December 2018 - Boost-users - lists.preview.boost.org

[interprocess] Determine size of managed_shared_memory
by Kai Benndorf 12 Oct '19

12 Oct '19

Hi all, I want to create a managed_shared_memory in exactly the size I needed. Therefore I do something like this: size = sizeof( common_managed_shared_memory ); size += sizeof( MyClassInSharedMem ); managed_shared_memory segment(create_only, "name", size ); MyClassInSharedMem allocates itself some memory by segment->allocate(). I added this also to the size, but get a boost::interprocess::bad_alloc exception on allocating this additional memory. Are there some magic bytes I forgot to add? Many thanks in advance Kai

3 2

boost::filtered_graph filtering on vertex color... if I only had a colormap
by Brian Davis 19 Jul '19

19 Jul '19

Lets say I have a graph g (of type graph_t) and a color map (boy I wish I had a colormap): typedef boost::property_map<graph_t, boost::vertex_color_t>::type color_map_t; or is it? auto indexmap = boost::get(boost::vertex_index, g); auto colors = boost::make_vector_property_map<boost::default_color_type>(indexmap); or maybe? auto indexmap = boost::get(boost::vertex_index, g); auto colors = boost::make_vector_property_map<boost::vertex_color_t>(indexmap); or possibly some goop found on the internet? boost::make_iterator_property_map(color_map.begin(), boost::get(boost::vertex_index, data_->g), color_map[0]) or other internet goop? typedef boost::property_map<GraphType, boost::vertex_color_t>::type color_map_t; color_map_t colorMap; boost::color_map(colorMap)) or other internet goop?? std::vector<int> colorMap(num_vertices(g), boost::white_color); so confused... (certainly not surprised why bgl is not a part of stl) Can someone help explain the various types and ways of specifying colormaps and the logic for each way and means? and a function (if I could get that far without VS syntax highlighter going bonkers and gobs of compiler errors.. you know the type): make_pretty(g, colormap ) which colors the graph colormap and a filter struct for boost::filtered_graph<Graph, EdgePredicate, VertexPredicate> (where I wonder why if I only want to filter vertices I have to set boost::keep_all for EdgePredicate... sigh... I know the answer... and also know how long it found me to find/figure out boost::keep_all - and if you are looking for boost::keep_all and found it here... your welcome ;-) ) and modifying (probably incorrectly) boost::filtered_graph example to: template <typename ColorMap> struct vertex_color_filter_eq { vertex_color_filter_eq() { } vertex_color_filter_eq(ColorMap colormap/*, boost::default_color_type c*/) : m_colormap(colormap)/*, m_c(c)*/ { } template <typename Vertex> bool operator()(const Vertex& v) const { return 0 < get(m_colormap, v); //return c < get(m_colormap, v); } ColorMap m_colormap; //boost::default_color_type m_c; }; auto indexmap = boost::get(boost::vertex_index, g); auto colors = boost::make_vector_property_map<boost::default_color_type>(indexmap); // some where here I would call make_pretty(g, colors) typedef property_map<graph_t, boost::vertex_color_t>::type ColorMap_t; vertex_color_filter<ColorMap_t> filter(colors); // boost::filtered_graph<graph_t, boost::keep_all, VertexPredicate> boost::filtered_graph<graph_t, boost::keep_all, vertex_color_filter> fg(g, filter); // Later say if I could get it compile and better if it compiled and worked I could move on and auto color = boost::red_color; typedef property_map<graph_t, boost::vertex_color_t>::type ColorMap_t; vertex_color_filter<ColorMap_t> filter(colors, color); boost::filtered_graph<graph_t, boost::keep_all, vertex_color_filter> fg(g, filter); Maybe if I color them yellow, I can then reach the wizard pull back the curtain and ask how to create a color map... If I only had a color map.

4 6

Question on Posix timezone string
by Mats Webjörn/WIKAB 10 Mar '19

10 Mar '19

I'm using Boost 1.66 and use the following code to get a Posix string to set my uCLinux TZ environment variable to get a local time from an UTC time #include <boost/date_time/local_time/tz_database.hpp> #include <boost/date_time/local_time/local_time.hpp> tz_database tz_db; tz_db.load_from_file("./date_time_zonespec.csv"); time_zone_ptr nyc = tz_db.time_zone_from_region("Europe/Stockholm"); auto s = nyc->to_posix_string(); // "CET+01CEST+01,M3.5.0/02:00,M10.5.0/03:00" setenv("TZ", s.c_str(), 1); The odd thing is that my localtime becomes incorrect, and is actually 1 hr wrong on the opposite side of UTC. But, when I read https://www.gnu.org/software/libc/manual/html_node/TZ-Variable.html it says "The offset specifies the time value you must add to the local time to get a Coordinated Universal Time value". Not add to UTC time, which seems like the Boost Posix string assumes !!!!! Which explains why 16:00 UTC give 15:00 CET instead of 17:00 which is the correct value, when using localtime_r(). Am I missing something fundamentally, or what? Regards Mats Webjörn Webjörn IngenjörsKonsult AB Solhemsbackarna 75A SE-163 56 SPÅNGA Sweden

3 9

Boost Test thread safety
by Dan Field 21 Feb '19

21 Feb '19

Hi folks: I'm trying to identify whether or not the boost Test library is thread safe in the sense that multiple threads can simultaneously interact with the library's various macros and API. I'm more interested in a single test being multi-threaded than being able to run multiple test cases simultaneously. In early documentation of the Test library (version 1.58.0 and before) , there's a section in the introduction called "Open issues" that indicates that the library is not thread-safe. However, the documentation after 1.58.0 was substantially re-written, and I don't see any discussion regarding thread safety at all. The question remains. is the Test library thread-safe or not? If there is a definitive statement regarding this issue, please let me know. Thanks in advance!

3 3

[boost.test] Colors
by Udo Steinbach 20 Feb '19

20 Feb '19

Ahoi! I use boost.test and have now updated to boost v1.68 on Windows, working in standard console. Now Test has colored output, YEAH. The result you see in the picture https://www2.pic-upload.de/img/36193998/no_error_detected.png Notice that I don't use the standard color scheme. With set BOOST_TEST_COLOR_OUTPUT=1 text is light green on gray, bad to read. set BOOST_TEST_COLOR_OUTPUT=0 does black on gray. After that the colors are set to my scheme. Now error result with set BOOST_TEST_COLOR_OUTPUT=1: https://www2.pic-upload.de/img/36193997/failure.png Light red on gray, OK to read, but reset to dark red on gray. I have to use the color command to reset the colors. With command line options it's all the same. Hmmm, how useful can colors be, if the standard output (no error detected) is only one short line? Both pics on https://www.pic-upload.de/gal-1197957/om664y/1.html -- Fahrradverkehr in Deutschland: http://radwege.udoline.de/ GPG: A245 F153 0636 6E34 E2F3 E1EB 817A B14D 3E7E 482E

2 1

ASIO: Writing Composed Operations
by Cristian Morales Vega 26 Jan '19

26 Jan '19

Hi, I was looking at https://www.boost.org/doc/libs/develop/libs/beast/doc/html/beast/using_io/w… and came up with a question I can't answer: In https://github.com/boostorg/beast/blob/develop/example/echo-op/echo_op.cpp#… why does it use the AsyncStream associated executor without even looking at the potentially different (right?) Handler associated executor? Regards.

3 13

C++/CLI and >=VS2015
by liong＠gmx.ch 09 Jan '19

09 Jan '19

6 5

ASIO vs Beast DynamicBuffer use
by Cristian Morales Vega 02 Jan '19

02 Jan '19

Hi, I have just noticed that while ASIO passes DynamicBuffers as && (https://www.boost.org/doc/libs/1_68_0/doc/html/boost_asio/reference/async_r…) Beast passes them via reference (https://www.boost.org/doc/libs/1_68_0/libs/beast/doc/html/beast/ref/boost__…) I'm guessing this happens because ASIO thinks you are going to use dynamic_string_buffer/dynamic_vector_buffer as DynamicBuffer, while Beast thinks you are going to use flat_buffer/multi_buffer. In the second case doesn't make sense to give ownership of the DynamicBuffer to the initiation function because then you would not be able to look at the received data when the operation completes. Since ASIO DynamicBuffer implementations are just adaptors over string/vectors that you keep, it makes sense to move those adaptors, you still keep the string/vector with the data. Have I got it right? But this raises the question: what should I use in my own initiating functions? Regards.

2 3

function template with a universal reference to a specific type
by Ireneusz Szcześniak 01 Jan '19

01 Jan '19

Hi, I'm writing to ask a C++ question, not specifically Boost-related, but this list is the best place I know. How can I write a function template, which has a parameter of the universal reference to a specific type? I can write a function template like this: template <typename T> void foo(T &&t) { // The primary template implementation. cout << "Primary: " << __PRETTY_FUNCTION__ << endl; } And so I can call this function with an expression of any value category, and any type, cv-qualified or not, like this: int main() { foo(1); int i = 1; foo(i); const int ci = 1; foo(ci); } In my code I need to provide different implementations of foo for different types (and these types are templated). How can I accomplish that? Best, Irek

5 11

Error: no member named 'GetModuleFileNameA' in namespace 'boost::winapi'; did you mean simply 'GetModuleFileNameA'?
by Osman Zakir 31 Dec '18

31 Dec '18

Hi again, everyone. When trying to build Boost 1.69.0 after having placed my installation of Python 3.7.1 above the one of Python 2.7.15, I got the error mentioned in the title of this email. It says at the end, " ...failed updating 4 targets... ...skipped 8 targets... ...updated 2381 targets... " The project-config.jam file I used is attached to this message.

1 0

Errors with 1.69 when compiling with /clr on MSVC
by Stian Zeljko Vrba 30 Dec '18

30 Dec '18

I get the following warning and error when compiling a C++/CLI file that includes boost 1.69. 1>c:\quinesoftwarebase\include\boost\type_traits\detail\is_member_function_pointer_cxx_11.hpp(114): error C2764: 'C': template parameter not used or deducible in partial specialization 'boost::is_member_function_pointer<Ret(Args...)>' 1>c:\quinesoftwarebase\include\boost\type_traits\detail\is_member_function_pointer_cxx_11.hpp(125): warning C4575: '__vectorcall' incompatible with the '/clr' option: converting to '__stdcall' Is there a workaround for this? -- Stian

3 3

Re: [Boost-users] [cfe-dev] [llvm-dev] Error: ISO C++17 does not allow 'register' storage class specifier [-Wregister], when building Boost 1.69.0
by Osman Zakir 25 Dec '18

25 Dec '18

I'll build using C++14 mode, then. I wonder if Python 3.7.1 has this issue as well? If it doesn't, then maybe Boost should switch over to that version of Python. ________________________________ From: Matthieu Brucher <matthieu.brucher(a)gmail.com> Sent: Tuesday, December 25, 2018 2:56 PM To: Oleg Smolsky Cc: Craig Topper; Osman Zakir; clang developer list; cfe-users(a)lists.llvm.org; llvm-dev(a)lists.llvm.org Subject: Re: [cfe-dev] [llvm-dev] Error: ISO C++17 does not allow 'register' storage class specifier [-Wregister], when building Boost 1.69.0 The issue comes in part form the Python headers that have the keyword. Unfortunately, this also means that there is an issue while building user code. But it's not a Boost code issue for this library, but a Python one. Regards, Matthieu Le mar. 25 déc. 2018 à 02:08, Oleg Smolsky via cfe-dev <cfe-dev(a)lists.llvm.org<mailto:cfe-dev@lists.llvm.org>> a écrit : Yes, this keyword is no longer valid starting from C++17: https://en.cppreference.com/w/cpp/keyword/register I realize that the "why" probably doesn't help you move forth with your problem, so here are a couple suggestions how to get over it: * Build Boost in C++14 mode with -std=c++14 * Cut out "register" from the Boost source and rebuild * Try a newer Boost version, it may have been fixed Oleg. On 2018-12-23 09:26, Craig Topper via cfe-dev wrote: This seems a better question for cfe-dev or cfe-users ~Craig On Sat, Dec 22, 2018 at 12:23 PM Osman Zakir via llvm-dev <llvm-dev(a)lists.llvm.org<mailto:llvm-dev@lists.llvm.org>> wrote: When trying to build Boost using LLVM, I had this error: " error: ISO C++17 does not allow 'register' storage class specifier [-Wregister] " I attached the project-config.jam file I used to this message. Someone please help me out (if there's someone on here who can). _______________________________________________ LLVM Developers mailing list llvm-dev(a)lists.llvm.org<mailto:llvm-dev@lists.llvm.org> http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-dev _______________________________________________ cfe-dev mailing list cfe-dev(a)lists.llvm.org<mailto:cfe-dev@lists.llvm.org> http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-dev _______________________________________________ cfe-dev mailing list cfe-dev(a)lists.llvm.org<mailto:cfe-dev@lists.llvm.org> http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-dev -- Quantitative analyst, Ph.D. Blog: http://blog.audio-tk.com/ LinkedIn: http://www.linkedin.com/in/matthieubrucher

2 1

Problem downloading 1.69.0 for Windows
by tim＠raisedbar.net 24 Dec '18

24 Dec '18

Hi, Windows Defender is blocking my download from https://www.boost.org/users/history/version_1_69_0.html as it is identified as being unsafe. Can anybody advise, please? Best wishes. Tim Burgess Raised Bar Ltd. E: <mailto:tim@raisedbar.net> tim(a)raisedbar.net M: +44 (0)7989 486976

2 2

[Asio/Beast] Any non-obvious reasons for no bind_executor_and_allocator?
by Sorin Fetche 23 Dec '18

23 Dec '18

Hi, As I'm looking to modernize an older C++03 project based on Boost.Asio to use C++17 and the net TS api, I am not ready yet to give up on using lambdas for completion handlers. Boost.Asio and Boost.Beast have some utilities for correctly implementing asynchronous composed operations but from them only asio::bind_executor seems to be applicable to lambdas. And the problem is that it only associates an executor with a (lambda) object; I didn't find any utility for associating an allocator too. Could there be any reason for a utility that associates both an executor and an allocator with an object to not work correctly when implementing asynchronous operations? Or is it just that this utility has not been implemented yet? I did come across the paper "Networking TS Associations for Call Wrappers"[1], but I didn't see any mention there about bind_executor or something similar for allocators or why this kind of approach wouldn't work. I would be very happy if I could write something like: asio::post( socket_.get_executor(), bind_executor_and_allocator_from( handler_, [...] { self_.handler_(...); })); and, unless I'm missing something, such a utility could be implemented similarly to the existing asio::bind_executor. Thank you, Sorin [1] http://open-std.org/JTC1/SC22/WG21/docs/papers/2018/p1133r0.html

2 7

Update for C++17 register keyword error question
by Osman Zakir 23 Dec '18

23 Dec '18

This is the error I get when trying to build Boost 1.69.0 with LLVM: "error: ISO C++17 does not allow 'register' storage class specifier [-Wregister]". I've also sent an email about this to the LLVM developer mailing list. I've attached the project-config.jam file to this message. Please let me know how to fix this. Thanks in advance.

2 2

[uuid] constexpr uuid from string and ""_uuid literal
by Tobias Loew 23 Dec '18

23 Dec '18

Hi, while searching for constexpr GUID from string generators, I found https://gist.github.com/AlexBAV/b58e92d7632bae5d6f5947be455f796f . Since I still work with VS 2015, I wrote a C++11 constexpr version incl. boost-uuid generators. If there's interest adding this to boost uuid, please have a look at https://github.com/tobias-loew/constexpr-GUID-cpp-11 Tobias -- Sent from: http://boost.2283326.n4.nabble.com/Boost-Users-f2553780.html

2 1

Patch for using boost-1_69_0 and clang-cl
by JR Cary 23 Dec '18

23 Dec '18

I found that I had to apply the attached patch to build another project that uses boost with clang-cl, or else I got an error that complained about the non-existence of std::snprintf. John Cary

2 1

Error when trying to build Boost 1.69.0 with LLVM
by Osman Zakir 22 Dec '18

22 Dec '18

When trying to build Boost 1.69.0 with LLVM, I had an error about the "register" keyword, regarding the removal of it in C++17. I searched on the Web and found out that I can use the "-Wno-register" flag to calm down Clang 6.0.0. What do I do in Clang 8.0.0? The same?

1 0

boost_1_19_0 fails to build random using clang-cl
by JR Cary 21 Dec '18

21 Dec '18

This is now more of an FYI, as I can disable random and get the build I need. boost 1_69_0 clang version 8.0.0 (trunk 349323) Windows 10 Build command: ./b2 toolset=clang-win runtime-link=shared link=shared define=BOOST_TEST_DYN_LINK cxxflags='-DBOOST_THREAD_DONT_USE_CHRONO' linkflags='-DBOOST_THREAD_DONT_USE_CHRONO' address-model=64 threading=multi variant=release -s NO_COMPRESSION=1 --layout=system --abbreviate-paths -j2 --without-python --without-mpi --without-log --without-math stage The first error is Creating library bin.v2\libs\random\build\clng-win-8.0.0\rls\adrs-mdl-64\thrd-mlt\boost_random.lib and object bin.v2\libs\random\build\clng-win-8.0.0\rls\adrs-mdl-64\thrd-mlt\boost_random.exp^M random_device.obj : error LNK2019: unresolved external symbol __imp_CryptEnumProvidersA referenced in function "public: __cdecl boost::random::random_device::impl::impl(class std::basic_string<char,struct std::char_traits<char>,class std::allocator<char> > const &)" (??0impl@random_device@random@boost@@QEAA@AEBV?$basic_string@DU?$char_traits@D@std@@V?$allocator@D@2@@std@@@Z)^M random_device.obj : error LNK2019: unresolved external symbol __imp_CryptAcquireContextA referenced in function "public: __cdecl boost::random::random_device::impl::impl(class std::basic_string<char,struct std::char_traits<char>,class std::allocator<char> > const &)" (??0impl@random_device@random@boost@@QEAA@AEBV?$basic_string@DU?$char_traits@D@std@@V?$allocator@D@2@@std@@@Z)^M random_device.obj : error LNK2019: unresolved external symbol __imp_CryptReleaseContext referenced in function "public: __cdecl boost::random::random_device::impl::~impl(void)" (??1impl@random_device@random@boost@@QEAA@XZ)^M random_device.obj : error LNK2019: unresolved external symbol __imp_CryptGenRandom referenced in function "public: unsigned int __cdecl boost::random::random_device::operator()(void)" (??Rrandom_device@random@boost@@QEAAIXZ)^M bin.v2\libs\random\build\clng-win-8.0.0\rls\adrs-mdl-64\thrd-mlt\boost_random.dll : fatal error LNK1120: 4 unresolved externals^M clang-cl: error: linker command failed with exit code 1120 (use -v to see invocation)^M The build succeeds upon adding --without-random to the build invocation. I tried attaching the full log, but then the message was rejected for being too large. ..................John Cary

1 0

Questions about Boost.Multiprecision..
by A Z 20 Dec '18

20 Dec '18

I want to be performing some algebraic functions work on cpp_rational which I will be using templates with the following type to configure precision (and scale?) #include <boost/multiprecision/cpp_int.hpp> in order to avoid the use of floating point types. --What is the #include statement for cpp_rational? I can't manage to find it. -When using non-floating point, is there a method to obtain a variable precision pi, or do I have to use arccos(-1); ? What is the file to import for that? -If I want to import a non-floating-point (cpp_rational) library power, nth root, trigonometry, all the useful function, what is the #include and the file in boost for that?

2 1

Having some trouble with bootstrapping Boost to use LLVM
by Osman Zakir 20 Dec '18

20 Dec '18

I changed the project-config.jam file in my Boost source tree to have Boost.Build use LLVM, but then when I used "bootstrap" command after, I saw that it was still using MSVC. When I looked at the project-config.jam file again at that point, I saw that its content had been reverted back to the default one for building with MSVC. What's going on? What do I need to do? Is there an argument I need to provide to the bootstrap command here so that it'll keep the configuration file content that I want it to? And is there no help command that would let me see what arguments, if any, I can pass to the tool? Another thing I want to ask is whether or not I need to have Boost.Built emit library files with a .bc extension to be able to link them with code I want to build using the Emscripten compiler. I ask because Emscripten needs compiled libraries to have been compiled to LLVM bitcode, and when Emscripten itself does this it adds the .bc extension instead of the .lib one.

2 2

integer_log2.hpp header deprecation using Boost
by Michael Powell 19 Dec '18

19 Dec '18

Hello, I am receiving a deprecation warning: d:\dev\boost.org\boost_1_69_0\boost\pending\integer_log2.hpp(7): note: This header is deprecated. Use <boost/integer/integer_log2.hpp> instead. The immediate includes are these: #include <boost/uuid/uuid.hpp> #include <boost/uuid/uuid_io.hpp> #include <boost/uuid/random_generator.hpp> But I am also leveraging Spirit, Format, Variant, Optional, to name a few, that I can determine. I am not including "integer_log2.hpp" directly, so I wondered which library(ies) might have this as a dependency? Spirit? Phoenix, via Spirit? Format? Thoughts? Suggestions? Cheers, Michael Powell

1 1

odeint: how to solve ODE as you go without deciding upfront what the interval is?
by Yuri 19 Dec '18

19 Dec '18

The function 'integrate' solves ODE for a pre-defined interval of time. What if I don't know what the interval is? I just need a few values, and then I would like to decide if I need more values, and when to stop. I don't want to just solve for some arbitrarily large interval, because this would waste CPU. On the other hand, choosing the shorter original interval and then continuing would introduce unnecessary errors if one would do this too many times. It seems like either the 'observer' function should have another argument: 'stop', or the interface should allow continuation. Yuri

1 0

[Spirit] X3 thread-safety
by Igor R 19 Dec '18

19 Dec '18

Hello, In my project I run x3-based parsers in parallel, using openmp. I've just moved from boost 1.64 to 1.68, and the parsers now cause segfault (heap corruption) - quite consistently, every time I run the program. OTOH, with 1.64 it never crashed. Before I begin to debug it in depth, I'd like to know if there's some known issue with Spirit's thread-safety in Boost 1.68. Did anyone encounter thread-safety issie with the recent Spirit versions? Thanks.

1 0

Build boost asio http server examples single threaded using c++03
by Eric Curtin 19 Dec '18

19 Dec '18

Hi Guys, I was wondering if anyone could help me out building all the boost asio http server examples from: https://github.com/boostorg/asio/tree/develop/example/cpp03/http But catch is, I need to use c++03/98, single threaded without dependencies on std::thread, boost::thread or posix threads. I also need static linking to all the other boost libraries. Also want to compile with g++ -O3 -flto on x86_64 Linux. How can I build these examples like this? This is also a question on stackoverflow: https://stackoverflow.com/questions/53822438/building-boost-asio-without-th… Lets say the build environment is this one: curtine/dockerfiles-fat:ubuntu18.04 Regards, Eric

2 1