- Boost-users - lists.preview.boost.org

uBlas blas and lapack binds
by geert_ceuppens 25 Mar '03

25 Mar '03

When will the blas and lapack binds be included with uBlas? I hoped to see them in the 1-30 release : ( Thanks - Geert

2 1

ublas: questions from a newcomer
by stuart macgregor 25 Mar '03

25 Mar '03

I am interested in ublas for linear algebra. In addition to the simple operations supported by ublas I will need comprehensive access to lapack for the real work. I see in the wiki that some work has been done in this area. Is this numeric/binding stuff working, and can it be readily extended to the full set of lapack routines? Some kind of meta program to generate lapack interfaces from a parameter spec would be nice (ruby/perl? or just a description of what needs to be done) Then I could happily bring up the interfaces I need. Is there any way to generate vectore and matrices which do not own the data buffer which they provide access to - 'view' or 'sub' objects in other lin alg packages? This would be very valuable to me for interfacing existing code, without duplicating storage and doing unnecessary copies. I found it difficult to determine from the current documentation. I have one worry: Ublas, and most of the boost modules, seems to me not for the faint hearted. They are clearly powerful, but seem to expect a lot from a potential user. Are these facilities aimed at library writers rather than users whoes skill focus may lie in other domains? I am thinking of apl (and related J++ (not java)) which is very powerful, but has only minority appeal because it is too alien and unnatural for ordinary mortals. The advocates take great pride in how powerful a small and almost incomprehensible line of code can be (FFT in one short line etc.), and implicitly how clever they are to generate and understand it. I expect that a user level doc may eventually make ublas/lapack much more approachable by a non computer science mathematician or engineer. Forgive me if the above sounds critical and ungracious for free software, but I want to use boost and I am worried that it may be too demanding for its benefites to be justified for use by the group of people I work with. -- Stuart

3 2

RE: [Boost-Users] shared_array with memcpy, help request
by bjorn.karlsson＠readsoft.com 25 Mar '03

25 Mar '03

> From: Blake D [mailto:BlakeTNC@yahoo.com] > I am a new boost user, and I'm trying to convert a large > project to use the > smart pointers. I have a use for shared_array buy the memcpy > line from the > following code snippet(, and all memcpy's) throw an exception with the > shared_arrays and I don't know why. Am I doing something > obvious wrong? [snip] > boost::shared_array<uint8> p_data(new uint8[data_length]); The above declaration of a local variable called p_data is probably not what you intended (you want to initialize the member variable p_data). If you cannot (or won't) perform the initialization in the constructor initializer, you'll need to reset the shared_array like so: p_data.reset(new uint8[data_length]); Or, in the initializer: c_Packet::c_Packet (const c_Packet & r_packet) : p_data(new uint8[r_packed.data_length]) { data_length = r_packet.data_length; ... } ...and then perform the memcpy. By the way, you are actually sharing this data, right? If not, scoped_array would probably be a better choice for managing resource lifetime. Bjorn Karlsson

1 0

shared_array with memcpy, help request
by Blake D 25 Mar '03

25 Mar '03

Hello everyone :) I am a new boost user, and I'm trying to convert a large project to use the smart pointers. I have a use for shared_array buy the memcpy line from the following code snippet(, and all memcpy's) throw an exception with the shared_arrays and I don't know why. Am I doing something obvious wrong? Thank you, Blake //------------------class declaration snippet--------------------------------- // Pointer to the packet boost::shared_array<uint8> p_data; // uint8 * p_data; <--- old working line //------------------class cpp file snippet--------------------------------- // copy Constructor c_Packet::c_Packet (const c_Packet & r_packet) { data_length = r_packet.data_length; boost::shared_array<uint8> p_data(new uint8[data_length]); // p_data = new uint8[data_length]; <--- old working line // Copy data from our source object to this object memcpy(p_data.get(), r_packet.p_data.get(), data_length); // <- throws exception! // memcpy(p_data, r_packet.p_data, data_length); <---- old working line #ifdef DO_DEBUG // cout << "packet copy constructor called" << endl ; #endif // DO_DEBUG }

2 1

regarding the quality of math functions in stl/boost
by adam_speight 24 Mar '03

24 Mar '03

I've often heard that you should never trust your system implementation of functions like sqrt, exp, sin, cos, rand, etc. Instead, most mathematicians say that one should use an implementation that is known to be accurate up to some order. Unfortunately, using third-party code for these functions hinders portability. I know that boost has a formidable random number generation library that uses the cmath versions of basic math functions. Does the stl standard, or various implementations of stl, make any gaurantees about the accuracy of these basic math functions? Are they safe to use for mathematical applications in practice? Thanks, Adam

2 1

First problems with 1.30.0
by klapshin 24 Mar '03

24 Mar '03

I am trying new 1.30.0 release and got couple of problems on MSVC6 SP5. First of all, compiling #include <boost/date_time/gregorian/gregorian.hpp> int main() { boost::gregorian::from_string( "" ); return 0; } yields warning: I:\SW\EXTERNAL\BOOST_1_30_0\boost/lexical_cast.hpp(147) : warning C4800: 'void *' : forcing value to bool 'true' or 'false' (performance warning) I:\SW\EXTERNAL\BOOST_1_30_0\boost/lexical_cast.hpp(146) : while compiling class-template member function 'bool __thiscall boost::detail::lexical_stream<int,class std::basic_string<char,struct std::char_traits<char>,class std::allocator<char> > >::operator <<(const class std::basic_string<char,struct std::char_traits<char>,class std::allocator<char> > &)' Much bigger problem is that compiling simple #include <boost/random/uniform_01.hpp> int main(){return 0;} fails with: I:\SW\EXTERNAL\BOOST_1_30_0\boost/random/uniform_01.hpp(44) : error C2275: 'std::numeric_limits<`template-parameter258'>::is_integer' : illegal use of this type as an expression I:\SW\EXTERNAL\BOOST_1_30_0\boost/random/uniform_01.hpp (95) : see reference to class template instantiation 'boost::uniform_01<UniformRandomNumberGenerator,RealTyp e>' being compiled I:\SW\EXTERNAL\BOOST_1_30_0\boost/random/uniform_01.hpp(44) : error C2027: use of undefined type 'STATIC_ASSERTION_FAILURE' I:\SW\EXTERNAL\BOOST_1_30_0\boost/static_assert.hpp(29) : see declaration of 'STATIC_ASSERTION_FAILURE' I:\SW\EXTERNAL\BOOST_1_30_0\boost/random/uniform_01.hpp (95) : see reference to class template instantiation 'boost::uniform_01<UniformRandomNumberGenerator,RealTyp e>' being compiled

7 9

RE: [Boost-Users] boost 1_30_0
by bjorn.karlsson＠readsoft.com 24 Mar '03

24 Mar '03

> From: Barnali [mailto:yg-boost-users@m.gmane.org] > > Has anyone figured out how to make boost 1_30_0 work with > Intel compiler > version 7 ? > None of the prescribed methods seem to work ! > I had to hard code it in some places to get it work. > > Barnali Any specific libraries? Error messages? What did you need to change? Without information about the problems you're experiencing, it's impossible for anyone to provide answers - please be more specific than "it doesn't work"! Bjorn Karlsson

1 0

run-tests.sh
by Barnali 24 Mar '03

24 Mar '03

The ~tools/regression/run-tests.sh does not work for multiple compilers. Barnali

2 1

Regex / boost 1.30.0 / HP-UX 11i / gcc 3.2
by Tom Matelich 24 Mar '03

24 Mar '03

Problem compiling c_regex_traits.cpp: g++ -c -DNO_PLT_CALLS -fPIC -ftemplate-depth-30 -Wall -W -g -BOOST_REGEX_NO_LIB -DBOOST_REGEX_STATIC_LINK -DDBG_ENABLED -DQT_NO_ASCII_CAST -DQT_ALTERNATE_QTSMANIP -DBOOST_REGEX_NO_LIB -DBOOST_REGEX_STATIC_LINK -I/usr/local/eddynet/opt/qt/mkspecs/hpux-g++ -I. -I. -I/usr/local/eddynet/opt/qt/include -IMOCS/ -o OBJS/c_regex_traits.o libs/regex/src/c_regex_traits.cpp libs/regex/src/c_regex_traits.cpp: In static member function `static void boost::c_regex_traits<char>::update()': libs/regex/src/c_regex_traits.cpp:655: call of overloaded `getlocale(int)' is ambiguous /usr/include/locale.h:114: candidates are: locale_data* getlocale(int) libs/regex/src/c_regex_traits.cpp:62: const char* <unnamed>::getlocale(int) For the time being I'm going to introduce a named namespace inside the anonymous namespace and scope the call. Thanks, Tom ----------------------------------------------------------------------- DISCLAIMER: Information contained in this message and/or attachment(s) may contain confidential information of Zetec, Inc. If you have received this transmission in error, please notify the sender by return email. -----------------------------------------------------------------------

2 1

Newbie help with date_time
by willw_at_ais 24 Mar '03

24 Mar '03

I am updating an old 1992-style application that previously used time_t based time to produce trend graphs of logged data. I am putting in ptime instead, and have come unstuck. The time range of the X-axis in this app is essentially under user control. It can show any period from a few seconds (sub seconds in the new version) up to three years. The labels on the x axis are placed at sensible points for the current range. For example if the graph shows 30 minutes, the ticks on the axis occur at 5 minute intervals; if it is displaying a week they occur on day boundaries. The old version of the program achieves this label placing by using the % operator and doing arithmetic with the seconds that underly time_t. I would like to do something similar with ptime. My attempt went like this: // Get next time (>= start) which falls on an itvl boundary ptime GetNextBoundary2(const ptime& start, const time_duration& itvl) { // Use 1/1/1970 as a base - not ideal but good enough for this const ptime origin(date(1970,1,1)); time_duration start_offset = start - origin; return ptime (start + itvl - nanosec(start_offset.ticks() % itvl.ticks())); } This fails because - among other reasons - although I can get ticks out, nanosec() won't put them back in again. I guess it is normalising on the way in? I feel sure I am going about this the wrong way, but am baffled as to what I should do next. All help gratefully received. Please be gentle with me - I am essentially a Pascal programmer who is busking, and finding this C++ template stuff hard. TIA Will Watts AIS Ltd

3 5