-2.8.7

2007-10-16 Thread David Joyner
On 10/16/07, mabshoff <[EMAIL PROTECTED]> wrote: > > > > On Oct 16, 4:26 pm, "David Joyner" <[EMAIL PROTECTED]> wrote: > > Hi David, > > > This version does not upgrade or install on suse 9.1 amd64. > > As far as I can tell Suse 9.1 AMD64 ships with gcc 3.3, which is not > C99 conform. So flint wo

[sage-devel] Re: sage-2.8.7

2007-10-16 Thread mabshoff
On Oct 16, 4:26 pm, "David Joyner" <[EMAIL PROTECTED]> wrote: Hi David, > This version does not upgrade or install on suse 9.1 amd64. As far as I can tell Suse 9.1 AMD64 ships with gcc 3.3, which is not C99 conform. So flint won't build, you need at least gcc 3.4 or higher. > > I would not m

[sage-devel] Re: sage-2.8.7

2007-10-16 Thread David Joyner
This version does not upgrade or install on suse 9.1 amd64. I would not make it a priority since these machines are due to be upgraded, but thought I'd let you know. However, it does not install on suse 10.2 either! In case this helps: sage: An error occurred while installing flint-0.2.p4 Plea

[sage-devel] Re: sage-2.8.7

2007-10-16 Thread Robert Miller
On sage.math, I tried to run the public version: [EMAIL PROTECTED]:~$ sage -- | SAGE Version 2.8.7, Release Date: 2007-10-15 | | Type notebook() for the GUI, and license() for information.| -

[sage-devel] Re: sage-2.8.7.alpha0

2007-10-15 Thread Jaap Spies
William Stein wrote: > Hello, > > I have released sage-2.8.7.rc1 here: > >http://sage.math.washington.edu/tmp/ > > In particular, this link: > > http://sage.math.washington.edu/tmp/sage-2.8.7.rc1.tar > > Hopefully this will work with no doctest failures on some systems. > Let me know

[sage-devel] Re: sage-2.8.7.alpha0

2007-10-15 Thread David Joyner
On 10/15/07, William Stein <[EMAIL PROTECTED]> wrote: > > Hi, > > I've posted rc2, which resolved a few remaining > issues (a major bug in Cython and issues I had > with two of the patches that fixed things): > > http://sage.math.washington.edu/tmp/sage-2.8.7.rc2.tar > > Also, it would be great if

[sage-devel] Re: sage-2.8.7.alpha0

2007-10-15 Thread William Stein
Hi, I've posted rc2, which resolved a few remaining issues (a major bug in Cython and issues I had with two of the patches that fixed things): http://sage.math.washington.edu/tmp/sage-2.8.7.rc2.tar Also, it would be great if somebody with SUSE could test the build. I had a major problem buildi

[sage-devel] Re: sage-2.8.7.alpha0

2007-10-15 Thread Pablo De Napoli
All test passed on Gentoo x86 linux (gcc-4.1.2) On 10/15/07, Justin C. Walker <[EMAIL PROTECTED]> wrote: > > > On Oct 14, 2007, at 6:56 AM, William Stein wrote: > > > > > Hello, > > > > I have released sage-2.8.7.rc1 here: > > > >http://sage.math.washington.edu/tmp/ > > > > In particular, th

[sage-devel] Re: sage-2.8.7.alpha0

2007-10-14 Thread Justin C. Walker
On Oct 14, 2007, at 6:56 AM, William Stein wrote: > > Hello, > > I have released sage-2.8.7.rc1 here: > >http://sage.math.washington.edu/tmp/ > > In particular, this link: > > http://sage.math.washington.edu/tmp/sage-2.8.7.rc1.tar > > Hopefully this will work with no doctest failures on

[sage-devel] Re: sage-2.8.7.alpha0

2007-10-14 Thread David Joyner
On 10/14/07, William Stein <[EMAIL PROTECTED]> wrote: > > Hello, > > I have released sage-2.8.7.rc1 here: > >http://sage.math.washington.edu/tmp/ > > In particular, this link: > > http://sage.math.washington.edu/tmp/sage-2.8.7.rc1.tar > > Hopefully this will work with no doctest failures o

[sage-devel] Re: sage-2.8.7.alpha0

2007-10-14 Thread David Roe
All tests pass on my machine as well. David On 10/14/07, William Stein <[EMAIL PROTECTED]> wrote: > > > On 10/14/07, mabshoff > On Oct 14, 3:56 pm, "William Stein" > <[EMAIL PROTECTED]> wrote: > > > Hello, > > > > > > I have released sage-2.8.7.rc1 here: > > > > > >http://sage.math.washington.

[sage-devel] Re: sage-2.8.7.alpha0

2007-10-14 Thread William Stein
On 10/14/07, mabshoff > On Oct 14, 3:56 pm, "William Stein" <[EMAIL PROTECTED]> wrote: > > Hello, > > > > I have released sage-2.8.7.rc1 here: > > > >http://sage.math.washington.edu/tmp/ > > > > In particular, this link: > > > >http://sage.math.washington.edu/tmp/sage-2.8.7.rc1.tar > > > >

[sage-devel] Re: sage-2.8.7.alpha0

2007-10-14 Thread mabshoff
On Oct 14, 3:56 pm, "William Stein" <[EMAIL PROTECTED]> wrote: > Hello, > > I have released sage-2.8.7.rc1 here: > >http://sage.math.washington.edu/tmp/ > > In particular, this link: > >http://sage.math.washington.edu/tmp/sage-2.8.7.rc1.tar > > Hopefully this will work with no doctest fa

[sage-devel] Re: sage-2.8.7.alpha0

2007-10-14 Thread William Stein
Hello, I have released sage-2.8.7.rc1 here: http://sage.math.washington.edu/tmp/ In particular, this link: http://sage.math.washington.edu/tmp/sage-2.8.7.rc1.tar Hopefully this will work with no doctest failures on some systems. Let me know what happens. William --~--~-~--~--

[sage-devel] Re: sage-2.8.7.alpha0

2007-10-13 Thread Jaap Spies
cwitty wrote: > Well, I could be convinced otherwise. I think it's a good thing that > ZZ(3.0) == 3, but Sloane sequences could do more input validation than > just calling ZZ(). I don't actually care whether an index of 20.0 > works or not. > Yes, you are right, there should be more input va

[sage-devel] Re: sage-2.8.7.alpha0

2007-10-13 Thread cwitty
On Oct 13, 1:56 pm, Jaap Spies <[EMAIL PROTECTED]> wrote: > cwitty wrote: > > On Oct 13, 10:30 am, Jaap Spies <[EMAIL PROTECTED]> wrote: > >> In sloane_functions.py sequence can now be indexed by reals!? > >> a(1.0) returns a value. Formerly an error was raised. > > > In SloaneSequence.__call__()

[sage-devel] Re: sage-2.8.7.alpha0

2007-10-13 Thread Jaap Spies
Hi Michael, mabshoff wrote: > Carl Witty has opened tickets for all the issues in trac and there are > already some patches. So if you want to help out joins us in IRC, so > that way we do not duplicate efforts. > No, sorry but I will not join the party. I need a good night rest! Do you ever sl

[sage-devel] Re: sage-2.8.7.alpha0

2007-10-13 Thread Justin C. Walker
On Oct 13, 2007, at 1:10 AM, William Stein wrote: > > Hi, > > I've posted the first alpha release of sage-2.8.7 here: > >http://sage.math.washington.edu/tmp/ > > It is probably somewhat broken. Building it, contributing > fixes, etc., would be greatly appreciated. Here're the test results

[sage-devel] Re: sage-2.8.7.alpha0

2007-10-13 Thread mabshoff
Carl Witty has opened tickets for all the issues in trac and there are already some patches. So if you want to help out joins us in IRC, so that way we do not duplicate efforts. The additional number of doctests failing on FC7 in 32 bit mode might be related to coercion issues, see #833 for the b

[sage-devel] Re: sage-2.8.7.alpha0

2007-10-13 Thread Jaap Spies
cwitty wrote: > On Oct 13, 10:30 am, Jaap Spies <[EMAIL PROTECTED]> wrote: >> In sloane_functions.py sequence can now be indexed by reals!? >> a(1.0) returns a value. Formerly an error was raised. > > In SloaneSequence.__call__() (which is shared among all sequences) the > first thing it does is

[sage-devel] Re: sage-2.8.7.alpha0

2007-10-13 Thread cwitty
On Oct 13, 10:30 am, Jaap Spies <[EMAIL PROTECTED]> wrote: > In sloane_functions.py sequence can now be indexed by reals!? > a(1.0) returns a value. Formerly an error was raised. In SloaneSequence.__call__() (which is shared among all sequences) the first thing it does is "m = ZZ(n)", to coerce i

[sage-devel] Re: sage-2.8.7.alpha0

2007-10-13 Thread Justin C. Walker
On Oct 13, 2007, at 11:22 AM, Jason Grout wrote: mabshoff wrote: >> On Oct 13, 10:10 am, "William Stein" <[EMAIL PROTECTED]> wrote: [snip] > This is odd. The test passes on my Ubuntu 32-bit P4 system. It looks > like the edges are returned in a different order for you, so the > answer > is co

[sage-devel] Re: sage-2.8.7.alpha0

2007-10-13 Thread Jason Grout
Jaap Spies wrote: > Jason Grout wrote: > >> This is odd. The test passes on my Ubuntu 32-bit P4 system. It looks >> like the edges are returned in a different order for you, so the answer >> is correct, just output differently. >> >> I can't build the alpha right now, but someone that has the a

[sage-devel] Re: sage-2.8.7.alpha0

2007-10-13 Thread Jaap Spies
Jason Grout wrote: > This is odd. The test passes on my Ubuntu 32-bit P4 system. It looks > like the edges are returned in a different order for you, so the answer > is correct, just output differently. > > I can't build the alpha right now, but someone that has the alpha built, > can you see

[sage-devel] Re: sage-2.8.7.alpha0

2007-10-13 Thread Jason Grout
mabshoff wrote: > > > On Oct 13, 10:10 am, "William Stein" <[EMAIL PROTECTED]> wrote: >> Hi, >> >> I've posted the first alpha release of sage-2.8.7 here: >> >>http://sage.math.washington.edu/tmp/ >> >> It is probably somewhat broken. Building it, contributing >> fixes, etc., would be great

[sage-devel] Re: sage-2.8.7.alpha0

2007-10-13 Thread Jaap Spies
William Stein wrote: > > It is probably somewhat broken. Building it, contributing > fixes, etc., would be greatly appreciated. A lot of test failures on Fedora 7: The following tests failed: sage -t devel/sage-main/sage/calculus/wester.py sage -t devel/sage-main/sage/li

[sage-devel] Re: sage-2.8.7.alpha0

2007-10-13 Thread William Stein
Hi, I just landed in Albuquerque, NM (free wifi, excellent power outlets, etc., at their airport!). Anyway, I get similar doctest failures on most architecture. As Michael says, if anybody wants to help on this, report the failure as a trac ticket, assign it to sage-2.8.7, and fix it. Then hopef

[sage-devel] Re: sage-2.8.7.alpha0

2007-10-13 Thread mabshoff
On Oct 13, 10:10 am, "William Stein" <[EMAIL PROTECTED]> wrote: > Hi, > > I've posted the first alpha release of sage-2.8.7 here: > >http://sage.math.washington.edu/tmp/ > > It is probably somewhat broken. Building it, contributing > fixes, etc., would be greatly appreciated. Hello, build

[sage-devel] Re: SAGE-2.8.7 "sprint"

2007-10-12 Thread mabshoff
On Oct 12, 4:59 pm, "William Stein" <[EMAIL PROTECTED]> wrote: > Hi, > > I'm going to try to put sage-2.8.7 together this evening, starting > at about 5pm my time. If anybody is interested in helping out, > I'll be on IRC #sage-devel (and in the Sage lab if you're local). Count me in, assuming