December 2018 - Boost-users

[Asio] Combining use_future with strands
by Sorin Fetche 18 Dec '18

18 Dec '18

Hi, Is there a way to combine the special use_future completion token with a strand? Something with the meaning of: - the asynchronous operation should pass the result to the future it returns, and - the internal operations of it (if any) should run in a strand Code snippet: asio::io_context io_context; asio::io_context::strand strand(io_context); std::future<void> f = async_my_operation(io_context, <params>, asio::use_future); //________________________________________How to add the strand here ^^^ ? The code area where I would like to apply this is in this test program: https://github.com/sorf/cpp-dev-new/blob/abf9d831a0b82fb81751eee6cfa498d3b9… The self contained program and its output is here: https://wandbox.org/permlink/pF5Hbm6jqPOmEKbO As a background, the test program I wrote is based on these Boost.Asio examples: - for strand usage (i.e. bind_executor): https://github.com/boostorg/asio/blob/develop/example/cpp03/tutorial/timer5… - for composed operations: https://github.com/boostorg/asio/blob/develop/example/cpp11/operations/comp… and implements a composed operation that runs multiple internal operations in parallel and in series. Similar to the constructor of a socket or a timer object, the operation receives as a parameter an io_context; it doesn't offer an overload that receives a strand object directly. The operation is then tested with callback completion, callback completion in a strand, future and future + strand (where the test fails). Thank you, Sorin

5 7

How to implement a visitor pattern over an AST
by Michael Powell 18 Dec '18

18 Dec '18

Hello, I have an AST successfully building, generating text file, and subsequently Spirit Qi parsing for verification purposes. Now I want to implement a visitor pattern over that AST, somehow... For "calculators" and such, "evaluating" those expressions is one thing; however, this AST is not that. It is closer akin to a parsed Xml or Json model/tree. In other words, I'd like for it to feel sort of like an iterator if that's even possible. It should have contextual awareness where it is at all times, etc, perhaps "incrementing" is a depth first analysis. In particular I have in mind to iterate the expected and actual AST results in order to perform the verification. In other words, I might do something like this: auto expected_it = ast_visitor{&expected}; auto actual_it = ast_visitor{&actual}; REQUIRE(expected_it); REQUIRE(actual_it); REQUIRE(*expected_it == *actual_it); // Rinse and repeat ++expected_it; ++actual_it; REQUIRE(*expected_it == *actual_it); I wonder if something like this is even possible from the Boost libraries? Or interested to hear any insights on the subject. For that matter, I wonder it is more appropriate to simply implement the logical operators. Thanks! Michael Powell

4 7

Understanding fibers
by Stephan Menzel 17 Dec '18

17 Dec '18

Hello @ll, I am in the process of transforming a sync server application to an async approach. The framework in use (not asio) instantiates and calls an object representing the call when a request comes in, allows me to do things, fill a response object and then expects me to call a finish() method which will send the response and start the next request. This is not what this question is about but more like my circumstances. All code is pseudocode please... void rpc_work() { const value = do_stuff(request); response.fill_value(value) parent->finish(); } Now some of the calls will imply calling methods, which block on some operations, waiting for external resources to respond. Obviously in an async server with only one thread, this will block everything. So I'm looking for a clever way to structure this. My approach so far (in the prototype) is to go boost::async() for those operations. Like this: void rpc_work() { // prepare work boost::async(launch::any, [request, response, parent]() { const value = do_blocking_stuff(request); response.fill_value(value) parent->finish(); }); } The way I understand boost::async or indeed std::async is, that normally it will just spawn a thread for the operation. The best I can hope for is a thread pool but this is not guaranteed either (on Linux x64 gcc 7.3). So I started looking into fibers but I'm having a hard time understanding them, which is why I post here. From what I gather, what I have here is a prime use case as it matches many things the docs talk about. And yet they also suggest that the fibers are very intrusive and everything underneath do_blocking_stuff() would have to be aware of being in a fiber and yield() when appropriate. Is this correct? What would this mean? Most if not all of the blocking operations are using boost::thread::futures in their blocking things. They would do things like... value_t do_blocking_stuff(request) { boost::future<value_t> fv = retrieve_value_from_somewhere(); fv.wait_for(timeout); return fv.get() } Now, considering I would use fibers in the server, would I have to use 'fiber futures' here to make this work? To make the blocking functions automatically 'fiber aware' and yield() when they would block? This would imply replacing the thread::futures? And if so, how will I know when the value is there and they are ready to continue? And how will I cause this continuation? Do I have to keep track of them and somehow poll() them in a loop or something? I am aware of the broad nature of my question but perhaps someone can chime in and give some hints that allow me to understand them better. They look like they would fit my use case perfectly but I am a bit reluctant of dragging the concept all throughout the code until I realize it won't work. Perhaps someone can eli5 this to me ;-) Cheers, Stephan

4 13

Boost Process fails on Solaris 10 in async mode
by Sanket Sharma 17 Dec '18

17 Dec '18

Hi, I tried installing boost process on my Solaris 10 Sparc machine. On compiling it using gcc-6.3.0, I ran into a static assert failure at compile time at https://github.com/boostorg/process/blob/boost-1.68.0/include/boost/process…. In order to fix that, I changed static_assert(!WIFEXITED(still_active) && !WIFSIGNALED(still_active), "Internal Error"); to static_assert(!WIFEXITED(still_active), "Internal Error"); There is some discussion about this at https://www.openwall.com/lists/musl/2018/08/10/1 Now, the following code works well onCentOS 7(where no change was made to Boost process), but fails on Solaris 10 (where I made the above change as a workaround). On Solaris 10, the return code is 127 and stdout and stderr are empty. #include <iostream>#include <thread>#include <boost/process.hpp>#include <boost/asio.hpp> namespace bp = boost::process ;int main(int argc, char** argv){ std::future<std::string> dataOut; std::future<std::string> dataErr; std::string command = "ls"; boost::asio::io_service ios; bp::child c(command, bp::std_out > dataOut, bp::std_err > dataErr, ios); ios.run(); c.wait(); int result = c.exit_code(); std::string stdOut = dataOut.get(); std::string stdErr = dataErr.get(); std::cout << "Exit Code " << result << std::endl; std::cout << "stdOut " << stdOut << std::endl; std::cout << "stdErr " << stdErr << std::endl; return 0;} Output on CentOS 7 Exit Code 0 stdOut a.out main.cppProcess.cppProcess.h simple.cpp stdErr Output on Solaris 10 Exit Code 127 stdOut stdErr 1. Has my workaround to install Boost process on Solaris 10 broken it? 2. Is Boost-1.68.0 not supported on Solaris? 3. Is there a way I can get Boost Process to work on Solaris 10 in async mode? https://stackoverflow.com/questions/53780273/boost-process-fails-on-solaris… -- Regards, Sanket

1 0

Boost version 1.69.0 already included in list of known versions in FindBoost.cmake module
by Osman Zakir 17 Dec '18

17 Dec '18

I looked in the FindBoost.cmake module and saw that version 1.69.0 is already listed as a known version. But when I configured CMake to build Jinja2Cpp after building Boost 1.69.0, I couldn't get it to find some modules of Boost. There were warning messages about imported targets and dependences or some such. Jinja2Cpp uses an earlier version of Boost from its own submodule in this case (it has a GitHub repository that I cloned from and built), but I'd like it if it used the latest version of Boost instead of an earlier version. My Boost installation is in C:/boost_1_69_0/, by the way. Could this be causing problems, or will it still be fine I configure CMake correctly? What CMake option should I use to find Boost? -DBOOST_ROOT?

3 5

[Spirit] Comment skipper or first class member of grammar
by Michael Powell 17 Dec '18

17 Dec '18

Hello, I am turning a corner in my Protocol Buffers grammar, and it is now time for me to consider how best to approach parsing comments. Chiefly, I am weighing two options, I think: 1) A white-space, comments skipper; if memory serves there has been discussion about such a thing on either here, Boost mailing list, or other forums. 2) Or, as with some compiler environments, I'm thinking .NET Roslyn/C#, in particular, where "comment" elements are first class members of the grammar/tokens, quite literally, I think. I'm not quite sure how that would be expressed in terms of synthesized AST, when you could literally have a comment appearing virtually anywhere between tokens, especially of the C variety, i.e. /* this is a comment which may appear virtually anywhere between tokens, potentially */. Or the single line C++ variety, for example, // this is a comment through the rest of this line. // Yet, there could be multiple lines of these... optional myVar = true; // And can appear at the end of a line Thoughts? Suggestions? My first approach I think would be for (1) the skipper; however, for what I am wanting to accomplish, I can see there potentially being some value in capturing the comments and carrying them forward with the target language adapter. Best regards, Michael Powell

1 1

[process] distinguish normal exit from termination by signal - complete
by Oliver Kowalke 17 Dec '18

17 Dec '18

Hello, I'm evaluating boost.process. It seams to me that it is impossible to distinguish a normal exit from a termination by a signal (for instance SIGBUS etc.). child::exit_code() calls eval_exit_status() that expands on POSIX to: if (WIFEXITED(code)) { return WEXITSTATUS(code); } else if (WIFSIGNALED(code)) { return WTERMSIG(code); } else { return code; } With the code above it is impossible if the returned exit_code is a `exit-code` or the number of a signal. Or do I miss something? regards, Oliver

1 0

[process]distinguish normal exit from signaled not possible?
by Oliver Kowalke 17 Dec '18

17 Dec '18

Hello, I'm evaluating boost.process. It seams to me that it is impossible to distinguish a normal exit from a termination by a signal (for instance SIGBUS etc.). child::exit_code() calls eval_exit_status() that expands on POSIX to: if (WIFEXITED(code)) { return WEXITSTATUS(code); } else if (WIFSIGNALED(code)) { return WTERMSIG(code); } else { return code; }

1 0

[contextual async] generator support added
by Christophe Bailly 16 Dec '18

16 Dec '18

Hello, I am still waiting for some feedbacks. I really think it makes possible to add asynchronous code very easiy, I have added the support of generators which can run with asynchronous.code. You will find the added article in the blog: https://pythonc.home.blog/ The project has been updated and the examples are available if you want to test. Regards, Chris

2 2

[corountine2/context] forced_unwind exception on Windows with TDM-GCC
by Hao Jiang 15 Dec '18

15 Dec '18

Hello all, I am porting my program from Linux to Windows with boost coroutine2. I found 32bit program throwing boost::context::detail::forced_unwind every time when coroutine destructed. But 64bit program works like charm. terminate called after throwing an instance of 'boost::context::detail::forced_unwind' The program works well on Linux(Ubuntu). Then I tried the official example: typedef boost::coroutines2::coroutine<int> coro_t; coro_t::pull_type source( [&](coro_t::push_type& sink){ int first=1,second=1; sink(first); sink(second); for(int i=0;i<8;++i){ int third=first+second; first=second; second=third; sink(third); } }); for(auto i:source) std::cout << i << " "; Same thing happened. Platform: windows 7 64 bit Boost: 1.67/1.69 Compiler: TDM-GCC 5.1 64bit. Best,

3 4

project-config.jam File for building Boost with LLVM toolchain?
by Osman Zakir 14 Dec '18

14 Dec '18

Sorry, but I need to see the project-config.jam file I need to use for building Boost with the LLVM toolchain again. I'm going to build it with the MSVC toolchain first and then again with the LLVM toolchain as well so that I have both the MSVC- and Clang-generated binaries. I downloaded Boost version 1.69.0 and I want to build it now. Thanks in advance.

5 11

[boost] mixing async + sync code easily via boost::context
by Christophe Bailly 14 Dec '18

14 Dec '18

Hello, I have worked on a prototype to avoid the current limitations with async code, the use of sync + async code is really difficult, it does not co-exist easily. If you introduce async code, you have to migrate all your code ... There are alternative solutions (threads..) but here is a suggestion to overcome these limitations. It is currently a very simple prototype but I would be interested to have your feedback. I think this kind of solution can be really powerfull, but I will wait to know what you think about it. Here are the links: https://pythonc.home.blog/ https://github.com/chbailly/5a5 Regards, Chris

1 0

C++Now (BoostCon) 2019 Call for Submissions
by Jon Kalb 13 Dec '18

13 Dec '18

The C++Now (formerly BoostCon) 2019 Call for Submissions is now open. http://cppnow.org/announcements/2018/12/2019-call-for-submissions/ C++Now will be held in Aspen, Colorado on May 5th - 10th. Colorado has become trendy for C++ conferences, but BoostCon was here first. If you are interested in Boost Libraries and have never attended C++Now, you owe it to yourself to have the BoostCon experience. The best way to get your company to send you is to have an accepted submission. The deadline is early in the new year, so don’t delay. Jon Kalb Conference Chair C++Now

1 0

[release] Boost 1.69.0
by Marshall Clow 12 Dec '18

12 Dec '18

Release 1.69.0 of the Boost C++ Libraries is now available. These open-source libraries work well with the C++ Standard Library, and are usable across a broad spectrum of applications. The Boost license encourages both commercial and non-commercial use. This release contains numerous enhancements and bug fixes for existing libraries. For details, including download links, see < http://www.boost.org/users/news/version_1.69.0> You can also download directly from: < https://dl.bintray.com/boostorg/release/1.69.0/> SHA-256 sums: af05616f71006f97833e130aad886c96136457511ace4b5496d6566e69cbe0ca boost_1_69_0.7z 8f32d4617390d1c2d16f26a27ab60d97807b35440d45891fa340fc2648b04406 boost_1_69_0.tar.bz2 9a2c2819310839ea373f42d69e733c339b4e9a19deab6bfec448281554aa4dbb boost_1_69_0.tar.gz d074bcbcc0501c4917b965fc890e303ee70d8b01ff5712bae4a6c54f2b6b4e52 boost_1_69_0.zip To install this release on your system, see < http://www.boost.org/doc/libs/release/more/getting_started/index.html> Thanks to everyone who participated in this release. Thanks, -- The Boost Release Team Vladimir Prus, Marshall Clow, Daniel James, Micheal Caisse

3 2

The Master branch is still closed; it will be opened tomorrow
by Marshall Clow 11 Dec '18

11 Dec '18

Thanks for your patience -- The boost release managers

1 0

Formatting Boost.Uuid as string
by Michael Powell 11 Dec '18

11 Dec '18

Hello, I see that you can do "basic" formatting of boost::uuids::uuid via boost::format. This is good for basic formatting, but are there other format specifiers one can furnish? I'm thinking along the lines of formats such as with or without hyphens, etc. Or an ability to extend such a feature for that matter. See my C# WPF GuidGen for some examples: http://github.com/mwpowellhtx/guidgenwpf/tree/master/src/GuidGen.Wpf/Formats This is based on the Microsoft GuidGen, but using WPF technology, and with some additional formats. Thanks! Best regards, Michael Powell

1 0

[release] 1.69.0 closed for release
by Marshall Clow 10 Dec '18

10 Dec '18

The master branch is now closed for all changes except by permission. The commit-bot isn't running on master, so the super project will need to be manually updated for any changes. If all goes well, the release candidate should be out later today. Thanks, -- The Boost Release Team

3 4

travis runners version
by Alain Miniussi 10 Dec '18

10 Dec '18

Hi, I see that travis runners uses open MPI 1.6.5 mpirun.openmpi --version mpirun.openmpi (OpenRTE) 1.6.5 which is considered "retired" on the OpenMPI download page (https://www.open-mpi.org/) current version is 4.0.0 and most linux ditrib I see uses 2.x.x or hirgher. What is the travis runners distrib/version ? Thanks -- ----- Alain Miniussi DSI, Pôles Calcul et Genie Log. Tél. : +33492003009 (Mont-Gros) Tél. : +33483618544 (Sophia Antipolis)

1 0

Boost.Filesystem perms on Windows
by Lars 09 Dec '18

09 Dec '18

Hello, The documentation<https://www.boost.org/doc/libs/1_68_0/libs/filesystem/doc/reference.html#En…> (boost version 1.68) for boost::filesystem::perms enumerator says; "Windows: All permissions except write are currently ignored. There is only a single write permission; setting write permission for owner, group, or others sets write permission for all, and removing write permission for owner, group, or others removes write permission for all." Does "All permissions except write are currently ignored" restriction apply to both getting permission from a file and setting permission to a file? In other words, does it apply to both boost::filesystem::status(..) and boost::filesystem::permissions(..) methods? Why does the fileysystem have this restriction? Is this some limitation in the design or a bug or missing feature? Kind regards, Lars

2 3

Mocking with boost.test
by Sean Farrow 09 Dec '18

09 Dec '18

Hi all, I was just wondering whether boost.test provides any mocking facilities? If it doesn't is there a recommended mocking framework? Any help appreciated. Kind regards Sean.

4 3

GIL: New file io, almost there, but can't compile 'write_view'
by Dan Bloomquist 08 Dec '18

08 Dec '18

So, I had to deal with a lib I worked on last a couple of years ago. Great news, the file IO has been greatly simplified in 1.68. I was able to retire several dozens of cryptic,(for me), lines. The old code won't compile as the Adobe extensions are gone. I'm almost there, I can read files, but I don't understand why I can't get the writer to compile. I get to my code, in case it matters: bg::rgb8_image_t destImg( res, x ); resize_view( view( sourceImg ), view( destImg ), bg::bilinear_sampler( ) ); and rgb8_image_t should be a good one from what I read: <https://www.boost.org/doc/libs/1_68_0/libs/gil/doc/html/io.html> With either: bg::write_view( path_to, destImg, bg::jpeg_tag() ); or: bg::image_write_info<bg::jpeg_tag> write_setttings; bg::write_view( path_to, destImg, write_setttings ); ms140 complains: 1>C:\cpp\boost_1_68_0\boost/gil/extension/toolbox/metafunctions/get_pixel_type.hpp(33): error C2039: 'reference': is not a member of 'boost::gil::image<boost::gil::rgb8_pixel_t,false,std::allocator<unsigned char>>' 1> c:\cpp\boost_1_68_0\boost\gil\typedefs.hpp(176): note: see declaration of 'boost::gil::image<boost::gil::rgb8_pixel_t,false,std::allocator<unsigned char>>' 1> console_test.cpp(248): note: see reference to class template instantiation 'boost::gil::get_pixel_type<boost::gil::rgb8_image_t>' being compiled 1>C:\cpp\boost_1_68_0\boost/gil/extension/toolbox/metafunctions/get_pixel_type.hpp(34): error C2146: syntax error: missing '>' before identifier 'reference' What am I missing? Do I need to supply more information? I'll put the whole thing in a url if needed. Thanks, Dan.

2 3

[release] Boost 1.69.0 Release Candidate 1
by Marshall Clow 07 Dec '18

07 Dec '18

The release candidates for the 1.68.0 release are now available at: <https://dl.bintray.com/boostorg/release/1.69.0/source/> The SHA256 checksums are as follows: af05616f71006f97833e130aad886c96136457511ace4b5496d6566e69cbe0ca boost_1_69_0_rc1.7z 8f32d4617390d1c2d16f26a27ab60d97807b35440d45891fa340fc2648b04406 boost_1_69_0_rc1.tar.bz2 9a2c2819310839ea373f42d69e733c339b4e9a19deab6bfec448281554aa4dbb boost_1_69_0_rc1.tar.gz d074bcbcc0501c4917b965fc890e303ee70d8b01ff5712bae4a6c54f2b6b4e52 boost_1_69_0_rc1.zip 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 Boost Release Managers

2 2

error_code using c++11 and boost
by Lars 06 Dec '18

06 Dec '18

Hello, A project that uses Boost 1.6X (filesystem, uuid, asio) and C++14 on multiplatform. Does it make sense to use std::error_code in this project ? The concern is compatibility between boost and std. The filesystem_error exception will return a boost ::system::error_code which is not compatible with std::error_code as far as I can tell. Exception is not an option. Would it be possible for boost to implement some sort of convertion method from boost::system::error_code to std::errror_code? Thanks kind regards, Lars

3 2

Re: [Boost-users] error_code using c++11 and boost
by Bjorn Reese 06 Dec '18

06 Dec '18

On 11/16/18 4:06 PM, Lars via Boost-users wrote: > Would it be possible for boost to implement some sort of convertion > method from boost::system::error_code to std::errror_code? The upcoming 1.69 release of Boost.System has built-in support for std::error_code. If you need to use earlier versions, then you can create a bridge between the two: http://breese.github.io/2016/06/18/unifying-error-codes.html

3 2

[Spirit] How to prohibit reserved names from matching
by Michael Powell 06 Dec '18

06 Dec '18

Hello, Let's say I have some Qi rules: qi::rule<It, char()> mych; qi::rule<It, std::string()> myname; struct reserved_names_t : qi::symbols<path_t, path_t> { reserved_names_t() { this->add("CON", "CON") ("PRN", "PRN") ("AUX", "AUX") ("NUL", "NUL") ("COM1", "COM1") ("COM2", "COM2") ("COM3", "COM3") ("COM4", "COM4") ("COM5", "COM5") ("COM6", "COM6") ("COM7", "COM7") ("COM8", "COM8") ("COM9", "COM9") ("LPT1", "LPT1") ("LPT2", "LPT2") ("LPT3", "LPT3") ("LPT4", "LPT4") ("LPT5", "LPT5") ("LPT6", "LPT6") ("LPT7", "LPT7") ("LPT8", "LPT8") ("LPT9", "LPT9") ; } } reserved_names; Assuming I've defined mych accordingly, the simple approach would be: myname %= +myname; This gets me the name in the character set. Now, I want to prohibit the reserved names from matching. myname %= ~qi::no_case[reserved_names] | +myname; However, this does not seem to work. I suppose it is one thing ruling out invalid characters, but what about prohibiting whole strings? Thanks in advance for any insights... Michael Powell

1 0