- Boost-users - lists.boost.org

[tuple] Tuple arithmetic
by Bill Buklis 13 Feb '12

13 Feb '12

Is there any facility to perform arithmetic operations on a tuple? For example: struct data { double a, b; }; data d1 = {1.0, 2.0}; data d2 = {2.5, 5.5}; // These two lines are not valid, but simulate what I want. boost::tie(d1.a, d1.b) *= 10.0; // Multiply each element by 10.0 boost::tie(d1.a, d1.b) += boost::tie(d2.a, d2.b); // Add the elements from the second object After calculations, d1 should equal {12.5, 25.5}; Thanks, -- Bill

4 6

[Format] No implicit conversion to string?
by John M. Dlugosz 13 Feb '12

13 Feb '12

It's shown prominently in the docs that a boost::format object can be sent to an ostream. cout << format("%2% %1%") % 36 % 77; But, there is no easy way to use it where a plain string is needed. E.g. throw runtime_error ( format("%2% %1%") % 36 % 77 ); One must use extra parens and a named function, either throw runtime_error ( (format("%2% %1%") % 36 % 77).str() ); or throw runtime_error ( str(format("%2% %1%") % 36 % 77) ); either of which detracts from the expression. Is there any good reason why an implicit conversion operator to the matching string class was omitted? I'd like to find out if there is some reason, before looking into submitting an update. —John

3 3

Poor/erratic boost::interprocess named_semaphore performance
by Davidson, Josh 13 Feb '12

13 Feb '12

I'm experiencing performance issues using named semaphores on Windows 7 x64. Currently, I'm on 1.49 beta 1, but the behavior was similar on 1.48 and 1.47. Below, I'm copying two sample programs. They simply synchronize with one another using a pair of semaphores. To run them, start the first program and then the second. They will synchronize with each other 100k times and then the second program will spit out the elapsed time. You can restart the second program to run again without bringing down the first program. On Windows, the elapsed time the first time around is anywhere from 20 - 40 seconds. If you leave the second program running, the following run will be only about 2-3 seconds. A third run it will be back up to 20-40 seconds, and it will keep bouncing back and forth like that. On Linux, the total elapsed time is consistently less than half a second. I'm guessing the Windows implementation isn't as efficient, but I wouldn't expect it to be 100 times slower than Linux. The other perplexing thing is how wildly different the results are on Windows from run to run. =====================Test1.cpp========================================== #include <boost/date_time/posix_time/posix_time.hpp> using namespace boost::posix_time; #include <boost/interprocess/sync/named_semaphore.hpp> using namespace boost::interprocess; #include <iostream> using namespace std; int main() { named_semaphore::remove("sem1"); named_semaphore::remove("sem2"); named_semaphore sem1(create_only_t(), "sem1", 0); named_semaphore sem2(create_only_t(), "sem2", 0); while(true) { sem1.wait(); sem2.post(); } return 0; } =====================Test2.cpp========================================== #include <boost/interprocess/sync/named_semaphore.hpp> using namespace boost::interprocess; #include <boost/date_time/posix_time/posix_time.hpp> using namespace boost::posix_time; #include <iostream> using namespace std; int main() { const size_t iterations = 100000; named_semaphore sem1(open_only_t(), "sem1"); named_semaphore sem2(open_only_t(), "sem2"); ptime start = boost::posix_time::microsec_clock::local_time(); for(size_t i = 0; i < iterations; ++i) { sem1.post(); sem2.wait(); } ptime end = boost::posix_time::microsec_clock::local_time(); time_duration delta = end - start; double seconds = double(delta.total_nanoseconds())/1000000000.0; cout << "Total elapsed time: " << seconds << endl;; cout << "Time per iteration: " << (seconds/double(iterations)) << endl; return 0; }

1 0

How to build Boost without any libraries
by Spain, Dr. Jeffry A. 12 Feb '12

12 Feb '12

I would like to be able to build Boost 1.48.0 and install the header-only libraries without any of the libraries that require separate build and installation steps. './bootstrap.sh --without-libraries=all' doesn't work. I don't see this discussed in the getting started guide, or perhaps I am just not picking up on something obvious. Thanks for your suggestions. Jeffry A. Spain Network Administrator Cincinnati Country Day School

2 1

Boost timer library : formatting a cpu_times value always prints zeros ?
by Jean-Francois Romang 12 Feb '12

12 Feb '12

Hello, I'm trying to use the boost::timer library, but I can't get to 'format' function to work : it always prints zero values. I followed the documentation : http://www.boost.org/doc/libs/1_48_0/libs/timer/doc/cpu_timers.html#Default… Here is my code : { boost::timer::cpu_timer cput; someFunctionTakingTime(); std::cout<<cput.elapsed().user/1000000000.0<<std::endl; std::cout<<cput.format(3,"%ws\n")<<std::endl; } Here is the result I get : 0.390002 0.000s Using auto_cpu_timer does not help. Is there something wring in my code ?

1 0

Tokenizer Vs. String Algos
by asif saeed 12 Feb '12

12 Feb '12

Hi, I haven't looked into the code but I wonder if just using string algos would be faster than using Tokenizer in that I can customize. Best regards, Asif

1 0

Tokenizer / offset_separator
by asif saeed 12 Feb '12

12 Feb '12

Hi, I created a class template where the template parameter, which is the size of the array, is used to create a static array of offsets (for offset_separator) in a static function (offsets()) returning a const reference to that local static array. I then call create offset_separator like this: offset_separator sep(offsets(), offsets+T); However, if I have observed it correctly, offset_separator (1.47.0) seems to copy the sequence of offsets into an internal vector. What's wrong with just using the passed array? Best regards, Asif

1 0

[uBLAS] Strange bug with nvcc
by Chris Jewell 12 Feb '12

12 Feb '12

Hi all, I have a weird bug with the CUDA compiler nvcc (v. 4.1, Linux) and Boost::uBLAS (1.48.0). I need to copy data out of uBLAS vectors/matrices into GPU space, hence I need to compile a file containing Boost::ublas structures as well as CUDA code -- hence compiling Boost::uBLAS with nvcc. I am including ublas/operation.hpp, ublas/matrix.hpp, ublas/matrix_sparse.hpp, ublas/matrix_proxy.hpp, ublas/vector.hpp, and ublas/io.hpp. The underlying version of GCC is 4.3.2 on a Suse Linux platform. If I comment out all the CUDA code and compile with raw g++, then everything is fine. However, if I compile with nvcc, then I get many errors like the following: /home/stats/stsiab/include/boost/numeric/ublas/vector_sparse.hpp: In member function ‘const T* boost::numeric::ublas::mapped_vector<T, A>::find_element(typename A::size_type) const’: /home/stats/stsiab/include/boost/numeric/ublas/vector_sparse.hpp:390: error: ‘__T13’ has not been declared What is "__T13"? grep on the Boost source doesn't find it, so I assume that I'm missing a header from somewhere else? The other odd thing is that there is no reference to my CUDA file. Furthermore, everything works well on my MacBook Pro with GCC 4.2.1 and nvcc 4.1. Any ideas? Thanks, Chris -- Dr Chris Jewell Department of Statistics University of Warwick Coventry CV4 7AL UK Tel: +44 (0)24 7615 0778

2 2

std:string to boost::filesystem::path, then, create directories.
by JIA Pei 11 Feb '12

11 Feb '12

Hi, all: After upgrading to boost 1.48.0, my code seems to stopping working. The original code is pretty simple: char* folderName = "./test"; if (!boost::filesystem::is_directory( folderName ) ) boost::filesystem::create_directories( folderName ); It used to work fine but after upgrading to boost 1.48.0, I got the following annoying errors always: Unhandled exception at 0x51d8ee9e (msvcr90d.dll) in myTest.exe: 0xC0000005: Access violation writing location 0xcdcdcdcd. The program stops at the line "if (!boost::filesystem::is_directory( folderName ) )", and the call stack seems to hint it's an error coming from memcpy.asm " TrailUp1: mov al,[esi] ;U - get byte from source ;V - spare mov [edi],al ;U - put byte in destination mov eax,[dst] ;V - return pointer to destination pop esi ;U - restore esi pop edi ;V - restore edi M_EXIT " Can anybody help me out? Thank you very much.... Best Regards Pei -- Pei JIA Email: jp4work(a)gmail.com cell: +1 604-362-5816 Welcome to Vision Open http://www.visionopen.com

3 5

[asio] compiling on cygwin
by Red Taza 11 Feb '12

11 Feb '12

Hi,I have a program that works fine on Solaris and Linux and I want to port it to work to on Cygwin, but am having issues getting Boost.asio to compile.The problem is it wants me to set some defines which I believe makes it use the native Windows socket interface (winsock) where as I want it to use the Cygwin socket interface because the program uses other libraries that expect the Cygwin socket interface. I want to be able to compile a program like this: $ cat prog.cc #include <net-snmp/net-snmp-config.h>#include <net-snmp/net-snmp-includes.h>#include <net-snmp/agent/net-snmp-agent-includes.h>#include <boost/asio.hpp> int main() {} $ To get the above code to compile when only including boost/asio.hpp I have to use the following: g++ -D_WIN32_WINNT=0x0501 -D__USE_W32_SOCKETS \ -I$BOOST_ROOT/include -I$NETSNMP_HOME/include \ prog.cc \ -L$BOOST_ROOT/lib -lboost_system -lws2_32 But the net-snmp headers fail when added, where as when I remove the include for the asio header file and remove -D_WIN32_WINNT=0x0501, -D__USE_W32_SOCKETS and -lws2_32 from the compile line then the net-snmp headers compile fine - but the asio will fail.g++ -I$BOOST_ROOT/include -I$NETSNMP_HOME/include prog.cc -L$BOOST_ROOT/lib -lboost_system Is there a define I can set so that boost.asio uses the Cygwin sockets? I would have thought that this would be the expected behaviour when compiling on Cygwin. The version of Cygwin is:$ uname -aCYGWIN_NT-6.0 toshiba-PC 1.7.10(0.259/5/3) 2012-02-05 12:36 i686 Cygwin The version of boost is:boost_1_48_0 Thanks

1 0