May 2005 - Boost - lists.boost.org

Iterator adaptors <-> boost::bind interaction
by Aleksey Gurtovoy 10 Jun '05

10 Jun '05

The following code currently fails to compile because of a subtle interaction between bind-produced function objects and the current implementation of transform_iterator's 'dereference' function (the same applies to 'filter_iterator' and most likely the rest of the adaptors in the library): #include <boost/iterator/transform_iterator.hpp> #include <boost/bind.hpp> int add_two( int x ) { return x + 2; } int main() { int x[] = { 0 }; *boost::make_transform_iterator( boost::make_transform_iterator( x, boost::bind( &add_two, _1 ) ) , boost::bind( &add_two, _1 ) ); } The problem here is that the result of 'boost::bind( &add_two, _1 )' takes its argument by reference, while transform_iterator's 'dereference' implementation is: typename super_t::reference dereference() const { return m_f(*this->base()); } ^^^^^^^^^^^^^ In our case, '*this->base()' produces a temporary 'int' which cannot be directly passed to 'm_f'. Effectively, the issue prevents you from layering any iterator adaptors built with 'boost::bind' on top of 'transform_iterator', which, needless to say, is BAD. My sugesstion would be to replace the above with typename super_t::reference dereference() const { typename iterator_reference<Iterator>::type x( *this->base() ); return m_f( x ); } Comments? -- Aleksey Gurtovoy MetaCommunications Engineering

2 13

[mscv config]
by Thorsten Ottosen 06 Jun '05

06 Jun '05

Hi, Is there any good reason that VC++ is set to warn about ADL by default? Otherwise, we should put the equivalent of <toolset>msvc <cxxflags>/wd4675 into the main configuration for those compilers -Thorsten -- Thorsten Ottosen ---------------------------- www.dezide.com http://www.cs.aau.dk/index2.php?content=Research/mi www.boost.org www.open-std.org/JTC1/SC22/WG21/

3 5

compressed_pair requirements?
by David Abrahams 06 Jun '05

06 Jun '05

Can arguments to compressed_pair be const-qualified? The documentation only lists this restriction: Note that compressed_pair can not be instantiated if either of the template arguments is a union type, unless there is compiler support for boost::is_union, or if boost::is_union is specialised for the union type. I believe I am getting an error from Intel C++ when an argument to compressed_pair is const. The question is whether we need a workaround in compressed_pair.hpp or a change to the html documentation. -- Dave Abrahams Boost Consulting www.boost-consulting.com

6 22

[assign][range][ptr_container] using-induced problem
by Joaquín Mª López Muñoz 06 Jun '05

06 Jun '05

Martin Wille posted about this a couple of days ago (http://lists.boost.org/boost-testing/2005/05/1216.php) but I'm afraid it could have passed unnoticed. Some tests from Boost.(Assign|Range|PointerContainer) have the following layout #include<...> using namespace std; using namespace boost; ... #include <boost/test/unit_test.hpp> having the using directives before boost/test/unit_test.hpp is AFAICS illegal and does indeed cause problems with GCC 3.4.x and GCC 4.x, see for instance http://tinyurl.com/8payc I cannot test it, but I guess it suffices to move all the includes before any using directive. This will probably clear up dozens of tests, including some toolsets that are currently marked as unusable. Joaquín M López Muñoz Telefónica, Investigación y Desarrollo

4 10

Boost Array Initialization Technique
by christopher diggins 06 Jun '05

06 Jun '05

The fact that the Boost array class does not support constructors in favour of the limited and somewhat obscure aggregate initialization syntax has me concerned. IMO it would be much more flexible and powerful to support overloading of the comma operator. The following works on Visual C++ template<typename array_T, unsigned int N> struct array_comma_helper { array_comma_helper(array_T& x) : m(x) { } array_comma_helper(const array_comma_helper& x) : m(x.m) { } array_comma_helper& operator=(typename array_T::value_type x) { m[0] = x; return *this; } mutable array_T& m; }; template<class array_T, unsigned int N, class Scalar_T> array_comma_helper<array_T, N+1> operator,(array_comma_helper<array_T, N> h, const Scalar_T& x) { array_comma_helper<array_T, N+1> ret(h.m); ret.m[N + 1] = x; return ret; } template<class array_T> array_comma_helper<array_T, 0> initialize(array_T& x) { return array_comma_helper<array_T, 0>(x); } This enables us to write: int n = 2; array<int, 3> a; initialize(a) = 1, n, 3; This is much more flexible and it allows us to have constructors in boost::array thus making it a full reversible container. Christopher Diggins http://www.cdiggins.com

15 36

Math constants - updated.
by Paul A Bristow 06 Jun '05

06 Jun '05

I have recently posted to the Boost Sandbox vault an update of my collection of mathematical constants, previously reviewed as a Boost library and ultimately not accepted because of lack of agreement on how to present the constants, especially for User Defined Types, and an outbreak of macrophobia. However the values are still valid and some may find them a useful and convenient source, with even more reassuringly sufficient decimal places to give you a warm feeling ;-) The values are take from Internet sources and then computed with NTL a arbitrary (very high) precision library and presented as 40 decimal digit strings, suitable for all hardware as yet envisioned. I have tried to check as many as possible against two source to try to avoid mistakes, but please tell me of any that you find. The same values are actually presented in header files as macros, floats, doubles, and long doubles, (and also as the functions allowing pi to call a function pi() proposed by Michael Kenniston, even though I am not entirely convinced of their usefulness). So including double_constants.hpp will give lots of values, for example: static const double cubeRootTwo = 1.25992104989487316476721060727822835057; // 2^1/3 If you prefer a macro, the define_constants.h will give you values like #define M_EPOWPI 23.14069263277926900572908636794854738027L /* e^pi */ You can of course also cut and paste invididual lines, if you just want pi. No choice of names will suit everyone - so I have indulged myself. You can change them of course. http://tinyurl.com/7p6es for two zips of headers and documentation as html. Enjoy? Paul PS Some of these values are to be used in the math functions for stats library that I am in danger of getting around to Real Soon Now. Paul A Bristow Prizet Farmhouse, Kendal, Cumbria UK LA8 8AB +44 1539 561830 +44 7714 330204 mailto: pbristow(a)hetp.u-net.com

10 44

[program_options] Patch for Tru64/cxx65
by Markus Schöpflin 06 Jun '05

06 Jun '05

Attached patch fixes the last failure on Tru64. (See http://tinyurl.com/dfslr) Ok to commit? Markus Index: parsers_test.cpp =================================================================== RCS file: /cvsroot/boost/boost/libs/program_options/test/parsers_test.cpp,v retrieving revision 1.12 diff -u -w -r1.12 parsers_test.cpp --- parsers_test.cpp 27 Apr 2005 07:02:39 -0000 1.12 +++ parsers_test.cpp 17 May 2005 10:12:16 -0000 @@ -126,7 +126,13 @@ // Regression test: check that '0' as style is interpreted as // 'default_style' vector<option> a4 = - parse_command_line(5, cmdline3_, desc, 0, additional_parser).options; + parse_command_line(5, +#if BOOST_WORKAROUND(__DECCXX_VER, BOOST_TESTED_AT(60590042)) + static_cast<char const * const *>(cmdline3_), +#else + cmdline3_, +#endif + desc, 0, additional_parser).options; BOOST_CHECK_EQUAL(a4.size(), 4u); check_value(a4[0], "foo", "4");

3 5

Customer Friendlier Boost Installation
by Paul A Bristow 03 Jun '05

03 Jun '05

Recently we had a flurry of discussion of why Boost is not more widely used. I have used some bits of Boost happily for years, but recently I felt I had to use it 'properly' so concluded it was time I used the bjam build system - something I had felt looked a bit clever. Indeed my suspicion was proved correct - I couldn't get it to work. The instructions are far too Unix/Make oriented for this mere Windows hacker; the Wiki Windows specific instructions are much better but suggest copying (duplicating) a lot of files to avoid an addition to the include directories, and don't apply to the most recent compilers. Victor Wagner gave some hints recently using environment variables, but not the full text. I don't doubt the usefulness of the bjam to the congescenti - only with this could the incredibly impressive build & test system allow all the cracks and chasms of compiler non-compliance and language imprecision to be papered over - Boost's great strength. But if we are to interest your average Windows IDE programmer, for whom MAKE is a blank or a distant nightmare, IMO we need some much better and simpler instructions, including lots of examples of input and output, and to be as well tested as the rest of Boost code. Like it or not, these are the potential customers and I spoken to many who are put off by the 'build barrier'. I suggest that we explicitly say that you can use all of Boost (except the ones that must have built libraries) by adding to the include list - and say exactly how in Windows IDE-ese. If you share my concern, would some someone like to prepare something? Or at least provide details of their way of doing it. I will be willing to help with checking/editing. Paul Paul A Bristow Prizet Farmhouse, Kendal, Cumbria UK LA8 8AB +44 1539 561830 +44 7714 330204 mailto: pbristow(a)hetp.u-net.com

33 194

set arithmetic proposal
by Dominic Herity 02 Jun '05

02 Jun '05

I've often wished that STL sets supported set arithmetic operations. Addition, multiplication and subtraction operators could be overloaded with set union, intersection and difference respectively. For example: set<char> s1; // s1 <- {} s1 += 'a'; // s1 <- {a} set<char> s2 = s1 + 'b'; // s2 <- {a,b} set<char> s3 = s2 - s1; // s3 <- {b} set<char> s4 = s1 * s2; // s4 <- {a} Achieving the above results with the existing STL library is by comparison cumbersome, error-prone and unclear. Although this proposal does not follow the STL design philosophy, it meets a real developer need. It can be implemented using global operator overloads, which should avoid conflicts with other uses of STL sets. Comments and suggestions welcome. Best Regards Dominic Herity Red Plain Technology (www.redplain.com) T:+353-87-2208233 F: +1-603-909 6570 E: dominic.herity(a)redplain.com

6 7

[filesystem] Windows DLL causes invalid bad path exception
by David Daeschler 01 Jun '05

01 Jun '05

was: [filesystem] WIN32 Can not use drive name in directory I found posts on boost-users from January that describe the problem that I am having. Was there ever a resolution to this? Mr. Dawes, you mention a re-work in the next version of the library ( http://lists.boost.org/boost-users/2005/01/9607.php ) that would fix this problem. Is that rework complete, or not yet? Should I just use the static version for now and not worry about this problem? Comment reference for January 2005 http://lists.boost.org/boost-users/2005/01/9560.php http://lists.boost.org/boost-users/2005/01/9581.php http://lists.boost.org/boost-users/2005/01/9583.php http://lists.boost.org/boost-users/2005/01/9584.php http://lists.boost.org/boost-users/2005/01/9586.php http://lists.boost.org/boost-users/2005/01/9607.php http://lists.boost.org/boost-users/2005/01/9610.php http://lists.boost.org/boost-users/2005/01/9639.php Thank you, - Dave

2 2