1 Oct 2003 10:15
1 Oct 2003 10:27
config.h:1: error: parse error before '<<' token error
Nils Wagner <nwagner <at> mecha.uni-stuttgart.de>
2003-10-01 08:27:44 GMT
2003-10-01 08:27:44 GMT
Hi all, python setup.py build from latest cvs failed In file included from /export/home/wagner/cvs/scipy/Lib/xplt/src/play/unix/dir.c:16: /export/home/wagner/cvs/scipy/Lib/xplt/src/play/unix/config.h:1: error: parse error before '<<' token error: command 'gcc' failed with exit status 1 Nils
1 Oct 2003 09:39
Re: config.h:1: error: parse error before '<<' token error
Pearu Peterson <pearu <at> scipy.org>
2003-10-01 07:39:01 GMT
2003-10-01 07:39:01 GMT
On Wed, 1 Oct 2003, Nils Wagner wrote: > python setup.py build from latest cvs failed > > In file included from > /export/home/wagner/cvs/scipy/Lib/xplt/src/play/unix/dir.c:16: > /export/home/wagner/cvs/scipy/Lib/xplt/src/play/unix/config.h:1: error: > parse error before '<<' token > error: command 'gcc' failed with exit status 1 This probably means that there were conflicts when mergeing the changes in CVS and your local config.h. This is is a known issue that we should evenutally fix (generate files only under build/ tree and never commit generated files to CVS). As a workaround, remove scipy/Lib/xplt/src/play/unix/config.h and do `cvs update` again. Pearu
1 Oct 2003 11:06
Re: config.h:1: error: parse error before '<<' tokenerror
Nils Wagner <nwagner <at> mecha.uni-stuttgart.de>
2003-10-01 09:06:03 GMT
2003-10-01 09:06:03 GMT
Pearu Peterson schrieb: > > On Wed, 1 Oct 2003, Nils Wagner wrote: > > > python setup.py build from latest cvs failed > > > > In file included from > > /export/home/wagner/cvs/scipy/Lib/xplt/src/play/unix/dir.c:16: > > /export/home/wagner/cvs/scipy/Lib/xplt/src/play/unix/config.h:1: error: > > parse error before '<<' token > > error: command 'gcc' failed with exit status 1 > > This probably means that there were conflicts when mergeing the changes in > CVS and your local config.h. This is is a known issue that we should > evenutally fix (generate files only under build/ tree and never commit > generated files to CVS). > > As a workaround, remove scipy/Lib/xplt/src/play/unix/config.h and > do `cvs update` again. > > Pearu > Just another question. What is the meaning of ? scipy/_ft.cpp ? scipy/_kiva.cpp ? scipy/Lib/linalg/build ? scipy/Lib/linalg/cblas.pyf ? scipy/Lib/linalg/clapack.pyf(Continue reading)
1 Oct 2003 10:16
Re: config.h:1: error: parse error before '<<' tokenerror
Pearu Peterson <pearu <at> scipy.org>
2003-10-01 08:16:12 GMT
2003-10-01 08:16:12 GMT
On Wed, 1 Oct 2003, Nils Wagner wrote: > Just another question. What is the meaning of > > ? scipy/_ft.cpp > ? scipy/_kiva.cpp > ? scipy/Lib/linalg/build > ? scipy/Lib/linalg/cblas.pyf > ? scipy/Lib/linalg/clapack.pyf > ? scipy/Lib/linalg/fblas.pyf > ? scipy/Lib/linalg/flapack.pyf > ? scipy/Lib/xplt/pygist/Make.cfg > ? scipy/scipy_core/scipy_base/mconf_lite.h > ? scipy/scipy_core/scipy_base/mconf_lite.h > > Is it necessary to remove these files ? No. But they all illustrate a bad habit to generate files under CVS tree. While *.pyf,*.cpp files are platform/compiler independent then other files are not. These other files must be removed whenever you switch a compiler or a platform but using the same CVS tree. Actually, also *.pyf files depend on the ATLAS version and may cause problems when switching ATLAS libraries. `?` indicates that those files are unknown to the CVS repository and this is a good thing. Pearu
1 Oct 2003 11:40
Different failures in scipy.test()
Nils Wagner <nwagner <at> mecha.uni-stuttgart.de>
2003-10-01 09:40:43 GMT
2003-10-01 09:40:43 GMT
Dear experts, I have observed different failures in scipy.test() on different installations (python2.2, python2.3) with different versions of ATLAS. What is the reason for that ? Which versions of ATLAS and python are recommended ? Nils
1 Oct 2003 10:34
Re: Different failures in scipy.test()
Pearu Peterson <pearu <at> scipy.org>
2003-10-01 08:34:42 GMT
2003-10-01 08:34:42 GMT
On Wed, 1 Oct 2003, Nils Wagner wrote: > Dear experts, > > I have observed different failures in scipy.test() on different > installations (python2.2, python2.3) with different versions of ATLAS. > What is the reason for that ? Could you include also the nature of these failures? It is rather difficult to comment with the given (lack of) information. > Which versions of ATLAS and python are recommended ? See INSTALL.txt. In practice, less trouble are with versions that scipy developers useI usually test scipy with the boundary versions of software as given in INSTALL.txt. Pearu
1 Oct 2003 12:43
Improvements for the iterative computation of signm
Nils Wagner <nwagner <at> mecha.uni-stuttgart.de>
2003-10-01 10:43:18 GMT
2003-10-01 10:43:18 GMT
Dear experts,
The current iterative algorithm for signm can be improved.
The theory is given in a recent paper by Bai et. al.
The spectral decomposition of nonsymmetric matrices
on distributed memory parallel computers,
SIAM J. Sci. Comput. Vol. 18 (5) pp. 1446-1461
Any cooments ?
Nils
1 Oct 2003 17:47
Re: Different failures in scipy.test()
Nils Wagner <nwagner <at> mecha.uni-stuttgart.de>
2003-10-01 15:47:35 GMT
2003-10-01 15:47:35 GMT
Pearu Peterson schrieb: > > On Wed, 1 Oct 2003, Nils Wagner wrote: > > > Dear experts, > > > > I have observed different failures in scipy.test() on different > > installations (python2.2, python2.3) with different versions of ATLAS. > > What is the reason for that ? > > Could you include also the nature of these failures? Yes. >> scipy.__version__ '0.2.0__220.4275' >>> Numeric.__version__ '23.0' Python 2.2.2 (#1, Mar 17 2003, 15:17:58) [GCC 3.3 20030226 (prerelease) (SuSE Linux)] on linux2 Type "help", "copyright", "credits" or "license" for more information. BTW, Is it possible to dtermine the version of ATLAS in a similar way ? ====================================================================== FAIL: check_pro_ang1_cv (test_basic.test_cephes) ---------------------------------------------------------------------- Traceback (most recent call last): File "/usr/lib/python2.2/site-packages/scipy/special/tests/test_basic.py", line 357, in check_pro_ang1_cv(Continue reading)
1 Oct 2003 21:30
Re: Different failures in scipy.test()
Pearu Peterson <pearu <at> scipy.org>
2003-10-01 19:30:41 GMT
2003-10-01 19:30:41 GMT
On Wed, 1 Oct 2003, Nils Wagner wrote: > >> scipy.__version__ > '0.2.0__220.4275' > >>> Numeric.__version__ > '23.0' > Python 2.2.2 (#1, Mar 17 2003, 15:17:58) > [GCC 3.3 20030226 (prerelease) (SuSE Linux)] on linux2 > Type "help", "copyright", "credits" or "license" for more information. > > > BTW, Is it possible to dtermine the version of ATLAS in a similar way ? Yes, see INSTALL.txt: section TROUBLESHOOTING, item (6) > ====================================================================== > FAIL: check_pro_ang1_cv (test_basic.test_cephes) > ---------------------------------------------------------------------- > Items are not equal: > DESIRED: (1.0, 0.0) > ACTUAL: (0.99999999999999989, 0.0) This could be related to Python 2.2.2, not sure. On Python 2.3.1 this test succeeds. > ################################################################################################################# > > The same cvs version of scipy > but(Continue reading)
RSS Feed