Olena-patches September 2006

olena-patches@lrde.epita.fr

3 participants
92 discussions

Re: [Olena-patches] 527: Add image2d and fix topo-related classes.

by Roland Levillain

Roland Levillain <roland(a)lrde.epita.fr> writes: > Thierry GERAUD <theo(a)lrde.epita.fr> writes: > >> Index: oln/core/abstract/image.hh >> =================================================================== >> --- oln/core/abstract/image.hh (revision 526) >> +++ oln/core/abstract/image.hh (working copy) >> @@ -29,216 +29,11 @@ > > [...] > >> + /// Typedefs. >> + typedef oln_type_of(E, topo) topo_t; >> + typedef oln_type_of(E, psite) psite_t; >> + typedef oln_type_of(E, rvalue) rvalue_t; > > Shouldn't we use type names ending in `_type', rather than `_t', in > accordance with > > https://olena.lrde.epita.fr/cgi-bin/twiki/view/Olena/CodingStyle > > ? We've discussed this point with Théo; he said names ending in `_type' are for public types, while the ones ending in `_t' are for private types. That makes sense to me, so I suggest sticking to this convention. (BTW, we have to create a new coding style for Olena 1.0 on the Wiki, derived from Olena 0.10's.)

18 years, 2 months

oln-0.10 10.230: Work around a bug in SWIG 1.3.29

by Roland Levillain

I haven't checked whether this bug has been fixed in SWIG's Subversion repository yet (but it wasn't present in SWIG 1.3.28). 2006-07-04 Roland Levillain <roland(a)lrde.epita.fr> Work around a bug in SWIG 1.3.29. * generate_arith_instantiations.py (write_algorithms) * generate_morpho_instantiations.py (write_algorithms): Work around a bug in SWIG 1.3.29 in the generation of C++ wrappers. Index: 10.229/tools/swilena/generate_morpho_instantiations.py --- 10.229/tools/swilena/generate_morpho_instantiations.py Fri, 30 Jun 2006 17:26:15 +0200 levill_r (oln/v/24_generate_m 1.8.1.1 700) +++ 10.229(w)/tools/swilena/generate_morpho_instantiations.py Mon, 03 Jul 2006 12:23:42 +0200 levill_r (oln/v/24_generate_m 1.8.1.1 700) @@ -57,7 +57,13 @@ "ntg_int_u8", "ntg_int_u32", "ntg_int_s8", "ntg_int_s32", "ntg_float" ]: - img_type = "::oln::image%(dim)sd< %(type)s >" % vars() + # FIXME: SWIG 1.3.29 Bug. We used to refer to `oln::image' from + # the global (top-level) namespace, i.e. `::oln::image', + # but it makes swig 1.3.29 generate invalid C++ code when + # used ad a template argument: the space between `<' and + # `::' is eaten by swig, and `<:' is understood as a + # trigraph (for `[') by the C++ compiler. + img_type = "oln::image%(dim)sd< %(type)s >" % vars() win_type = "::oln::window%(dim)sd" % vars() neighb_type = "::oln::neighborhood%(dim)sd" % vars() @@ -135,10 +141,12 @@ instantiate(idx, "top_hat_contrast_op", img_type, img_type, win_type) instantiate(idx, "fast_top_hat_contrast_op", img_type, img_type, win_type) # Watershed - img_ret_type = "::oln::image%(dim)sd< ntg_int_u32 >" % vars() + # FIXME: SWIG 1.3.29 Bug (same as above). + img_ret_type = "oln::image%(dim)sd< ntg_int_u32 >" % vars() instantiate(idx, "watershed_seg", img_ret_type, img_type, neighb_type) instantiate(idx, "watershed_con", img_ret_type, img_type, neighb_type) - bin_img_type = "::oln::image%(dim)sd< ntg_bin >" % vars() + # FIXME: SWIG 1.3.29 Bug (same as above). + bin_img_type = "oln::image%(dim)sd< ntg_bin >" % vars() instantiate(idx, "sure_minima_imposition", img_type, img_type, bin_img_type, neighb_type) instantiate(idx, "sequential_minima_imposition", img_type, img_type, bin_img_type, neighb_type) instantiate(idx, "hybrid_minima_imposition", img_type, img_type, bin_img_type, neighb_type) Index: 10.229/tools/swilena/generate_arith_instantiations.py --- 10.229/tools/swilena/generate_arith_instantiations.py Fri, 30 Jun 2006 17:26:15 +0200 levill_r (oln/v/51_generate_a 1.1.1.1 700) +++ 10.229(w)/tools/swilena/generate_arith_instantiations.py Mon, 03 Jul 2006 11:44:33 +0200 levill_r (oln/v/51_generate_a 1.1.1.1 700) @@ -59,13 +59,19 @@ "ntg_int_u8", "ntg_int_u32", "ntg_int_s8", "ntg_int_s32", "ntg_float" ]: - img_type = "::oln::image%(dim)sd< %(type)s >" % vars() + # FIXME: SWIG 1.3.29 Bug. We used to refer to `oln::image' from + # the global (top-level) namespace, i.e. `::oln::image', + # but it makes swig 1.3.29 generate invalid C++ code when + # used ad a template argument: the space between `<' and + # `::' is eaten by swig, and `<:' is understood as a + # trigraph (for `[') by the C++ compiler. + img_type = "oln::image%(dim)sd< %(type)s >" % vars() bigger_type = get_bigger_type(type) - return_img_type = "::oln::image%(dim)sd< %(bigger_type)s >" % vars() + return_img_type = "oln::image%(dim)sd< %(bigger_type)s >" % vars() # FIXME: these algorithms do not work with bin if type != "ntg_bin": - for algo in [ "plus", "minus", "times", "div" ]: + for algo in [ "plus", "minus", "times", "div" ]: instantiate(dim, algo, return_img_type, img_type, img_type) instantiate(dim, "min", return_img_type, img_type, img_type)

18 years, 2 months

596: Use TESTS_CXXFLAGS in Olena.

by Roland Levillain

https://svn.lrde.epita.fr/svn/oln/trunk/olena Index: ChangeLog from Roland Levillain <roland(a)lrde.epita.fr> Use TESTS_CXXFLAGS in Olena. * tests/Makefile.am (TESTS_CXXFLAGS): New. Append it to... (CXXFLAGS): ...this variable. * tests/fill.cc: Typo in comment. Makefile.am | 6 ++++-- fill.cc | 2 +- 2 files changed, 5 insertions(+), 3 deletions(-) Index: tests/Makefile.am --- tests/Makefile.am (revision 595) +++ tests/Makefile.am (working copy) @@ -9,8 +9,10 @@ # AM_CXXFLAGS = $(CXXFLAGS_STRICT) $(CXXFLAGS_DEBUG) # # when oln.m4 is available in the distribution. -# Meanwhile, alter CXXFLAGS to turn off any optimization. -CXXFLAGS = -O0 -ggdb +# Meanwhile, alter CXXFLAGS to turn off any optimization (Automake will +# warn against this, but that's OK). +TESTS_CXXFLAGS = @TESTS_CXXFLAGS@ +CXXFLAGS += $(TESTS_CXXFLAGS) check_PROGRAMS = \ Index: tests/fill.cc --- tests/fill.cc (revision 595) +++ tests/fill.cc (working copy) @@ -29,7 +29,7 @@ #include <cassert> // FIXME: We should not include oln/basics2d.hh, but -// oln/core/2d/image2d.hh (and oln/core/2d/neigh2d.hh ?). +// just oln/core/2d/image2d.hh. #include <oln/basics2d.hh> #include <oln/level/fill.hh>

18 years, 2 months

595: Use TESTS_CXXFLAGS in Static.

by Roland Levillain

https://svn.lrde.epita.fr/svn/oln/trunk/static Index: ChangeLog from Roland Levillain <roland(a)lrde.epita.fr> Use TESTS_CXXFLAGS in Static. * tests/Makefile.am (TESTS_CXXFLAGS): New. Append it to... (CXXFLAGS): ...this variable. * tests/tour.cc (dynamic_hierarchy::A, dynamic_hierarchy::B) (dynamic_hierarchy::C): Add missing virtual destructors. Makefile.am | 3 ++- tour.cc | 6 ++++++ 2 files changed, 8 insertions(+), 1 deletion(-) Index: tests/Makefile.am --- tests/Makefile.am (revision 594) +++ tests/Makefile.am (working copy) @@ -9,7 +9,8 @@ # when oln.m4 is available in the distribution. # Meanwhile, alter CXXFLAGS to turn off any optimization (Automake will # warn against this, but that's OK). -CXXFLAGS = -O0 -ggdb +TESTS_CXXFLAGS = @TESTS_CXXFLAGS@ +CXXFLAGS += $(TESTS_CXXFLAGS) check_PROGRAMS = \ Index: tests/tour.cc --- tests/tour.cc (revision 594) +++ tests/tour.cc (working copy) @@ -79,6 +79,8 @@ virtual int foo() { return 1; } // A virtual pure method. virtual int bar() = 0; + + virtual ~A() {} }; // `B' is a concrete class. @@ -88,6 +90,8 @@ virtual int foo() { return 2; } // `B::bar' is defined. virtual int bar() { return 3; } + + virtual ~B() {} }; // `C' is a concrete class. @@ -95,6 +99,8 @@ { // `B::foo' is redefined. virtual int foo() { return 4; } + + virtual ~C() {} }; } // end of namespace dynamic_hierarchy

18 years, 2 months

594: Use TESTS_CXXFLAGS in Extended.

by Roland Levillain

https://svn.lrde.epita.fr/svn/oln/trunk/extended Index: ChangeLog from Roland Levillain <roland(a)lrde.epita.fr> Use TESTS_CXXFLAGS in Extended. * tests/Makefile.am (TESTS_CXXFLAGS): New. Append it to... (CXXFLAGS): ...this variable. * tests/optraits.cc (main): Make dummy use of all variables to avoid compiler warnings. Makefile.am | 8 ++++++-- optraits.cc | 2 ++ 2 files changed, 8 insertions(+), 2 deletions(-) Index: tests/Makefile.am --- tests/Makefile.am (revision 593) +++ tests/Makefile.am (working copy) @@ -1,12 +1,16 @@ ## Process this file through Automake to create Makefile.in -*- Makefile -*- AM_CPPFLAGS = -I$(top_srcdir)/extended -I$(top_srcdir)/metalic + # FIXME: Add # -# AM_CXXFLAGS = $(CXXFLAGS_STRICT) -ggdb +# AM_CXXFLAGS = $(CXXFLAGS_STRICT) $(CXXFLAGS_DEBUG) # # when oln.m4 is available in the distribution. -AM_CXXFLAGS = -O0 -ggdb +# Meanwhile, alter CXXFLAGS to turn off any optimization (Automake will +# warn against this, but that's OK). +TESTS_CXXFLAGS = @TESTS_CXXFLAGS@ +CXXFLAGS += $(TESTS_CXXFLAGS) check_PROGRAMS = \ Index: tests/optraits.cc --- tests/optraits.cc (revision 593) +++ tests/optraits.cc (working copy) @@ -55,6 +55,8 @@ typeof(1+2u) tmp; // 2 * cos(_1); + // Dummy usage. + tmp = tmp; test_plus(2 * cos(_1), sin(_1)); }

18 years, 2 months

593: Use TESTS_CXXFLAGS in Metalic.

by Roland Levillain

https://svn.lrde.epita.fr/svn/oln/trunk/metalic Index: ChangeLog from Roland Levillain <roland(a)lrde.epita.fr> Use TESTS_CXXFLAGS in Metalic. * tests/gcase.cc (print), tests/protected.cc (reveal), * tests/switch.cc (print): Make dummy use of all variables to void compiler warnings. * tests/Makefile.am (TESTS_CXXFLAGS): New. Append it to... (CXXFLAGS): ...this variable. Makefile.am | 8 ++++++-- gcase.cc | 2 ++ protected.cc | 2 ++ switch.cc | 2 ++ 4 files changed, 12 insertions(+), 2 deletions(-) Index: tests/switch.cc --- tests/switch.cc (revision 592) +++ tests/switch.cc (working copy) @@ -157,6 +157,8 @@ void print() { int tmp; + // Dummy usage. + tmp = tmp; } Index: tests/gcase.cc --- tests/gcase.cc (revision 592) +++ tests/gcase.cc (working copy) @@ -55,6 +55,8 @@ void print() { int tmp; + // Dummy usage. + tmp = tmp; } Index: tests/Makefile.am --- tests/Makefile.am (revision 592) +++ tests/Makefile.am (working copy) @@ -1,12 +1,16 @@ ## Process this file through Automake to create Makefile.in -*- Makefile -*- AM_CPPFLAGS = -I$(top_srcdir)/metalic + # FIXME: Add # -# AM_CXXFLAGS = $(CXXFLAGS_STRICT) -ggdb +# AM_CXXFLAGS = $(CXXFLAGS_STRICT) $(CXXFLAGS_DEBUG) # # when oln.m4 is available in the distribution. -AM_CXXFLAGS = -O0 -ggdb +# Meanwhile, alter CXXFLAGS to turn off any optimization (Automake will +# warn against this, but that's OK). +TESTS_CXXFLAGS = @TESTS_CXXFLAGS@ +CXXFLAGS += $(TESTS_CXXFLAGS) check_PROGRAMS = \ Index: tests/protected.cc --- tests/protected.cc (revision 592) +++ tests/protected.cc (working copy) @@ -39,6 +39,8 @@ void reveal() { T tmp; + // Dummy usage. + tmp = tmp; }

18 years, 2 months

592: Create a configure variable for C++ compiler flags used for the tests.

by Roland Levillain

https://svn.lrde.epita.fr/svn/oln/trunk Index: ChangeLog from Roland Levillain <roland(a)lrde.epita.fr> Create a configure variable for C++ compiler flags used for the tests. * configure.ac (TESTS_CXXFLAGS): New variable. configure.ac | 7 +++++++ 1 file changed, 7 insertions(+) Index: configure.ac --- configure.ac (revision 591) +++ configure.ac (working copy) @@ -32,6 +32,13 @@ CXXFLAGS="$CXXFLAGS -pipe" fi +# C++ compiler flags for tests. +AC_ARG_VAR([TESTS_CXXFLAGS]) +# We want no optimization for the tests (too slow), and a lot of debugging. +if test "$GXX" = yes && test -z "$TESTS_CXXFLAGS"; then + TESTS_CXXFLAGS="-O0 -ggdb -Wall -pipe" +fi + ## ------- ## ## Tests. ##

18 years, 2 months

591: Work around a bug in g++ 4.1 and 4.2 in tests/vtypes-and-exact.cc.

by Roland Levillain

https://svn.lrde.epita.fr/svn/oln/trunk/static Index: ChangeLog from Roland Levillain <roland(a)lrde.epita.fr> Work around a bug in g++ 4.1 and 4.2 in tests/vtypes-and-exact.cc. * tests/vtypes-and-exact.cc (my_direct_type_of), (my_direct_type_of_): New macros. (my::A::foo_type, my::A::bar_type, my::A::baz_type) (my::B::foo_type, my::B::bar_type, my::B::baz_type) (my::B::quux_type, my::B::yin_type) (my::C::foo_type, my::C::quux_type, my::C::zorg_type): Rename typedefs as... (my::A::foo_t, my::A::bar_t, my::A::baz_t) (my::B::foo_t, my::B::bar_t, my::B::baz_t) (my::B::quux_t, my::B::yin_t) (my::C::foo_t, my::C::quux_t, my::C::zorg_t): ...this. Use my_direct_type_of and my_direct_type_of_ (on the exact type) instead of my_type_of and my_type_of_ (on the ``self'', current type) to work around a bug in g++ 4.1 and 4.2. (main): Adjust. * tests/Makefile.am (AM_CXXFLAGS): Rename as... (CXXFLAGS): ...this. Makefile.am | 7 ++- vtypes-and-exact.cc | 103 ++++++++++++++++++++++++++++++++++++++++------------ 2 files changed, 85 insertions(+), 25 deletions(-) Index: tests/vtypes-and-exact.cc --- tests/vtypes-and-exact.cc (revision 590) +++ tests/vtypes-and-exact.cc (working copy) @@ -44,6 +44,12 @@ #define my_type_of_(FromType, Typedef) \ stc_type_of_(my, my::category::my_cat, FromType, Typedef) +#define my_direct_type_of(FromExactType, Typedef) \ + typename my_direct_type_of_(FromExactType, Typedef) + +#define my_direct_type_of_(FromExactType, Typedef) \ + stc_direct_type_of_(my, my::category::my_cat, FromExactType, Typedef) + // Namespace equipment. stc_scoop_equipment_for_namespace(my); @@ -98,10 +104,33 @@ template <typename Exact> struct A : public stc::any<Exact> { + typedef A<Exact> self_t; + typedef Exact exact_t; + // Aliases. - typedef my_type_of(A, foo) foo_type; - typedef my_type_of(A, bar) bar_type; - typedef my_type_of(A, baz) baz_type; + + /* FIXME: Work around a bug that affects both g++ 4.1 and + gcc-snapshot (4.2) from Debian. The versions of the compiler + used at the moment this text was written were: + + g++ 4.1 : g++ (GCC) 4.1.2 20060920 (prerelease) (Debian 4.1.1-14) + gcc-snapshot : g++ (GCC) 4.2.0 20060922 (experimental) + + + Problem description: + + Using my_type_of (i.e., using stc_to_exact) breaks the + assertions ``Ensure stc::is_any_ works properly.'' below (in + main). Using my_direct_type_of (on the exact type) solves + this! */ +#if 0 + typedef my_type_of(self_t, foo) foo_t; + typedef my_type_of(self_t, bar) bar_t; + typedef my_type_of(self_t, baz) baz_t; +#endif + typedef my_direct_type_of(exact_t, foo) foo_t; + typedef my_direct_type_of(exact_t, bar) bar_t; + typedef my_direct_type_of(exact_t, baz) baz_t; }; @@ -144,12 +173,27 @@ template <typename Exact> struct B : public stc_get_supers(B<Exact>) { + typedef B<Exact> self_t; + typedef Exact exact_t; + // Aliases. - typedef my_type_of(B, foo) foo_type; - typedef my_type_of(B, bar) bar_type; - typedef my_type_of(B, baz) baz_type; - typedef my_type_of(B, quux) quux_type; - typedef my_type_of(B, yin) yin_type; + + /* FIXME: Same as above; using my_type_of (i.e., using + stc_to_exact) breaks the assertions ``Ensure stc::is_any_ works + properly.'' below (in main). Using my_direct_type_of (on the + exact type) solves this! */ +#if 0 + typedef my_type_of(self_t, foo) foo_t; + typedef my_type_of(self_t, bar) bar_t; + typedef my_type_of(self_t, baz) baz_t; + typedef my_type_of(self_t, quux) quux_t; + typedef my_type_of(self_t, yin) yin_t; +#endif + typedef my_direct_type_of(exact_t, foo) foo_t; + typedef my_direct_type_of(exact_t, bar) bar_t; + typedef my_direct_type_of(exact_t, baz) baz_t; + typedef my_direct_type_of(exact_t, quux) quux_t; + typedef my_direct_type_of(exact_t, yin) yin_t; }; @@ -173,10 +217,23 @@ struct C : public stc_get_supers(C) { + typedef C self_t; + typedef self_t exact_t; + // Aliases. - typedef my_type_of_(C, foo) foo_type; - typedef my_type_of_(C, quux) quux_type; - typedef my_type_of_(C, zorg) zorg_type; + + /* FIXME: Same as above; using my_type_of_ (i.e., using + stc_to_exact) breaks the assertions ``Ensure stc::is_any_ works + properly.'' below (in main). Using my_direct_type_of_ (on the + exact type) solves this! */ +#if 0 + typedef my_type_of_(self_t, foo) foo_t; + typedef my_type_of_(self_t, quux) quux_t; + typedef my_type_of_(self_t, zorg) zorg_t; +#endif + typedef my_direct_type_of_(exact_t, foo) foo_t; + typedef my_direct_type_of_(exact_t, quux) quux_t; + typedef my_direct_type_of_(exact_t, zorg) zorg_t; }; } // end of namespace my @@ -204,21 +261,21 @@ // Check types associated to A<C>. - mlc::assert_<mlc_eq(A<C>::foo_type, int)>::check(); - mlc::assert_<mlc_eq(A<C>::bar_type, double)>::check(); + mlc::assert_<mlc_eq(A<C>::foo_t, int)>::check(); + mlc::assert_<mlc_eq(A<C>::bar_t, double)>::check(); // Check types associated to B<C>. - mlc::assert_<mlc_eq(B<C>::baz_type, char)>::check(); - mlc::assert_<mlc_eq(B<C>::quux_type, long)>::check(); - mlc::assert_<mlc_eq(B<C>::yin_type, unsigned long)>::check(); + mlc::assert_<mlc_eq(B<C>::baz_t, char)>::check(); + mlc::assert_<mlc_eq(B<C>::quux_t, long)>::check(); + mlc::assert_<mlc_eq(B<C>::yin_t, unsigned long)>::check(); - mlc::assert_<mlc_eq(B<C>::bar_type, A<C>::bar_type)>::check(); + mlc::assert_<mlc_eq(B<C>::bar_t, A<C>::bar_t)>::check(); // Check types associated to C. - mlc::assert_<mlc_eq(C::foo_type, int)>::check(); - mlc::assert_<mlc_eq(C::bar_type, double)>::check(); - mlc::assert_<mlc_eq(C::baz_type, char)>::check(); - mlc::assert_<mlc_eq(C::quux_type, long)>::check(); - mlc::assert_<mlc_eq(C::yin_type, unsigned long)>::check(); - mlc::assert_<mlc_eq(C::zorg_type, double)>::check(); + mlc::assert_<mlc_eq(C::foo_t, int)>::check(); + mlc::assert_<mlc_eq(C::bar_t, double)>::check(); + mlc::assert_<mlc_eq(C::baz_t, char)>::check(); + mlc::assert_<mlc_eq(C::quux_t, long)>::check(); + mlc::assert_<mlc_eq(C::yin_t, unsigned long)>::check(); + mlc::assert_<mlc_eq(C::zorg_t, double)>::check(); } Index: tests/Makefile.am --- tests/Makefile.am (revision 590) +++ tests/Makefile.am (working copy) @@ -1,12 +1,15 @@ ## Process this file through Automake to create Makefile.in -*- Makefile -*- AM_CPPFLAGS = -I$(top_srcdir)/static -I$(top_srcdir)/metalic + # FIXME: Add # -# AM_CXXFLAGS = $(CXXFLAGS_STRICT) -ggdb +# AM_CXXFLAGS = $(CXXFLAGS_STRICT) $(CXXFLAGS_DEBUG) # # when oln.m4 is available in the distribution. -AM_CXXFLAGS = -O0 -ggdb +# Meanwhile, alter CXXFLAGS to turn off any optimization (Automake will +# warn against this, but that's OK). +CXXFLAGS = -O0 -ggdb check_PROGRAMS = \

18 years, 2 months

590: xtd_is_binary use bexpr_.

by Thierry GERAUD

2006-09-29 Thierry GERAUD <theo(a)tegucigalpa.lrde.epita.fr> * xtd/valtraits.hh: Use bexpr_. Index: xtd/valtraits.hh =================================================================== --- xtd/valtraits.hh (revision 589) +++ xtd/valtraits.hh (working copy) @@ -28,6 +28,7 @@ #ifndef XTD_VALTRAITS_HH # define XTD_VALTRAITS_HH +# include <mlc/bexpr.hh> // # include <xtd/traits.hh> // # include <xtd/builtin/traits.hh> @@ -37,16 +38,14 @@ { template <typename T> - struct is_binary + struct is_binary : public mlc::bexpr_<false> { - typedef mlc::false_ ret; }; template <> - struct is_binary <bool> + struct is_binary <bool> : public mlc::bexpr_<true> { - typedef mlc::true_ ret; }; @@ -54,7 +53,7 @@ -# define xtd_is_binary(T) typename xtd::is_binary<T>::ret +# define xtd_is_binary(T) xtd::is_binary<T>

18 years, 2 months

589: Add a temporary simple version of xtd_is_binary for use in oln.

by Thierry GERAUD

2006-09-29 Thierry GERAUD <theo(a)tegucigalpa.lrde.epita.fr> Add a temporary simple version of xtd_is_binary for use in oln. * xtd/valtraits.hh: New. * xtd/Makefile.am (nobase_xtd_HEADERS): Update. Index: xtd/valtraits.hh =================================================================== --- xtd/valtraits.hh (revision 0) +++ xtd/valtraits.hh (revision 0) @@ -0,0 +1,61 @@ +// Copyright (C) 2006 EPITA Research and Development Laboratory +// +// This file is part of the Olena Library. This library is free +// software; you can redistribute it and/or modify it under the terms +// of the GNU General Public License version 2 as published by the +// Free Software Foundation. +// +// This library is distributed in the hope that it will be useful, +// but WITHOUT ANY WARRANTY; without even the implied warranty of +// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU +// General Public License for more details. +// +// You should have received a copy of the GNU General Public License +// along with this library; see the file COPYING. If not, write to +// the Free Software Foundation, 51 Franklin Street, Fifth Floor, +// Boston, MA 02111-1307, USA. +// +// As a special exception, you may use this file as part of a free +// software library without restriction. Specifically, if other files +// instantiate templates or use macros or inline functions from this +// file, or you compile this file and link it with other files to +// produce an executable, this file does not by itself cause the +// resulting executable to be covered by the GNU General Public +// License. This exception does not however invalidate any other +// reasons why the executable file might be covered by the GNU General +// Public License. + +#ifndef XTD_VALTRAITS_HH +# define XTD_VALTRAITS_HH + +// # include <xtd/traits.hh> +// # include <xtd/builtin/traits.hh> + + + +namespace xtd +{ + + template <typename T> + struct is_binary + { + typedef mlc::false_ ret; + }; + + + template <> + struct is_binary <bool> + { + typedef mlc::true_ ret; + }; + + +} // end of namespace xtd + + + +# define xtd_is_binary(T) typename xtd::is_binary<T>::ret + + + +#endif // ! XTD_VALTRAITS_HH Index: xtd/Makefile.am =================================================================== --- xtd/Makefile.am (revision 588) +++ xtd/Makefile.am (working copy) @@ -56,4 +56,5 @@ optraits.hh \ res.hh \ traits.hh \ + valtraits.hh \ vec.hh

18 years, 2 months

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

Olena-patches September 2006