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

does boost::variant support 128 bit integer?
by David Frank 09 Apr '21

09 Apr '21

especially the gnu __int128?Thanks,David.

6 16

Multiprecision
by shady mohamed 07 Apr '21

07 Apr '21

I need to understand the competency test of multiprecision project

1 0

select_stdlib_config.hpp and <version> 1.73
by Allen Sanderson 05 Apr '21

05 Apr '21

Hello, A casual boost user and abuser here. Some specifics on my OS OS X Big Sur 11.2 Xcode 12.4 Compiler clang : Apple land version 12.0.0 Boost 1.73 - forward When select_stdlib_config.hpp is included it gets to this initial code to include the file “version" ... #if defined(__cplusplus) && defined(__has_include) # if __has_include(<version>) // It should be safe to include `<version>` when it is present without checking // the actual C++ language version as it consists solely of macro definitions. // [version.syn] p1: The header <version> supplies implementation-dependent // information about the C++ standard library (e.g., version number and release date). # include <version> ... The issue is that many projects have a file “VERSION” in their top level src directory. On a traditional Linux system the case sensitivity would make sure the right file is included. However, many/most OS X file systems are case insensitive. As such, the file VERSION gets included instead. And does not compile. I have no solution to the issue but wanted to raise it. Cheers, Allen Sanderson SCI Institute University of Utah

1 0

std::any serialization
by Isaac Lascasas 04 Apr '21

04 Apr '21

Hello, I wonder if there is a workaround to serialize std::any objects. I want to replace some old code that had similar functionality and all the types were registered via BOOST_CLASS_EXPORT_GUID. Regards.

1 0

Tried to unzip a file , and followed the example in the boost.org as well, but it doesn't work
by asfskfdjkj kfasjfkasjf 03 Apr '21

03 Apr '21

Hi guys I tried to unzip a file with code below which exactly like the example from boost.org, but it doesn't work the example is in the link which is Zlib Filters | | | | Zlib Filters | | | I tried the code like the below try{ ifstream file("d:\\202012303333629003180526491.txt.zip", ios_base::in | ios_base::binary); boost::iostreams::filtering_streambuf<boost::iostreams::input> in; in.push(boost::iostreams::zlib_decompressor()); in.push(file); boost::iostreams::copy(in, cout); return 0;}catch (const std::exception& error) { std::cerr << error.what() << std::endl; return (EXIT_FAILURE); } it's not working, it get exception when it excuted boost::iostreams::copy(in, cout); I got the zlib error: iostream stream error exception thoughwhat wrong I have done though? thank you

2 2

[Spirit] [Fusion] Concise way to construct attribute object
by Phil Endecott 28 Mar '21

28 Mar '21

Dear Experts, I currently have something like: struct S { string name; string value; }; BOOST_FUSION_ADAPT_STRUCT( S, name, value ) using S_rule = qi::rule<ITER, skipper, S()>; S_rule r %= identifier >> lit(':') >> identifier; Note that I can use %= here and I don't need to write an explicit semantic action to assign to S. Now say I modify S: class S { public: S(string name_, string value_); // May be default-constructable, copyable, assignable etc. // - but name and value are not individually assignable. }; Question: what is the best way to use that with the qi parser rule? I was hoping to find something in Fusion that would allow me to adapt the new class so that it can be used with %= as before, but I don't see anything appropriate. Currently the best that I can think of is: S_rule r = ( identifier >> lit(':') >> identifier ) [ _val = boost::phoenix::construct<S>(_1,_2) ]; How can I improve on that? Thanks, Phil.

1 0

[Core] scoped_enum and template arguments
by Alan Barr 24 Mar '21

24 Mar '21

Hi, I'm a little confused about the following statement in the docs for scoped enum: > The emulated scoped enum can not be used in switch nor in > template arguments. https://www.boost.org/doc/libs/1_75_0/libs/core/doc/html/core/scoped_enum.h… I understand the rationale for the switch statement part (and indeed it won't compile in C++03). However, I'm struggling to grasp the template argument part (and I'll be the first to admit I'm not particularly strong with templates). Unlike the switch case, any basic scenario I've thrown at it compiles. So I'm not sure if this is a case of "should not" vs "can not". When would using a "scoped_enum_type" in a template argument as opposed to "BOOST_SCOPED_ENUM_NATIVE(scoped_enum_type)" become a problem? Or, why should I use "boost::optional<BOOST_SCOPED_ENUM_NATIVE(future_errc)>" over "boost::optional<future_errc>" ? I've got a compiler explorer link with some test code here if it helps: https://godbolt.org/z/dfr6M6vrW Regards, Alan

1 0

Boost 64bit in Embarcadero C++ Builder
by Michael Morelli 17 Mar '21

17 Mar '21

Hi folks, we are testing the Boost.Test-Framework (installed with GetIt-Package Boost). We use C++ Builder 10.4.1 with Boost 1.70.0. We have created a new C++ Console-Application (with VCL enabled). Under Project->Options we have set '_TCHAR as char', because Boost.Test seems to need this mapping. It is not working with ' _TCHAR as wchar_t' The application main-entry-function is imported via #include <boost/test/included/unit_test.hpp>. The application then is run with the params : --log_level=all --report_level=detailed --build_info=true The console-output seems to be fuzzy : 1. Why is the platform win32? We use target-platform win64 and have installed boost 64bit via getit... 2. Paths are outputted with '\' and '/' 3. The fatal-error is correct but the consoleoutput 'St9exception' seems not correctly interpreted (charset ?) Are we doing something wrong or do you have some tipps regarding this ? Thx a lot. This is the console-output : [cid:image001.png@01D71B23.032439B0] Kind regards, Michael Morelli

1 0

[release] Boost 1.76.0 beta 1 released
by Marshall Clow 16 Mar '21

16 Mar '21

Boost release 1.76.0 beta 1 is now available at: <https://dl.bintray.com/boostorg/beta/1.76.0.beta1/source/> The SHA256 checksums are as follows: 39eecd9672e78b95c8e3fdec1430a63788d528a66d3c8799c4639238e5d785bd boost_1_76_0_b1.7z c32edd34f767dd4bb67b86f76ebf58cf76822a5993a9b434c028b2dd08cfeefc boost_1_76_0_b1.zip 3ec1f92fa0a159b7073e39dd88f527a55d1f35e8986a3c560f202a75bdb8d584 boost_1_76_0_b1.tar.gz eb098d6d5352e10b17a3c49d2927ed61f09b6544f7c534627de35dfa2560fe20 boost_1_76_0_b1.tar.bz2 For details of what's in the release, see <https://www.boost.org/users/history/version_1_76_0.html>. Please download the beta, give it a try, and report any problems you encounter. Thanks, -- The Boost Release Team

1 0

[release] Boost 1.76.0 Beta 1 Release Candidate 2
by Marshall Clow 13 Mar '21

13 Mar '21

The second release candidates for the 1.76.0 beta release are now available at: <https://dl.bintray.com/boostorg/beta/1.76.0.beta1/source/> The SHA256 checksums are as follows: c32edd34f767dd4bb67b86f76ebf58cf76822a5993a9b434c028b2dd08cfeefc boost_1_76_0_b1_rc2.zip eb098d6d5352e10b17a3c49d2927ed61f09b6544f7c534627de35dfa2560fe20 boost_1_76_0_b1_rc2.tar.bz2 39eecd9672e78b95c8e3fdec1430a63788d528a66d3c8799c4639238e5d785bd boost_1_76_0_b1_rc2.7z 3ec1f92fa0a159b7073e39dd88f527a55d1f35e8986a3c560f202a75bdb8d584 boost_1_76_0_b1_rc2.tar.gz The only changes from RC1 is an update to Boost.Interprocess, which was failing to compile on Mac OS X. 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. -- The Release managers

3 3

[release] Boost 1.76.0 Beta 1 Release Candidate 1
by Marshall Clow 11 Mar '21

11 Mar '21

The first release candidates for the 1.76.0 beta release are now available at: <https://dl.bintray.com/boostorg/beta/1.76.0.beta1/source/> The SHA256 checksums are as follows: a222d325cd3c9f66951ec9026eca6ff6f48dd563c9a025911d598d2aca8770ce boost_1_76_0_b1_rc1.tar.bz2 6b90b510a3de38fc30a071ac798da37545fb60a1d6b21aed1130b47c9c598977 boost_1_76_0_b1_rc1.zip b01627bd9ce934b092037da81696bb4482a44e6a31c8791dbed82eafe04b69fa boost_1_76_0_b1_rc1.tar.gz dbef57adbab1d4395d14000f0a2420fc82f87476e10d48359ceff7125a74b77a boost_1_76_0_b1_rc1.7z 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. -- The Release managers

1 3

(no subject)
by Jinlong Shen 11 Mar '21

11 Mar '21

Dear Madam/Sir, I used the boost::statechart asynchronous_state_machine to create a simple state machine. The state machine, the events and states are regular classes and are not class templates. Everything builds and works well on linux and qnx platform. However, on windows I got message that the the initial state is undefined. I think I have dealt with forward declaration well since it is build on linux and qnx. On Windows microsoft this compiler is used: Microsoft (R) C/C++ Optimizing Compiler Version 19.00.24215.1 for x64 The error message is like this. (ClCompile target) -> C:\Program Files\boost_1_64_0\boost/statechart/state_machine.hpp(701): error C2027: use of undefined type 'xxxx::FirstState' C:\Program Files\boost_1_64_0\boost/statechart/state_machine.hpp(701): error C3861: 'initial_deep_construct': identifier not found Any hint how it could be wrong with Microsoft C++? Best regards,Jinlong Shen

2 1

Boost accepted in Google Summer of Code 2021
by David Bellot 09 Mar '21

09 Mar '21

We're in again! https://github.com/boostorg/wiki/wiki/Google-Summer-of-Code%3A-2021 David

1 0

MSM and multithreading. Not clear what pattern to follow.
by Andrea Fiorito 08 Mar '21

08 Mar '21

Hi, I'm trying to use MSM to implement a client server component for my app. The client is broadcasting a msg, that servers in a local network will detect and respond. The client at that point will connect to one of the servers and interact following the network protocol. For the network and threading I'm using boost ASIO. I have 2 submachines that manage one the discovering part, and the other the connection and dialogue with one server part. I found that would be beneficial to run the submachines in one thread managed by an io_context, with the idea of stopping that context to cancel all the asynchronous operations. Unfortunately there are some weird behaviours with submachines, and not getting if I missed the right way of doing things or else. the description of my last stopper is here: https://github.com/boostorg/msm/issues/37 Is anyone able to point me on the right direction? Regards, Andrea

2 1

CMake's PCH support and boost.test linker error
by Olaf Peter 04 Mar '21

04 Mar '21

Hello Boost users I have here a strange problem which occurs only if I try to build with precompiled headers using CMake. I've configured the project for use with dynamically linking. The error is: /usr/bin/ld: /usr/lib/gcc/x86_64-redhat-linux/10/../../../../lib64/crt1.o: in function `_start': (.text+0x24): undefined reference to `main' Something obvious? Thanks in advance The project's reduce is: ---8<--- test_driver.cpp --- 8< ---- #defineBOOST_TEST_MODULE "My Test Suite" #include<boost/test/unit_test.hpp> --->8------>8------>8--- ---8<--- test_x.cpp --- 8< ---- #include<boost/test/unit_test.hpp> .... --->8------>8------>8--- ---8<--- src/CMakeLists.txt --- 8< ---- ... set_source_files_properties(src/test_xxx.cpp APPEND PROPERTIES COMPILE_DEFINITIONS "BOOST_TEST_DYN_LINK" ) .... --->8------>8------>8------>8------>8------>8--- ---8<--- pch.pp --- 8< ---- #pragma once #include<boost/test/unit_test.hpp> --->8------>8------>8---

1 2

How should a "using zlib ..." jam statement look, for Boost to link to ZLib in the form of SO filename (libz.so) instead of by -lz?
by Bob 03 Mar '21

03 Mar '21

Hello Boost mailing list, I am building a code project in my user home directory on OpenBSD, and doing this involves both a local build of Boost and also a local build of ZLib. I do the whole build using GCC. This is a relatively ordinary build use case, however GCC on OpenBSD (at least in its current version) has a peculiarity to it, which is that it adds to LD the argument "-L/usr/lib" in the leading position, and this means that LD's search for libz.so will find and use the operating system's bundled ZLib binary which is /usr/lib/libz.so ALWAYS, and NEVER use my local libz.so . Can you help me understand if Boost can be customized to link to ZLib in the following way: I can force GCC/LD to use a particular libz.so file, by specifying it by filename, e.g. "gcc ... libz.so ...", "gcc ... /my/dir/libz.so ...", instead of by the conventional "-lz". Boost's build process is hardwired to "-lz" and I ask you to help me understand if there is a way to circumvent that. ZLIB_LIBRARY_PATH will not make "-lz" hit anything else than /usr/lib/libz.so , per the compiler's default -L/usr/lib in leading position, so it will not be of help to solve my problem. To my best awareness ZLIB_INCLUDE has nothing to do with linking, so it will not be of help to solve my problem. ZLIB_NAME would only be to customize the library name would it e.g. to "-lmycustomz", or what is it for? Please clarify so I could figure if it could be of help, though ideally i keep my local ZLib having its default name "libz.so". And this brings me to the path of solving the problem that I imagine could work, but that I don't understand fully from the Boost documentation: There is the possibility to customize ZLib build settings through the JAM configuration, that is user-config.jam . The syntax used is "using zlib ...". Can you please help me understand if this route works, if so how would that "using zlib ..." statement look, such that Boost's build process will NOT build ZLib itself, and then will NOT link to it via "-lz", but will instead link to it by "$MYCUSTOMBUILDDIR/libz.so"?

1 0

Boost.Python + `RecursionError` during deallocation
by Andrew Jones 01 Mar '21

01 Mar '21

Hi, I just raised the following issue on the Python issue tracker: https://bugs.python.org/issue43351 But I'm sharing it here to widen the net ... Basically, when moving from Python 2.7.14 to Python 3.7.5, and when using Boost.Python (Boost version 1.6.9), I'm seeing a crash in finalisation due to: ``` Exception ignored in: <Boost.Python.life_support object at 0x2acf473c5d10> RecursionError: maximum recursion depth exceeded while calling a Python object Fatal Python error: Cannot recover from stack overflow. ``` Everything else with the Python that's running, and the Boost.Python interface, is consistent between 2.7.14 and 3.7.5 -- it is just the Python version that's changing. I should say, Python 3.7.5 does otherwise seem operable, it just falls over when running this certain code (which just walks a linked-list of ~1400 elements via Boost.Python). Has anyone seen anything like this before? Does this list feel this is more likely a Python-proper issue, or a Boost.Python issue? Cheers, Andrew

1 0

[boost::units] Best practices for implementing custom units and quantities
by Alexandru Onea 01 Mar '21

01 Mar '21

Hello community, I am working on a project which works with units such as scale intervals (cents, moria, etc.) depending on the temperament system used (24-EDO, 72-EDO, etc.). My implementation defines a namespace for each tuning system. Within the namespaces I define base units, units and a system, as well as conversion factors between the systems. The base dimension is defined in only one of the namespaces (corresponding to the tuning system which I consider "standard" which is the 1200 cents per octave system). 1. Does such an approach qualify as a good approach in general (leaving aside the musical aspects)? 2. Is it a good practice to place all these namespaces in boost::units (extend boost::units)? Or should I keep them within my own namespaces? Learning the answer to #2 is more interesting to me, even in the absence of an answer to #1. Best regards, Alexandru Onea

2 1

Question about Boost Integer
by Emilio Paolini 28 Feb '21

28 Feb '21

Hi everyone, i'm pretty new to Boost. I need to use as a default type for the CNL. The problem is that i would need an integer type that can assume only 2^4 values (so basically that stays on 4 bits). What i am asking is that if there is a way to define an integer like int128_t but instead of 128 i would need to have 4. Thank you very much for your answers. Emilio

2 2

flyweights, nested flyweights and shared libraries
by Andreas Buykx 25 Feb '21

25 Feb '21

I want to use boost::flyweight to capture some structures containing const data. The documentation states that if flyweights are exported across shared library boundaries that the intermodule holder should be used. Since my flyweights are all created in the same shared library I would prefer to use the static holder but I keep getting crashes. I tried the intermodule holder but I'm having trouble compiling my sources with that. So my thinking was to wrap all flyweight types in accessor classes which are exposed instead, and the accessor classes create the flyweights within the same shared library to store the data. Would that be a valid approach? Also, is it possible to have nested flyweights (when they are wrapped by accessor classes)? And would that cause any additional problems in the above scenario? Below is an example of what I'm trying to achieve... Thanks for your advice! Andreas === Foo.h === struct FooDetails { FooDetails( std::string text_, std::vector<double> values ); std::string text; std::vector<double> values; }; class STUFF_EXPORT Foo { public: Foo( std::string text_, std::vector<double> values ); // rule-of-5 constructors declared, and default-implemented in cpp file const std::string& text() const; const std::vector<double>& values() const; private: boost::flyweight<FooDetails> m_data; }; === Foo.cpp === FooDetails( ( std::string text_, std::vector<double> values_ ) : text(text_), values( values_ ) { } Foo::Foo( std::string text_, std::vector<double> values_ ) m_data( FooDetails( text_, values_ ) ) { } std::string Foo::text() const { return m_data.get().text; } std::vector<double> Foo::values() const { return m_data.get().values; } === Bar.h === struct BarDetails { BarDetails( Foo foo_, std::vector<long> ids_ ); Foo foo; //< Foo holds a flyweight std::vector<long> ids; }; class STUFF_EXPORT Bar { public: Bar( std::string text_, std::vector<double> values_, std::vector<long> ids_ ); // rule-of-5 constructors declared, and default-implemented in cpp file const std::string& text() const; const std::vector<double>& values() const; const std::vector<long>& ids() const; private: boost::flyweight<BarDetails> m_bar; }; === Bar.cpp === BarDetails::BarDetails( Foo foo_, std::vector<long> ids_ ) : foo( foo_ ), ids( ids_ ) { } Bar::Bar( std::string text_, std::vector<double> values_, std::vector<long> ids_ ) : m_bar( Foo( text_, values), ids_ ) { } const std::string& Bar::text() const { return m_bar.get().m_foo.text(); } const std::vector<double>& Bar::values() const { return m_bar.get().foo.values(); } const std::vector<long>& Bar::ids() const { return m_bar.get().ids; }

3 8

Re: [Boost-users] Boost-users Digest, Vol 5456, Issue 1
by philippe DALET 23 Feb '21

23 Feb '21

Hi, I have built some binaries for boost and python https://sourceforge.net/projects/pcpu/files/Boost/ boost 1.71 python3.7&numpy3.7 for ubuntu 20.04 64 bits boost 1.75 python3.789&numpy3.789 for visual studio 2019 mscv14.2 - win10 Ph.DALET Le mar. 23 févr. 2021 à 11:03, <boost-users-request(a)lists.boost.org> a écrit : > Send Boost-users mailing list submissions to > boost-users(a)lists.boost.org > > To subscribe or unsubscribe via the World Wide Web, visit > https://lists.boost.org/mailman/listinfo.cgi/boost-users > or, via email, send a message with subject or body 'help' to > boost-users-request(a)lists.boost.org > > You can reach the person managing the list at > boost-users-owner(a)lists.boost.org > > When replying, please edit your Subject line so it is more specific > than "Re: Contents of Boost-users digest..." > > > Today's Topics: > > 1. Re: Boost cannot find numpy37 on Dockerized Ubuntu 20.04 > (Roy de Bokx) > 2. Re: flyweights, nested flyweights and shared libraries > (Andreas Buykx) > > > ---------------------------------------------------------------------- > > Message: 1 > Date: Mon, 22 Feb 2021 15:49:11 +0100 > From: Roy de Bokx <rdebokx1990(a)gmail.com> > To: boost-users(a)lists.boost.org > Subject: Re: [Boost-users] Boost cannot find numpy37 on Dockerized > Ubuntu 20.04 > Message-ID: > < > CAFLoxwXy8_an1jZh_PQMkjppbOWyixvR5Exw1DCeaUyV9xDfBA(a)mail.gmail.com> > Content-Type: text/plain; charset="utf-8" > > Op vr 19 feb. 2021 om 22:05 schreef Nathan Ernst via Boost-users < > boost-users(a)lists.boost.org>: > > > The missing symbols you quoted come from python itself. It looks like > > you're missing a "-lpython" linker flag. > > > Thanks! That definitely put me on the right track. Eventually I got it > working using the "-lpython3.6m" flag, since I found that the Python3.6 .so > file was situated at /usr/lib/x86_64-linux-gnu/libpython3.6m.so > Many thanks, this is much appreciated! I've pasted the full Dockerfile in > https://github.com/boostorg/boost/issues/462, in case anyone is > interested. > > > > > Regards, > > Nathan > > > > On Fri, Feb 19, 2021, 4:47 AM Roy de Bokx via Boost-users < > > boost-users(a)lists.boost.org> wrote: > > > >> > >> > >> Op wo 17 feb. 2021 om 20:30 schreef Anonymous Maarten < > >> anonymous.maarten(a)gmail.com>: > >> > >>> Thanks for your answer Maarten! > >>>> Please bare with me as I'm a little new to this. Is there some way I > >>>> can change the interpreter that is used for installing boost? Or do > you > >>>> think that adapting the --prefix flag may also work? > >>>> > >>> > >>> When using the pre-built boost packages of a linux distribution > >>> (ubuntu), you're generally stuck with the versions/options that its > >>> packagers have chosen. > >>> A distribution freezes versions to make a more robust experience. > >>> > >>> Adding a `--prefix` will not help because, first of all, it's not a > >>> valid gcc option, and second, ubuntu 18.04 does not provide a > >>> libboost-python-py37 shared/static library. > >>> Do you really need python 3.7? Doesn't 3.6 suffice? > >>> > >> > >> Thanks! This helped me understanding the issue a bit further. I think > 3.6 > >> should suffice indeed, so I tried using 3.6 by removing the 3.7 > >> installation from the Dockerfile and using the -lboost_numpy3 flag > instead. > >> It seems I ran into some linking issue after this. It was able to find > >> the right numpy3.so and libboost_python3-py36.so, but I'm getting a lot > of > >> errors like these: > >> > >> //usr/lib/x86_64-linux-gnu/libboost_numpy3.so: undefined reference to > >> `PyExc_ValueError' > >> > >> //usr/lib/x86_64-linux-gnu/libboost_python3-py36.so.1.65.1: undefined > >> reference to `PyLong_AsLong' > >> > >> //usr/lib/x86_64-linux-gnu/libboost_python3-py36.so.1.65.1: undefined > >> reference to `PyNumber_InPlaceFloorDivide' > >> > >> //usr/lib/x86_64-linux-gnu/libboost_python3-py36.so.1.65.1: undefined > >> reference to `PyBool_Type' > >> etc... > >> > >> I've pasted the full docker file and logs in > >> https://github.com/boostorg/boost/issues/462 > >> Thanks again for any help. > >> > >>> > >>> If you really want to use Boost.Python + python 3.7, you can do 2 > things: > >>> - stay on ubuntu bionic and build boost yourself (or use an alternative > >>> c/c++ package manager, e.g. [conan](https://conan.io/) has a [boost > >>> package](https://conan.io/center/boost)) > >>> - more to a more recent ubuntu release. e.g. ubuntu 20.10 has python > >>> 3.8: > >>> > https://packages.ubuntu.com/groovy/amd64/libboost-python1.71-dev/filelist > >>> > >> I've also given 20.10 a try, however I ran into the same linking issues > >> as mentioned above. > >> > >>> > >>> > >>> > >>> _______________________________________________ > >> Boost-users mailing list > >> Boost-users(a)lists.boost.org > >> https://lists.boost.org/mailman/listinfo.cgi/boost-users > >> > > _______________________________________________ > > Boost-users mailing list > > Boost-users(a)lists.boost.org > > https://lists.boost.org/mailman/listinfo.cgi/boost-users > > >

1 0

Re: [Boost-users] Boost cannot find numpy37 on Dockerized Ubuntu 20.04
by Roy de Bokx 22 Feb '21

22 Feb '21

Op wo 17 feb. 2021 om 20:30 schreef Anonymous Maarten < anonymous.maarten(a)gmail.com>: > Thanks for your answer Maarten! >> Please bare with me as I'm a little new to this. Is there some way I can >> change the interpreter that is used for installing boost? Or do you think >> that adapting the --prefix flag may also work? >> > > When using the pre-built boost packages of a linux distribution (ubuntu), > you're generally stuck with the versions/options that its packagers have > chosen. > A distribution freezes versions to make a more robust experience. > > Adding a `--prefix` will not help because, first of all, it's not a valid > gcc option, and second, ubuntu 18.04 does not provide a > libboost-python-py37 shared/static library. > Do you really need python 3.7? Doesn't 3.6 suffice? > Thanks! This helped me understanding the issue a bit further. I think 3.6 should suffice indeed, so I tried using 3.6 by removing the 3.7 installation from the Dockerfile and using the -lboost_numpy3 flag instead. It seems I ran into some linking issue after this. It was able to find the right numpy3.so and libboost_python3-py36.so, but I'm getting a lot of errors like these: //usr/lib/x86_64-linux-gnu/libboost_numpy3.so: undefined reference to `PyExc_ValueError' //usr/lib/x86_64-linux-gnu/libboost_python3-py36.so.1.65.1: undefined reference to `PyLong_AsLong' //usr/lib/x86_64-linux-gnu/libboost_python3-py36.so.1.65.1: undefined reference to `PyNumber_InPlaceFloorDivide' //usr/lib/x86_64-linux-gnu/libboost_python3-py36.so.1.65.1: undefined reference to `PyBool_Type' etc... I've pasted the full docker file and logs in https://github.com/boostorg/boost/issues/462 Thanks again for any help. > > If you really want to use Boost.Python + python 3.7, you can do 2 things: > - stay on ubuntu bionic and build boost yourself (or use an alternative > c/c++ package manager, e.g. [conan](https://conan.io/) has a [boost > package](https://conan.io/center/boost)) > - more to a more recent ubuntu release. e.g. ubuntu 20.10 has python 3.8: > https://packages.ubuntu.com/groovy/amd64/libboost-python1.71-dev/filelist > I've also given 20.10 a try, however I ran into the same linking issues as mentioned above. > > > >

2 2

Prooblem building software with fstream.hpp
by Mack Wallace 21 Feb '21

21 Feb '21

Hi all. I'm a noob and unfamiliar with boost. It has also been a long time since I have done any C++ programming. In any case, here are the details: I am trying to compile MecnRao/TCP-UDP-Proxy that uses the Boost Library. I am trying to compile this for Windows, ultimately Windows 10, but at the moment trying to compile on my Windows 7 VM. I am using Visual Studio Pro 2019. After trying the latest version of the Boost Library with some errors due to some deprecated members, I bounced around older versions until finding that Version 1.68.0 reduced my compile errors substantially, and of those 20 or so errors, are caused by some typedef problems in fstream.hpp. (I could be wrong, but the errors all seem to be related to fstream_fix.) Specifically, I get the following errors to begin with: 'fstream_fix<std::basic_ifstream<char,std::char_traits<char> > >' is not a variable template 'fstream_fix<std::basic_ofstream<char,std::char_traits<char> > >' is not a variable template This comes from the following lines in the fstream.hpp typedef basic_ifstream<char> ifstream; typedef basic_ofstream<char> ofstream; Where ifstream and ofstream are indicated as being redefined as #define ifstream fstream_fix<std::ifstream> #define ofstream fstream_fix<std::ofstream> Within the program with fstream_fix as template<class T> struct fstream_fix : public T { fstream_fix(){}; template<class T1> fstream_fix(T1 v1){ setlocale(LC_CTYPE, ""); T::open(v1); setlocale(LC_CTYPE, 0); } template<class T1,class T2> fstream_fix(T1 v1,T2 v2){ setlocale(LC_CTYPE, ""); T::open(v1,v2); setlocale(LC_CTYPE, 0); } template<class T1> void open(T1 v1){ setlocale(LC_CTYPE, ""); T::open(v1); setlocale(LC_CTYPE, 0); } template<class T1,class T2> void open(T1 v1,T2 v2){ setlocale(LC_CTYPE, ""); T::open(v1,v2); setlocale(LC_CTYPE, 0); } }; So I'm wondering if this is a newer/different compiler issue? Or the application itself? Or maybe just missing something like uses namespace std. I am hoping someone can help me nip this last problem so I can get this compiled. Thanks, Mack

1 0

boost::asio::read() gets stuck when there's no internet
by Child OfGod 20 Feb '21

20 Feb '21

In TgBot library i use under MSVC++, in tgbot/net/BoostHttpOnlySslClient.cpp, a call to boost::asio::read() does not return when internet is disconnected for a few minutes. I would appreciate any help to get this resolved, i.e. for the read function to return if it fails after some timeout has passed the code block in the tgbot/net/BoostHttpOnlySslClient.cpp file: [code] string BoostHttpOnlySslClient::makeRequest(const Url& url, const vector<HttpReqArg>& args) const { tcp::resolver resolver(_ioService); tcp::resolver::query query(url.host, "443"); ssl::context context(ssl::context::tlsv12_client); context.set_default_verify_paths(); ssl::stream<tcp::socket> socket(_ioService, context); connect(socket.lowest_layer(), resolver.resolve(query)); #ifdef TGBOT_DISABLE_NAGLES_ALGORITHM socket.lowest_layer().set_option(tcp::no_delay(true)); #endif //TGBOT_DISABLE_NAGLES_ALGORITHM #ifdef TGBOT_CHANGE_SOCKET_BUFFER_SIZE #if _arm64 || _WIN64 || __amd64__ || __x86_64__ || __MINGW64__ || __aarch64__ || __powerpc64__ socket.lowest_layer().set_option(socket_base::send_buffer_size(65536)); socket.lowest_layer().set_option(socket_base::receive_buffer_size(65536)); #else //for 32-bit socket.lowest_layer().set_option(socket_base::send_buffer_size(32768)); socket.lowest_layer().set_option(socket_base::receive_buffer_size(32768)); #endif //Processor architecture #endif //TGBOT_CHANGE_SOCKET_BUFFER_SIZE socket.set_verify_mode(ssl::verify_none); socket.set_verify_callback(ssl::rfc2818_verification(url.host)); socket.handshake(ssl::stream<tcp::socket>::client); string requestText = _httpParser.generateRequest(url, args, false); write(socket, buffer(requestText.c_str(), requestText.length())); string response; #ifdef TGBOT_CHANGE_READ_BUFFER_SIZE #if _M_ARM64 || _WIN64 || __amd64__ || __x86_64__ || __MINGW64__ || __aarch64__ || __powerpc64__ char buff[65536]; #else //for 32-bit char buff[32768]; #endif //Processor architecture #else char buff[1024]; #endif //TGBOT_CHANGE_READ_BUFFER_SIZE boost::system::error_code error; while (!error) { std::size_t bytes = read(socket, buffer(buff), error); response += string(buff, bytes); } return _httpParser.extractBody(response); } [/code] Thankyou in advance😀🙏

3 4

(no subject)
by Dk Jack 18 Feb '21

18 Feb '21

Hi, I am using boost 1.58 in my project. Ran into a crash in the boost filesystem library. I've included the relevant stack trace at the end of the email. My code is something like this: boost::filesystem::path fpath(mFilename.c_str()); if (!boost::filesystem::exists(fpath)) { // log a message and return. return; } auto curr_time = boost::filesystem::last_write_time(fpath); <== crash The crash is not consistent and happened randomly. I tested the above code in a small test program with just the above code. I made the input file unreadable etc, even then it was able to successfully read the last write timestamp of the file without any issues. Not sure why this particular call would throw an error. Since it's compiled with -O2, the error code is optimized out in the stack trace. Any ideas? Thanks for your help. Dk. (gdb) bt #0 0x00007f01651ba387 in raise () from /lib64/libc.so.6 #1 0x00007f01651bba78 in abort () from /lib64/libc.so.6 #2 0x00007f0160d087d5 in __gnu_cxx::__verbose_terminate_handler() () from /lib64/libstdc++.so.6 #3 0x00007f0160d06746 in ?? () from /lib64/libstdc++.so.6 #4 0x00007f0160d06773 in std::terminate() () from /lib64/libstdc++.so.6 #5 0x00007f0160d06993 in __cxa_throw () from /lib64/libstdc++.so.6 #6 0x00007f0161171f5f in boost::filesystem::detail::last_write_time(boost::filesystem::path const&, boost::system::error_code*) () from xxx.so #7 0x00007f0161089657 in last_write_time (p=...) at /usr/local/include/boost/filesystem/operations.hpp:594

2 1