- Boost-users - lists.boost.org

[1.58.0] Release candidates available
by Marshall Clow 06 Apr '15

06 Apr '15

Release candidate files for 1.58.0 are available at: http://boost.cowic.de/rc/ As always, the release managers would appreciate it if you download the candidate of your choice and give building it a try. Please report both success and failure, and anything else that is noteworthy. This helps ensure the candidates build OK before we push them out to SourceForge. The files (and associated md5s) are: MD5 (boost_1_58_0_rc1.7z) = 609ea035fd3b9f1da110f640c9b8d1e6 MD5 (boost_1_58_0_rc1.tar.bz2) = 7edf680e2532a0370bce64e8e847d661 MD5 (boost_1_58_0_rc1.tar.gz) = 0e435db6a5963e58ffa8d7df49ab5846 MD5 (boost_1_58_0_rc1.zip) = a8d1761306a2ea020e92de53a5cae82d Thanks! -- The release managers P.S. Release notes are here: http://www.boost.org/users/history/version_1_58_0.html

4 5

Call for Papers for Meeting C++ 2015 ends on Wednesday
by Jens Weller 05 Apr '15

05 Apr '15

Hello Boost & Qt Community, the call for papers for this years Meeting C++ Conference got extended and ends on Wednesday(April 8th) next week! Especially talks about libraries have still a good chance to get voted into the conference, and also the embedded theme track! CfP announcement: http://meetingcpp.com/index.php/newsreader/items/call-for-papers-for-meetin… The voting will begin shortly after the end of the CfP and run ~10 days, this will give me guidance to select the right talks for this years conference! thanks, Jens Weller

1 0

error: Name clash for '<p/home/nr/dev/dist/boost/lib>libboost_system.a'
by Nathan Ridge 04 Apr '15

04 Apr '15

Hi, I just downloaded the boost 1.58 beta. Tried to build with: ./b2 --with-filesystem --with-system --with-atomic --with-chrono --with-date_time --with-program_options --with-regex --with-test --with-thread --with-random --with-context --with-coroutine variant=release link=static runtime-link=static threading=multi toolset=gcc cxxflags=-Wno-unused-local-typedefs install Got the following error: Performing configuration checks - 32-bit : no - 64-bit : yes - arm : no - mips1 : no - power : no - sparc : no - x86 : yes - lockfree boost::atomic_flag : yes - lockfree boost::atomic_flag : yes - has_icu builds : no error: Name clash for '<p/home/nr/dev/dist/boost/lib>libboost_system.a' error: error: Tried to build the target twice, with property sets having error: these incompabile properties: error: error: - none error: - <address-model>64 <architecture>x86 error: error: Please make sure to have consistent requirements for these error: properties everywhere in your project, especially for install error: targets. Some googling reveals that people suggest adding '--without-context --without-coroutine' to work around the problem, but this is not an option for me because I use these libraries! What can I do to fix this? Thanks, Nate

2 2

[polygon] Voronoi diagram: weird results
by Jeremy Murphy 03 Apr '15

03 Apr '15

Hi, I started using the voronoi_diagram functionality recently and am getting some weird results that I need help interpreting. Here is an example: https://drive.google.com/open?id=0B9-L1xN12iKVYUlMb1hLY0FEMWs&authuser=0 What you can see here is a salmon (?) coloured layer of polygons (input) underneath a semi-transparent layer of pale blue polygons (output) and a boundary. So you can see that some of the output looks 'good', as in, regular and usable, but on the right-hand side there is some obvious 'bad' results that I cannot completely explain. The missing polygons are almost certainly because I have culled any polygons from the output that included points outside the boundary. But there are also appears to be some 'bad' polygons in the output that overlap 'good' ones. So I guess my initial question is: is this a typical symptom of something that other users of the voronoi_diagram recognize? I'm using Boost.Geometry for all other geometric processing. Thanks, cheers. Jeremy

1 2

FW: [log] Is there a way to copy the initial log of a rolling log set
by Marc F. Houde 02 Apr '15

02 Apr '15

Greetings, I am using boost::log with a text_file_backend sink with file rotation for logging on a long running application. I have set it up to clear earlier logs when the logging directory reaches a certain size, but I would like to preserve the initial log as it contains useful configuration information that may not be repeated in later logs. I thought I might be able to do this using set_close_handler, and making a copy but there does not appear to be a way to get the actual file information. Thanks, Marc [ASG]<http://www.asg.com/> Marc Houde, Sr Software Engineer marc.houde(a)asg.com<mailto:marc.houde@asg.com> | 703.464.1624 www.asg.com<http://www.asg.com/> | 800.333.8666 USA Only Facebook<https://www.facebook.com/ASGSoftwareSolutions> | LinkedIn<https://www.linkedin.com/company/asg> | Twitter<https://twitter.com/#!/asgsoftware> | YouTube<https://www.youtube.com/asgtech>

1 0

[boost][units] reciprocal conversion / inter unit conversion
by Matthias Wohlgemuth 02 Apr '15

02 Apr '15

Dear users, I just started to use boost::units for some simulations. However, I have some problems with unit conversions and I hope someone can help me. Let's say I have a quantity given in terms of energy in "electron volts" (eV) with the unit "energy". I defined the unit the following: BOOST_UNITS_DEFINE_BASE_UNIT_WITH_CONVERSIONS(si, electron_volts, "electron volts", "eV", PHYSICAL_CONSTANT_DVALUE(e), si::energy, 5000); namespace si { typedef make_system<electron_volts_base_unit>::type electron_volt_system; typedef unit<energy_dimension,si::electron_volt_system> electron_volt_type; BOOST_UNITS_STATIC_CONSTANT(eV,electron_volt_type); } where PHYSICAL_CONSTANT_DVALUE(e) gives me the double representation of si::constants::codata::e.value(). This works fine, for conversions like eV=>Joule etc. (By the way, is there a more straight forward way, to define non-rational scaled units, than defining a new unit sytem?) Now I want to convert eV to nanometers (wavelength) This is defined as: wavelength = c/frequency = h*c/energy (c=speed of light, h = Planck's constant) I would again define a system with the base unit of wavelength and the unit wavelength_type initialized e.g. with wavelength_nanometer. First Problem: What would be the derived dimension/unit: energy, length or a new dimension like wavelength_dimenson? If I would define it as something else than energy, I will have problems when using quantity<si::electron_volt_type>(x*si::wavelength_nanometer) or quantity<si::nanometer_type>(x*si::eV) Second problem: The relationship is reciprocal: wavelength ~ 1/energy. How do I define a proper conversion? Thanks for your help! Cheers, Matthias

1 0

Cross compiling boost 1.55 static libs
by Pham Ngoc Hai 01 Apr '15

01 Apr '15

Hi guys, I'm trying to cross compile boost for armhf but some of the functions seem to be missed.I'm using croos-tool ng, gcc-4.8 =================================For example in file system, boost::filesystem::detail::status is not there: nm stage/lib/linux_armv7/libboost_filesystem.a | grep status [returns nothing] ================================= I take a look on an existing ubuntu.14.04 armhf nm libboost_filesystem.a | grep status 00000048 t _ZN12_GLOBAL__N_113dir_itr_firstERPvS1_PKcRSsRN5boost10filesystem11file_statusES8_.isra.29.constprop.49 00001cf4 T _ZN5boost10filesystem6detail14symlink_statusERKNS0_4pathEPNS_6system10error_codeE 000019c0 T _ZN5boost10filesystem6detail6statusERKNS0_4pathEPNS_6system10error_codeE 00001b54 T _ZNK5boost10filesystem15directory_entry12m_get_statusEPNS_6system10error_codeE 00001e98 T _ZNK5boost10filesystem15directory_entry20m_get_symlink_statusEPNS_6system10error_codeE ================================= my project-config.jam: # Boost.Build Configuration # Automatically generated by bootstrap.sh import option ; import feature ; # Compiler configuration. This definition will be used unless # you already have defined some toolsets in your user-config.jam # file. #if ! gcc in [ feature.values <toolset> ] #{ using gcc : armhf : arm-unknown-linux-gnueabi-g++ : <compileflags>-fexpensive-optimizations <compileflags>-O3 <compileflags>-march=armv7-a <compileflags>-mtune=cortex-a9 <compileflags>-mfpu=neon <compileflags>-mfloat-abi=hard <compileflags>-ftree-vectorize <compileflags>-ffast-math <compileflags>-funsafe-math-optimizations <compileflags>-pipe <compileflags>-fwhole-program <compileflags>-fpermissive <compileflags>-fPIC ; #} project : default-build <toolset>gcc ; # Python configuration using python : 2.7 : ; # List of --with-<library> and --without-<library> # options. If left empty, all libraries will be built. # Options specified on the command line completely # override this variable. libraries = ; # These settings are equivivalent to corresponding command-line # options. option.set prefix : /usr/local ; option.set exec-prefix : /usr/local ; option.set libdir : /usr/local/lib ; option.set includedir : /usr/local/include ; # Stop on first error option.set keep-going : false ; ================================= I use this patch to fix the zlib problem in boost_1.55 cat no_zlib.diff Index: libs/iostreams/build/Jamfile.v2 --- libs/iostreams/build/Jamfile.v2 (revision 6521) +++ libs/iostreams/build/Jamfile.v2 (working copy) @@ -159,8 +159,6 @@ : $(sources) : <link>shared:<define>BOOST_IOSTREAMS_DYN_LINK=1 <define>BOOST_IOSTREAMS_USE_DEPRECATED - [ ac.check-library /zlib//zlib : <library>/zlib//zlib - <source>zlib.cpp <source>gzip.cpp ] : : <link>shared:<define>BOOST_IOSTREAMS_DYN_LINK=1 ; ================================= my b2 command: ./b2 link=static threading=multi --tool-set=gcc-armhf --without-python -sNO_COMPRESSION=1

2 2

[Graph] Reserving graph memory (again)
by Sensei 01 Apr '15

01 Apr '15

Dear all, Reading all the documentation online about memory reservation in a graph, I am still wondering if it would be possible to reserve memory in advance. With a vecS container it would be quite an advantage to reserve memory, but let me be clear, I am not saying that I know the number of edges in advance, so CRS isn't an option. Until now I know of two ways to do this. The first is to add & remove an edge: add_edge(0, num_vertices(g) + list.size() - 1, g); remove_edge(0, num_vertices(g) + list.size() - 1, g); however, it won't work in my context since I could (as you see from the code) add edges to an existing graph, and I am getting weird results. I am sure this is due to my code, however, add/remove isn't really a neat solution. The second option is weird: g.m_vertices.resize(list.size() - 1); This could be used, but m_* members should, as far as I understand, be private or protected. Is there any other nice possibility? For instance, a static function in the traits that reserves memory? In case, why this cannot be done by BGL APIs? Thanks!

1 0

[asio]
by georg＠schorsch-tech.de 31 Mar '15

31 Mar '15

Hi all, currently i try to understand boost asio a little better. Right now, i just had a very simple asio program and started 2 timers asyncronously. i expected the timer t to complete after the timer i. //-------<----------------- Current output: 10 s. 1 s. //-------<----------------- Expected output: 1 s. 10 s. //-------<----------------- Shouldn't complete the async_wait with 1 sec complete prior the 10 sec async_wait? I am on Windows 7 x64/ MSVC 2013(VS12) / Boost 1.57.0. Regards Georg //-------<----------------- #include <boost/asio.hpp> #include <boost/system/error_code.hpp> #include <iostream> void wait_handler(const boost::system::error_code &ec) { if (!ec) std::cout << "10 s." << std::endl; } void irq_handler(const boost::system::error_code &ec) { if (!ec) std::cout << "1 s." << std::endl; } int main() { boost::asio::io_service io_service; boost::asio::deadline_timer t(io_service, boost::posix_time::seconds(10)); boost::asio::deadline_timer i(io_service, boost::posix_time::seconds(1)); t.async_wait(&wait_handler); t.async_wait(&irq_handler); io_service.run(); } //-------<-----------------

3 2

Serialization cumulatively.
by Tony Camuso 30 Mar '15

30 Mar '15

I'm trying to serialize the data from multiple passes of an app on various files. Each pass generates data that must be serialized. If I simply append each serialization, then deserialization will only occur for the first instance, since the number of records read by the deserialization will only be for the first instance. What I'm doing is deserializing on each new pass, deleting the original file, and then serializing everything again with the new information. If there were only a few files to process, this would not be a problem. However there are thousands of files. Additionally, on each new pass, I am checking to see if a certain type of record has already been saved. So, with every pass, I must look up in a deeper and deeper database. Currently, it's taking almost an hour to process about 3000 files, with an average of 55,000 lines per file. It is a huge amount of data. However, I'm looking for a way to reduce the length of time it takes to do this processing. Does anybody have a better idea than to cycle through the serialize-deserialize-lookup-serialize sequence for each file?

5 14