actually, sage: standard_packages() says that the following standard spkg's are not installed in a t2 build of sage-4.3.4 I made.
['cddlib-094f.p5', 'flintqs-20070817.p4', 'gfan-0.4plus', 'moin-1.9.1.p1', 'palp-1.1.p1', 'pil-1.1.6.p2', 'polytopes_db-20100210', 'sagetex-2.2.3.p0', 'scipy-0.7.p4', 'scipy_sandbox-20071020.p4', 'sqlalchemy-0.5.8', 'tachyon-0.98beta.p10', 'weave-0.4.9.p0'] Is this normal? On Mar 22, 9:47 pm, Dima Pasechnik <dimp...@gmail.com> wrote: > Hi all, > > scipy is not compiled in that build, despite it being a standard package. > It just won't build by default on Solaris (while it builds there if I do > sage -f scipy-0.7.p4.spkg, etc). > > This is not only the issue with this build by Minh, > but I also see this when I build sage myself on t2. > > This should also explain why the size of the archive is smaller than > it it should be > (perhaps some other libs aren't there, either) > > Perhaps someone in another timezone (it's almost 10pm here, and my > teaching started at 8:30am today :-)) can have a look and perhaps open > (and even close :)) a ticket on this. > > Best, > Dima > > On 22 March 2010 14:16, Minh Nguyen <nguyenmi...@gmail.com> wrote: > > > > > Hi folks, > > > A binary of Sage 4.3.4 for SPARC Solaris 10 on the machine t2.math is > > available at > > >http://sage.math.washington.edu/home/release/sage-4.3.4/sage-4.3.4-t2... > > > Here is a summary of how I produced that binary. > > > (1) I logged into t2.math, took the source tarball of Sage 4.3.4, and > > compiled it under my $HOME. I didn't try to build under my > > /scratch/mvngu/ directory because I anticipated that other users of > > t2.math would want to use that /scratch partition. > > > (2) After a successful build of Sage 4.3.4 on t2.math, I compiled the > > GNU core utilities, i.e. coreutils-7.4, under my $HOME on t2.math. To > > do that, I put the following conditional in my $HOME/.profile: > > > if [ `uname -s` = "SunOS" ]; then > > # Sun linker/assembler > > . /usr/local/gcc-4.4.1-sun-linker/gcc441sun > > fi > > > I then extracted the coreutils tarball to $HOME/usr/src, configured it > > with the options > > > $ ./configure --prefix=/home/mvngu/usr > > > and then build and install with > > > $ make > > $ make install > > > The compilation process used GCC 4.4.1, as sourced in the above > > conditional. The resulting binaries were then installed under > > /home/mvngu/usr/. > > > (3) When coreutils-7.4 had completed building and installing, I > > changed my $PATH to put the path /home/mvngu/usr/bin first. In this > > way, basic commands such as cp, ln, etc. under /home/mvngu/usr/bin are > > picked up first, instead of the corresponding system-wide binaries on > > t2.math. > > > (4) From the SAGE_ROOT of the newly compiled Sage 4.3.4, I wrapped up > > a t2.math binary as follows: > > > $ ./sage -bdist 4.3.4-t2.math.washington.edu > > > After issuing the above command, I noticed heaps of warning/error > > messages along the line of "permission denied". But in the end, I > > still ended up with a t2.math binary wrapped up as a gzip'd tarball. > > > A funny thing about the t2.math binary I posted above is that it is > > much smaller in terms of MB than the corresponding sage.math binary. I > > suspect that something was amiss in my build or binary wrap up > > process. > > > If you need to test something on t2.math, feel free to use the binary > > I posted. Please test and report all problems. > > > -- > > Regards > > Minh Van Nguyen > > > To unsubscribe from this group, send email to > > sage-solaris+unsubscribegooglegroups.com or reply to this email with the > > words "REMOVE ME" as the subject. -- 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 To unsubscribe from this group, send email to sage-devel+unsubscribegooglegroups.com or reply to this email with the words "REMOVE ME" as the subject.
