- Boost - lists.boost.org

Automated error analysis with correlation
by Steven Watanabe 15 Apr '08

15 Apr '08

AMDG The topic of treating error analysis has come up on this list several times in the past. The main difficulty is that the uncertainties of variables are not always independent. I recently had an idea of how to overcome this weakness. The first assumption is that all the variables that we start with are independent. Then, each variable keeps track of how it depends on each of the initial variables. The simplest implementation is to create a unique integer id for every new variable constructed. The format of a variable in the follwing discussion is (value, [(id1, alternate_value1), (id2, alternate_value2)]) where the alternate_values are the values that we would get if the value were increased by their uncertainties. value_with_uncertainty x(1.2, 0.2); // holds (1.2, [(0, 1.4)]) value_with_uncertainty y(1.2, 0.2); // holds (1.2, [(1, 1.4)]) Now, adding x + x, we add the values to get (2.4, ...). We then see that both arguments are dependent on the input with id 0. So, we add these two alternate values to get (2.4, [(0, 2.8)]). The total uncertainty is then 0.4. If on the other hand, we add x + y, the value will still be 2.4, but only the first element of the list of uncertainties depends on id 0, so we use the value of y and the adjusted value of x yielding (2.4, [(0, 2.6), ...]). The same rule gives us the dependence on id 1, for a final result of (2.4, [(0, 2.6), (1, 2.6)]) Now, the combined uncertainty is sqrt(0.2*0.2 + 0.2*0.2) = 0.28. There are several ways this can be made more accurate. 1) instead of taking x + uncertainty_of_x we can take several sample points inside the range x +- uncertainty_of_x 2) it should be possible to track dependence on groups of variables in addition to dependence on single variables independently. All of this adds overhead of course, but I think that a reliable mechanism for error propagation is often going to be worth that cost. The code is attached. Comments? In Christ, Steven Watanabe // value_and_uncertainty.hpp // // Copyright (c) 2008 // Steven Watanabe // // Distributed under the Boost Software License, Version 10.0 (See // accompanying file LICENSE_1_0.txt or copy at // http://www.boost.org/LICENSE_1_0.txt) #ifndef VALUE_AND_UNCERTAINTY_HPP_INCLUDED #define VALUE_AND_UNCERTAINTY_HPP_INCLUDED #include <vector> #include <utility> #include <numeric> #include <cmath> #include <boost/lambda/lambda.hpp> #include <boost/lambda/bind.hpp> template<class T> class value_and_uncertainty { public: value_and_uncertainty(const T& val, const T& unc) : value_(val) { if(unc != T()) { alternate_values.push_back(std::make_pair(new_id(), val + unc)); } } template<class F> value_and_uncertainty(F f, const value_and_uncertainty& arg0) : value_(f(arg0.value_)) { typename alternate_t::const_iterator begin = arg0.alternate_values.begin(), end = arg0.alternate_values.end(); for(; begin != end; ++begin) { alternate_values.push_back(std::make_pair(begin->first, f(begin->second))); } } template<class F> value_and_uncertainty(F f, const value_and_uncertainty& arg0, const value_and_uncertainty& arg1) : value_(f(arg0.value_, arg1.value_)) { typename alternate_t::const_iterator begin0 = arg0.alternate_values.begin(), end0 = arg0.alternate_values.end(); typename alternate_t::const_iterator begin1 = arg1.alternate_values.begin(), end1 = arg1.alternate_values.end(); while(true) { if(begin0 == end0) { for(; begin1 != end1; ++begin1) { alternate_values.push_back(std::make_pair(begin1->first, f(arg0.value_, begin1->second))); } break; } else if(begin1 == end1) { for(; begin0 != end0; ++begin0) { alternate_values.push_back(std::make_pair(begin0->first, f(begin0->second, arg1.value_))); } break; } else { if(begin0->first < begin1->first) { alternate_values.push_back(std::make_pair(begin0->first, f(begin0->second, arg1.value_))); ++begin0; } else if(begin1->first < begin0->first) { alternate_values.push_back(std::make_pair(begin1->first, f(arg0.value_, begin1->second))); ++begin1; } else { alternate_values.push_back(std::make_pair(begin0->first, f(begin0->second, begin1->second))); ++begin0; ++begin1; } } } } const T& value() const { return(value_); } T uncertainty() const { namespace bll = boost::lambda; using namespace bll; using std::sqrt; return(sqrt(std::accumulate(alternate_values.begin(), alternate_values.end(), T(), _1 + bll::bind(bll::protect(_1 * _1), (bll::bind(&std::pair<id_type, T>::second, _2) - value_))))); } private: T value_; typedef int id_type; static id_type new_id() { static id_type result = 0; return(result++); } typedef std::vector<std::pair<id_type, T> > alternate_t; std::vector<std::pair<id_type, T> > alternate_values; }; template<class T> value_and_uncertainty<T> operator+(const value_and_uncertainty<T>& arg0, const value_and_uncertainty<T>& arg1) { using namespace boost::lambda; return(value_and_uncertainty<T>(_1 + _2, arg0, arg1)); } template<class T> value_and_uncertainty<T> operator*(const value_and_uncertainty<T>& arg0, const value_and_uncertainty<T>& arg1) { using namespace boost::lambda; return(value_and_uncertainty<T>(_1 * _2, arg0, arg1)); } template<class T> std::ostream& operator<<(std::ostream& os, const value_and_uncertainty<T>& arg) { return(os << arg.value() << " +- " << arg.uncertainty()); } #endif

4 5

Supported compilers: gcc on osx?
by Dirk Griffioen 15 Apr '08

15 Apr '08

Hi All, If I check the supported compilers page, gcc on darwin is supported - but I get the following error (boost_1_34_1/boost_1_35_0) * *The utf conversion in boost.program_options, for example to_utf8 gives std::runtime_error: locale::facet::_S_create_c_locale name not valid How do I repair this? The osx locale tools are not as conveniant as say ubuntu's ... (Should I not use Xcode's gcc, but build my own with a proper stdc++, or is there some fix in the os ?) Further: I have osx 10.5, but bjam thinks I have 10.4 - this is not a problem I guess, but it sure is not correct: bin/darwin/debug/link-static/macosx-version-10.4/threading-multi/ (the build path) Regards, - Dirk

2 2

Status of Boost.Process?
by Boris 15 Apr '08

15 Apr '08

Does anyone know what's the status of Boost.Process? I sent a mail to Julio ten days ago but didn't get any answer. There is some pending work as explained at http://www.netbsd.org/~jmmv/process/boost_process/project_position.html#boo… but it doesn't sound like there is missing a lot. There are various problems in the Windows code though but minor ones (there is for example std::string used everywhere but when compiled as a Unicode program on Windows wide-character API functions are used which doesn't work). Boris

1 0

[config] BOOST_HAS_PTHREADS config test
by Markus Schöpflin 15 Apr '08

15 Apr '08

Hello, I would like to commit attached patch to clear this error: http://tinyurl.com/5ymwv5. OK to commit? Regards, Markus Index: boost_has_pthreads.ipp =================================================================== --- boost_has_pthreads.ipp (revision 44408) +++ boost_has_pthreads.ipp (working copy) @@ -14,7 +14,7 @@ namespace boost_has_pthreads{ -void* thread_proc(void* arg) +extern "C" void* thread_proc(void* arg) { return arg; }

2 2

ARM help needed
by Peter Dimov 15 Apr '08

15 Apr '08

Can someone with access to ARM please run libs/smart_ptr/test/spinlock_test.cpp and/or take a look at boost/detail/spinlock_gcc_arm.hpp? I've verified that it compiles, but can't run it.

3 11

Re: [boost] Egg 2nd request for reviews
by dan marsden 15 Apr '08

15 Apr '08

><gpderetta(a)gmail.com> wrote: >> If there is a possibility to extend the review period of another week, >> I'll take sometime to try the library, possibly with different >> compilers. This may also let other boosters have time write their own >> review (even if strangely so far no one else seems to have shown >> interest). >> >> If you think that the review should still end the 13th, I'll wrap up a >> review for today. > >Ditto. I have an interest in the problem domain Egg addresses, and >from what I've read, I like Shunsuke's approach. But I'm just now >sitting down to take a real look at it. I can wrap up a review today, >but if you don't mind extending the review period, I think the >discussion could benefit from more time. > >Thanks, >Daniel Walker I do intend to extend the review period by a week. I appreciate that Egg is a big library, and it takes a bit of time to review fully. So no need to rush a review, but please do put them in before the end of next week. Currently the review count sadly still sits at zero, so your reviews will be gladly received thanks! Cheers Dan ___________________________________________________________ Yahoo! For Good helps you make a difference http://uk.promotions.yahoo.com/forgood/

4 6

boost.iterator's zip_iterator::value_type should not be reference
by thomas weidner 14 Apr '08

14 Apr '08

Hi, defining zip_iterator::value_type in the boost.iterator library to be the same as the reference type breaks some mutating algorithms on zip_iterators. for example sort (c++0x inside): template<class... Arrays> void array_multisort(Arrays& ... arrays) { using namespace boost; std::sort(make_zip_iterator(make_tuple(arrays.begin()...)), make_zip_iterator(make_tuple(arrays.end()...))); } won't work as expected. Patching zip_iterator to provide a more natural value_type makes the above code work as intended. The patch is attached to this message. mfg thomas

1 0

[1.35.0] Threads - build failure using gcc on Win XP
by Mat Marcus 14 Apr '08

14 Apr '08

It seems that I am unable to build/use boost threads when compiling with gcc-4.2.0 on windows. I get errors: libs\thread\src\win32\thread.cpp:147: error: 'heap_delete' is not a member of 'boost::detail' and other (see below). Can someone advise me as to how I can get boost_thread to build under gcc-4.2.0/32-bit XP machine wth boost 1.35.0. Thanks, Mat $ bjam -j2 debug gcc-4.2.0 --without-mpi --without-python threadapi=win32 libs/thread/build//boost_thread warning: Graph library does not contain optional GraphML reader. note: to enable GraphML support, set EXPAT_INCLUDE and EXPAT_LIBPATH to the note: directories containing the Expat headers and libraries, respectively. Building Boost.Regex with the optional Unicode/ICU support disabled. Please refer to the Boost.Regex documentation for more information (don't panic: this is a strictly optional feature). ...patience... ...found 676 targets... ...updating 3 targets... gcc.compile.c++ bin.v2\libs\thread\build\gcc-4.2.0\debug\threading-multi\thread.o libs\thread\src\win32\thread.cpp: In function 'void boost::<unnamed>::run_thread_exit_callbacks()': libs\thread\src\win32\thread.cpp:147: error: 'heap_delete' is not a member of 'boost::detail' libs\thread\src\win32\thread.cpp:149: error: 'heap_delete' is not a member of 'boost::detail' libs\thread\src\win32\thread.cpp:159: error: 'heap_delete' is not a member of 'boost::detail' libs\thread\src\win32\thread.cpp: At global scope: libs\thread\src\win32\thread.cpp:170: error: no 'unsigned int boost::thread::thread_start_function(void*)' member function declared in class 'boost::thread' libs\thread\src\win32\thread.cpp: In member function 'void boost::thread::start_thread()': libs\thread\src\win32\thread.cpp:194: error: ISO C++ forbids taking the address of an unqualified or parenthesized non-static member function to form a pointer to member function. Say '&boost::thread::thread_start_function' libs\thread\src\win32\thread.cpp:194: error: 'struct boost::detail::thread_data_base' has no member named 'id' libs\thread\src\win32\thread.cpp:194: error: '_beginthreadex' was not declared in this scope libs\thread\src\win32\thread.cpp:199: error: 'intrusive_ptr_add_ref' was not declared in this scope libs\thread\src\win32\thread.cpp:200: error: 'boost::detail::win32' has not been declared libs\thread\src\win32\thread.cpp: In constructor 'boost::<unnamed>::externally_launched_thread::externally_launched_thread()': libs\thread\src\win32\thread.cpp:215: error: 'count' was not declared in this scope libs\thread\src\win32\thread.cpp:216: error: 'interruption_enabled' was not declared in this scope libs\thread\src\win32\thread.cpp: In function 'void boost::<unnamed>::make_external_thread_data()': libs\thread\src\win32\thread.cpp:225: error: 'heap_new' is not a member of 'boost::detail' libs\thread\src\win32\thread.cpp:225: error: expected primary-expression before '>' token libs\thread\src\win32\thread.cpp:225: error: expected primary-expression before ')' token libs\thread\src\win32\thread.cpp: In constructor 'boost::thread::thread(boost::detail::thread_move_t<boost::thread>)': libs\thread\src\win32\thread.cpp:251: error: no match for 'operator=' in 'x. boost::detail::thread_move_t<T>::operator-> [with T = boost::thread]()->boost::thread::thread_info = 0' ./boost/shared_ptr.hpp:194: note: candidates are: boost::shared_ptr<T>& boost::shared_ptr<T>::operator=(const boost::shared_ptr<T>&) [with T = boost::detail::thread_data_base] libs\thread\src\win32\thread.cpp: In member function 'void boost::thread::join()': libs\thread\src\win32\thread.cpp:292: error: 'interruptible_wait' is not a member of 'boost::this_thread' libs\thread\src\win32\thread.cpp:292: error: 'boost::detail::timeout' has not been declared libs\thread\src\win32\thread.cpp:293: error: 'release_handle' was not declared in this scope libs\thread\src\win32\thread.cpp: In member function 'bool boost::thread::timed_join(const boost::system_time&)': libs\thread\src\win32\thread.cpp:302: error: 'interruptible_wait' is not a member of 'boost::this_thread' libs\thread\src\win32\thread.cpp:306: error: 'release_handle' was not declared in this scope libs\thread\src\win32\thread.cpp: In member function 'void boost::thread::detach()': libs\thread\src\win32\thread.cpp:313: error: 'release_handle' was not declared in this scope libs\thread\src\win32\thread.cpp: At global scope: libs\thread\src\win32\thread.cpp:316: error: no 'void boost::thread::release_handle()' member function declared in class 'boost::thread' libs\thread\src\win32\thread.cpp: In member function 'void boost::thread::release_handle()': libs\thread\src\win32\thread.cpp:319: error: no match for 'operator=' in '((boost::thread*)this)->boost::thread::thread_info = 0' ./boost/shared_ptr.hpp:194: note: candidates are: boost::shared_ptr<T>& boost::shared_ptr<T>::operator=(const boost::shared_ptr<T>&) [with T = boost::detail::thread_data_base] libs\thread\src\win32\thread.cpp: In member function 'void boost::thread::interrupt()': libs\thread\src\win32\thread.cpp:327: error: 'struct boost::detail::thread_data_base' has no member named 'interrupt' libs\thread\src\win32\thread.cpp: In member function 'bool boost::thread::interruption_requested() const': libs\thread\src\win32\thread.cpp:334: error: 'boost::detail::win32' has not been declared libs\thread\src\win32\thread.cpp:334: error: 'struct boost::detail::thread_data_base' has no member named 'interruption_handle' libs\thread\src\win32\thread.cpp: In static member function 'static unsigned int boost::thread::hardware_concurrency()': libs\thread\src\win32\thread.cpp:339: warning: missing braces around initializer for '_SYSTEM_INFO::<anonymous union>' libs\thread\src\win32\thread.cpp: At global scope: libs\thread\src\win32\thread.cpp:344: error: 'native_handle_type' in class 'boost::thread' does not name a type libs\thread\src\win32\thread.cpp:360: error: 'timeout' is not a member of 'boost::detail' libs\thread\src\win32\thread.cpp:361: error: expected ',' or ';' before '{' token libs\thread\src\win32\thread.cpp:628: error: expected `}' at end of input libs\thread\src\win32\thread.cpp:628: error: expected `}' at end of input libs\thread\src\win32\thread.cpp:628: error: expected `}' at end of input ./boost/detail/shared_count.hpp: In constructor 'boost::detail::shared_count::shared_count(P, D) [with P = boost::detail::thread_data_base*, D = bool]': ./boost/shared_ptr.hpp:177: instantiated from 'boost::shared_ptr<T>::shared_ptr(Y*, D) [with Y = boost::detail::thread_data_base, D = bool, T = boost::detail::thread_data_base]' libs\thread\src\win32\thread.cpp:135: instantiated from here ./boost/detail/shared_count.hpp:115: error: 'd' cannot be used as a function ./boost/detail/sp_counted_impl.hpp: In member function 'void boost::detail::sp_counted_impl_pd<P, D>::dispose() [with P = boost::detail::thread_data_base*, D = bool]': libs\thread\src\win32\thread.cpp:627: instantiated from here ./boost/detail/sp_counted_impl.hpp:144: error: '((boost::detail::sp_counted_impl_pd<boost::detail::thread_data_base*, bool>*)this)->boost::detail::sp_counted_impl_pd<boost::detail::thread_data_base*, bool>::del' cannot be used as a function "/opt/gcc-4.2.0/bin/g++.exe" -ftemplate-depth-128 -O0 -fno-inline -Wall -g -mthreads -DBOOST_ALL_NO_LIB=1 -DBOOST_THREAD_BUILD_DLL=1 -I"." -c -o "bin.v2\libs\thread\build\gcc-4.2.0\debug\threading-multi\thread.o" "libs\thread\src\win32\thread.cpp" ...failed gcc.compile.c++ bin.v2\libs\thread\build\gcc-4.2.0\debug\threading-multi\thread.o... ...skipped <pbin.v2\libs\thread\build\gcc-4.2.0\debug\threading-multi>boost_thread-gcc42-mt-d-1_35.lib for lack of <pbin.v2\libs\thread\build\gcc-4.2.0\debug\threading-multi>thread.o... ...skipped <pbin.v2\libs\thread\build\gcc-4.2.0\debug\threading-multi>boost_thread-gcc42-mt-d-1_35.dll for lack of <pbin.v2\libs\thread\build\gcc-4.2.0\debug\threading-multi>thread.o... ...failed updating 1 target... ...skipped 2 targets...

4 14

[python] Crazy crash
by Robert Dailey 14 Apr '08

14 Apr '08

Hi, This crash has always occurred, even since 1.34.1. It's now reproducible in 1.35. I use the following command line to build boost in such a way that Boost.Python links against the debug python libraries: bjam --toolset=msvc --layout=system link=shared threading=multi variant=debug python-debugging=on stage Note that Visual Studio 2008 is the only MSVC compiler installed on my system. I'm using Python 2.5.2. Any time I call into a Boost.Python function I get an access violation when writing to location 0x00000000. Note that I have been able to get crashes when calling the initMyInterface() function for BOOST_PYTHON_MODULE(MyInterface), I also have gotten this crash when calling into boost::list::insert(). I have no idea what the issue could be. Note that when I built Boost.Python (debug) using the visual studio project before it was removed in 1.35, it worked perfectly. Any help is greatly appreciated since this is a complete blocker issue for me.

1 3

Regarding boost.math library
by Anup Mandvariya 14 Apr '08

14 Apr '08

Hi All, I have some queries related with boost.math library as ststed below: a) Is it possible to write many of the functions in the boost.math library (like beta and gamma functions) so that they can be parallelized irrespective of the "way of parallelization"(such as using OpenMP, MPI, etc...) and the "environment" (like multicore or a cluster)? b) What is the possibility of extending boost.math libraries (particularly beta and gamma functions) to generic libraries using generic programming process? -- Regards, Anup Mandvariya +919985330660 "Truth Must Have No Compromise"

4 6