Hi Stewart
any reason to use your own suitesparse libs instead of the
cygwin ones ?
$ cygcheck -c -d |grep Suite
libSuiteSparse-devel 3.5.0-1
SuiteSparse 3.5.0-
...
make[3]: Entering directory `/usr/local/src/octave-3.3.54/src'
...
making oct-conf.h from oct-conf.h.in
oct-conf.h is unchanged
/bin/sh ../libtool --tag=CXX --mode=compile g++ -DHAVE_CONFIG_H -I.
...
c' || echo './'`DLD-FUNCTIONS/fltk_backend.cc
libtool: compile: g++ -DHAVE_CONFIG_H -I. -I.. -I../libgnu -I../libgnu
...
D_FUNCTIONS_fltk_backend_la-fltk_backend.Tpo -c
DLD-FUNCTIONS/fltk_backend.cc -DDLL_EXPORT -
DPIC -o DLD-FUNCTIONS/.libs/DLD_FUNCTIONS_fltk_backend_la-fltk_backend.o
In file included from ../liboctave/lo-mappers.h:30,
from ../liboctave/oct-inttypes.h:35,
from ../liboctave/idx-vector.h:35,
from ../liboctave/Array.h:37,
from ../liboctave/str-vec.h:32,
from ../liboctave/cmd-edit.h:32,
from DLD-FUNCTIONS/fltk_backend.cc:57:
../liboctave/oct-cmplx.h:30: error: expected unqualified-id before
numeric constant
On the disti Suitesparse there is a specific complex patch
--- origsrc/SuiteSparse/CXSparse/Include/cs.h 2009-11-07
18:23:34.000000000 +0 100
+++ src/SuiteSparse/CXSparse/Include/cs.h 2010-08-22
09:42:01.515625000 +0 200
@@ -17,7 +17,11 @@ typedef std::complex<double> cs_complex_
extern "C" {
#else
#ifndef NCOMPLEX
+#ifdef __CYGWIN__
+#define I 1.0I
+#else
#include<complex.h>
+#endif
#define cs_complex_t double _Complex
#endif
#endif
I have no particular problem to build latest octave dev versions ,
using all the packages already available on cygwin.
As reported on the just updated README.Cygwin,
I am building with
configure --enable-shared \
--enable-float-truncate \
CC=gcc-4 F77=gfortran-4 CXX=g++-4 CPP=cpp-4
lt_cv_deplibs_check_method=pass_all \
LDFLAGS=-no-undefined
Marco
_______________________________________________
Help-octave mailing list
address@hidden
https://mailman.cae.wisc.edu/listinfo/help-octave