- Boost-users - lists.boost.org

issues with the latest boost numeric bindings
by Simon Mpasi 11 Jan '11

11 Jan '11

To boost users: Greetings! I do not know if this the correct format but here are the issues that I am running into: /boost/numeric/bindings/lapack/detail/ The file clapack.h includes itself causing an error /boost/numeric/bindings/blas/detail/ The file cblas.h includes itself causing an error I changed them to lapack.h and blas.h respectively but I got several errors. I checked out the latest version of numeric bindings from boost.svn this morning. I am using Microsoft Visual Studio 2008 without a Fortran compiler (I need to use clapack) The issues here are that the l lapack routines come with 2 underscores, i.e. _sgesvd_ instead of sgesvd_ I declared BOOST_NUMERIC_BINDINGS_USE_CLAPACK in my driver. Thank you very much in advance for your assistance. Simon N. Mpasi

2 1

[Build] Mimicking my Fedora13 Boost installation.
by Robert Jones 11 Jan '11

11 Jan '11

I have Boost installed as part of my Fedora13/64-bit system, at Boost 1.41. I now want install Boost 1.45, but need to keep the original around for backwards compatibility issues, so I'm putting my 1.45 build somewhere else, and and I want it look as much like the original 1.41 build as possible, but be 1.45. So I built with this command > ./bjam --prefix=/usr/local/boost_1_45_0 threading=single --runtime-link=shared --link=shared --layout=tagged install then listing all the Boost regex (regex chosen at random) I find > find /usr -name 'libboost_regex*' 187532 0 lrwxrwxrwx 1 root root 24 Sep 21 16:25 /usr/lib64/libboost_regex.so -> libboost_regex.so.1.41.0 142821 972 -rwxr-xr-x 1 root root 994792 Jun 4 2010 /usr/lib64/libboost_regex.so.1.41.0 540038 2356 -rw-r--r-- 1 root root 2410606 Jan 11 11:56 /usr/local/boost_1_45_0/lib/libboost_regex.a 540030 1132 -rwxr-xr-x 1 root root 1156757 Jan 11 11:55 /usr/local/boost_1_45_0/lib/libboost_regex.so.1.45.0 540031 0 lrwxrwxrwx 1 root root 24 Jan 11 11:55 /usr/local/boost_1_45_0/lib/libboost_regex.so -> libboost_regex.so.1.45.0 which seems to show an extra .a compared to my Fedora 'default' installation. How come? Do I need it? Can I determine the bjam flags that Fedora used originally? Thx, - Rob.

2 2

[Iterator] advance and random_access_traversal iterators
by Anthony Foglia 11 Jan '11

11 Jan '11

What happened to the patch to add a boost::advance function designed to work with Boost.Iterator's concepts? Three years ago there was a thread pointing out the problem with std::advance not treating random access traversal iterators as random access (because of the reference type). Sebastian Redl posted a patch to add a boost::advance (and boost::distance) function. <http://lists.boost.org/Archives/boost/2007/09/127785.php> I don't see any proof of this patch ever being applied. Someone filed a bug report two years ago to bring attention to the patch, and it still hasn't been applied. <https://svn.boost.org/trac/boost/ticket/2539> Is there a reason it hasn't been applied (the collision with the boost::distance supplied by Boost.Range?) or has it just fallen through the cracks? -- Anthony Foglia Princeton Consultants (609) 987-8787 x233

2 1

Problem compiling under Codegear
by jose perez 10 Jan '11

10 Jan '11

Hello. I’m using CodeGear compiler with Boost 1.39 and Boost 1.45. In both of them I have found an error in mpl\base.hpp line 25 (typedef typename T::base type;) [BCC32 Error] base.hpp(25): E2404 Dependent type qualifier 'na' has no member type named 'base' The code is: namespace boost { namespace mpl { template< typename BOOST_MPL_AUX_NA_PARAM(T) // find in mpl\aux_\na_spec.hpp, line 152: #define BOOST_MPL_AUX_NA_PARAM(param) param = na > struct base { typedef typename T::base type; // <------- ERRORRR BOOST_MPL_AUX_LAMBDA_SUPPORT(1,base,(T)) }; BOOST_MPL_AUX_NA_SPEC(1, base) }} Help will be appreciated. Thanks!

1 0

[mpl] resulting sequence limits
by Igor R 10 Jan '11

10 Jan '11

Hello, I've got 2 mpl::map's, both have less than 20 elements. I copy one to another, and the result has more than 20 elements. I thought that BOOST_MPL_LIMIT_MAP_SIZE macro affects only explicit definitions of maps, not result values of metafunctions, but it turns out that the above case doesn't compile -- until I #include <boost/mpl/map/map30.hpp> Is it behavior "by design", or is it a bug? Thanks.

2 1

[Range] Adaptors
by Robert Jones 10 Jan '11

10 Jan '11

I might be asking for the impossible here, but given the lovely adaptor syntax, eg std::vector<int> vec; boost::copy( vec | boost::adaptors::reversed | boost::adaptors::uniqued, std::ostream_iterator<int>(std::cout) ); I notice it still uses the function call notation in the outermost operation( boost::copy() ), can it be written to eliminate function call syntax completely, say something like vec | boost::adaptors::reversed | boost::adaptors::uniqued | boost::adaptors::copy( std::ostream_iterator<int>(std::cout) ); or even vec | boost::adaptors::reversed | boost::adaptors::uniqued | std::ostream_iterator<int>(std::cout); Thx, - Rob.

5 9

Boost.Regex compile error with gcc 4.6
by Nathan Ridge 10 Jan '11

10 Jan '11

Hello, boost/regex.hpp fails to compile with gcc 4.6, giving the following errors: In file included from /lib/boost/boost/regex/v4/perl_matcher.hpp:578:0, from /lib/boost/boost/regex/v4/regex.hpp:88, from /lib/boost/boost/regex.hpp:31, from test.cpp:1: /lib/boost/boost/regex/v4/perl_matcher_non_recursive.hpp: In member function ‘bool boost::re_detail::perl_matcher<BidiIterator, Allocator, traits>::unwind_short_set_repeat(bool)’: /lib/boost/boost/regex/v4/perl_matcher_non_recursive.hpp:1394:85: error: invalid use of incomplete type ‘const struct boost::regex_traits_wrapper<traits>’ /lib/boost/boost/regex/v4/regex_traits.hpp:163:8: error: declaration of ‘const struct boost::regex_traits_wrapper<traits>’ /lib/boost/boost/regex/v4/perl_matcher_non_recursive.hpp:1394:85: error: invalid type argument of unary ‘*’ (have ‘BidiIterator’) It compiles fine with gcc 4.5. Is this an issue with boost or gcc? Should I report it as a bug? Thanks, Nate.

4 5

[Proto] Storing an Expression Template as a Data Member
by Hossein Haeri 10 Jan '11

10 Jan '11

Dear all, Is there any way I can have classes with data members of type (proto) expression templates? Here is where I get stuck: template<typename T> class MyClass { ... T expression_; }; What type should T be? If you're wondering why on earth I might need that, here is an explanation: Instances of MyClass need to store the lazy calculation they will be performing later on when enough initialisation data is available. Each instance of MyClass has its own FORMULATION for the lazy calculation in terms of the initialisation data. They need to be fed with these formulations and I thought (proto) expression templates will perhaps do the job. But, then, I don't know how to store these formulations until it's time for them to perform... TIA, --Hossein

4 15

multi_index_container: replace() modifies my iterator
by Nicolas Rannou 10 Jan '11

10 Jan '11

Hi all, I'm pretty new in boost and I am experimenting an issue I can't understand. It looks like the usage of "replace()" modifies the iterator I am iterating on: Here is a simple example: it iterates properly if I do: Iterator it2, it3; boost::tuples::tie(it2, it3) = m_Container.get< CollectionID >().equal_range( 30 ); while( it2!= it3 ) { int time = it2->TCoord; if( time < 500 ) // we want to change the collection ID in this case { Structure tempStructure(*it2); tempStructure.CollectionID = 40; * //m_Container.get< CollectionID >().replace(it2, tempStructure); comment this line* } ++it2; } whereas if I call "replace()" the iterator value changes to "it3" then it exists the loop: Iterator it2, it3; boost::tuples::tie(it2, it3) = m_Container.get< CollectionID >().equal_range( 30 ); while( it2!= it3 ) { int time = it2->TCoord; if( time < 500 ) // we want to change the collection ID in this case { Structure tempStructure(*it2); tempStructure.CollectionID = 40; * m_Container.get< CollectionID >().replace(it2, tempStructure);* } ++it2; } the replace() method returns true. I must be missing something but after reading the documentation and some researches on the web I couldn't find the answer. Any help/explanation would be greatly appreciated, Thanks, Nicolas

3 3

using '\0' as separator in char_separator and tokenizer classes
by kmulla＠mail.com 10 Jan '11

10 Jan '11

Hi All, I am trying to use boost::tokenizer to tokenize a string containing embedded '\0' characters. I am using boost::char_separator to specify the separator. I am not getting correct tokens when I am using '\0' as the separator. Here I am curious to know, if it is indeed possible to use '\0' as separator in boost::char_separator class. I am pretty much using the code in the example given on below link. I know my code works since I am getting correct results with different separator values. http://www.boost.org/doc/libs/1_45_0/libs/tokenizer/char_separator.htm Best, K Mulla

1 0