El jueves, 18 de abril de 2013 22:11:56 UTC-5, kcrisman escribió:
> We have much more recent binaries for PPC machines on 10.4, but you are
> right that we don't have a buildbot for making binaries on a 10.4 Intel box.
>
>
Actually, you should be able to build Sage from source if you have gcc, as
> you appear to, to get a newer Sage!
>
I should probably try this. I will attempt to build from source if I can't
get biopython to work, though I prefer the convenience of a binary (and I
am a bit nervous about attempting to build something as big as sage from
source).
> What happens if you try the same thing for biopython in your Mac-provided
> Python install?
>
My Mac-provided Python is version 2.3, which is not supported by Biopython.
However, I also have a Python 2.7 installation from python.org in my
system. Building and installing Biopython there works, but I don't know if
this is directly comparable, since one of the dependencies of biopython is
numpy, which I don't have in that installation. Unfortunately, I am
currently unable to test installing in a Python installation with numpy on
it (I ran into trouble when trying to install it to the python.org
installation).
Biopython complains about numpy not being there, but allows the
installation with a warning message that some features might not work.
However, the part which fails in sage's python ('building 'Bio.cpairwise2'
extension') seems to work fine in the python.org installation.
In case it helps, this is that part of the output (I didn't copy the whole
thing since it is huge):
building 'Bio.cpairwise2' extension
creating build/temp.macosx-10.3-fat-2.7
creating build/temp.macosx-10.3-fat-2.7/Bio
gcc-4.0 -fno-strict-aliasing -fno-common -dynamic -isysroot
/Developer/SDKs/MacOSX10.4u.sdk -arch ppc -arch i386 -g -O2 -DNDEBUG -g -O3
-IBio -I/Library/Frameworks/Python.framework/Versions/2.7/include/python2.7
-c Bio/cpairwise2module.c -o
build/temp.macosx-10.3-fat-2.7/Bio/cpairwise2module.o
gcc-4.0 -bundle -undefined dynamic_lookup -arch ppc -arch i386 -isysroot
/Developer/SDKs/MacOSX10.4u.sdk -isysroot /Developer/SDKs/MacOSX10.4u.sdk
-g build/temp.macosx-10.3-fat-2.7/Bio/cpairwise2module.o -o
build/lib.macosx-10.3-fat-2.7/Bio/cpairwise2.so
One difference I saw is that the install process is using a different gcc
('gcc' in sage vs 'gcc-4.0' in the python.org installation). I tried
symlinking gcc-4.0 to gcc in a directory in my path to try to get sage's
python to use that version, but I still get the same output (and I'm not
sure if I actually got sage to use gcc-4.0 or not).
Does http://forum.tinycorelinux.net/index.php?topic=13102.0 help? (I
> assume not.)
>
This doesn't seem to help. I do have the limits.h file in the correct
directory, but, as far as I can see, gcc just does not find it for some
reason.
Thanks for your help.
--
You received this message because you are subscribed to the Google Groups
"sage-support" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to sage-support+unsubscr...@googlegroups.com.
To post to this group, send email to sage-support@googlegroups.com.
Visit this group at http://groups.google.com/group/sage-support?hl=en.
For more options, visit https://groups.google.com/groups/opt_out.
