When matplot lib builds on 't2', it says:
BUILDING MATPLOTLIB
matplotlib: 0.99.1
python: 2.6.4 (r264:75706, Jun 5 2010, 17:43:53) [GCC
4.4.1]
platform: sunos5
REQUIRED DEPENDENCIES
numpy: 1.3.0
freetype2: 9.8.3
That version of freetype is not the one in Sage.
when matplot libs starts building with gcc, I see:
gcc -fno-strict-aliasing -g -O2 -DNDEBUG -g -fwrapv -O3 -Wall
-Wstrict-prototypes -fPIC -DPY_ARRAYAUNIQUE_SYMBOL=MPL_ARRAY_API
-I/rootpool2/local/kirkby/sage-4.4.3/local/lib/python2.6/site-packages/numpy/core/include
-I/usr/sfw/include -I/usr/sfw/include/freetype2 -I/usr/local/include -I.
-I/rootpool2/local/kirkby/sage-4.4.3/local/include/
-I/rootpool2/local/kirkby/sage-4.4.3/local/include/python2.6 -c src/ft2font.cpp
-o build/temp.solaris-2.10-sun4v-2.6/src/ft2font.o
Note how the include path has some directories like -I/usr/sfw/include and
-I/usr/local/include before the Sage directory
-I/rootpool2/local/kirkby/sage-4.4.3/local/include/
Therefore old versions of include files are being included.
There's a warnings given later
In file included from /usr/sfw/include/freetype2/freetype/freetype.h:51,
from src/ft2font.h:14,
from src/ft2font.cpp:1:
/usr/sfw/include/freetype2/freetype/config/ftconfig.h:65:1: warning:
"SIZEOF_LONG" redefined
The problem is, we do not want Sage including those files, but rather those in
Sage.
I don't know exactly what is happening, but it may be that the configuration
file
/usr/sfw/bin/freetype-config
is found before $SAGE_LOCAL/bin/freetype-config, so the locations of the include
directories from there are used.
This all looks wrong to me.
Comments?
Dave
--
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
