- Boost - lists.boost.org

[program_options] custom validator overrides
by Roger Martin 16 Oct '18

16 Oct '18

Hi, The Custom Validators documentation https://www.boost.org/doc/libs/1_68_0/doc/html/program_options/howto.html shows void validate(boost::any& v, const std::vector<std::string>& values, magic_number* target_type, int) { but the possible templates don't match; the last variable type is an int; the key is matching this last type: int goes with std::vector<T>* long goes with T* it also seems possible to hit the void validate(boost::any& v, const std::vector<std::basic_string<charT> >& s, boost::optional<T>*, int) { where it then calls for get_single_string and if you are trying for a multitokens switch args, it'll throw " only takes a single argument". which also varies if it was parsed by command_line_parser or parse_config_file command_line_parser: vector size can >= 1 up to number of switch args parse_config_file: vector size ==1 no matter what the number of args are on the input. just one string with all switch args. This was tested with 1.65 yet I don't see anything changing in https://github.com/boostorg/program_options/blob/develop/include/boost/prog…

2 1

[RFC] unique_val (handles, IDs, descriptors...)
by Miguel Ojeda 15 Oct '18

15 Oct '18

Hi all, While working on a medium-sized project, I had to wrap quite some handles, IDs, descriptors, etc. into RAII classes. Somehow, writing those and making them movable seemed more verbose than needed. I searched in the STL and Boost for something already implemented, but nothing seemed to come up. I decided to write my own prototype to see whether I was missing something important; but actually using it in my project felt quite natural and simple. At this point, I asked for a second opinion from a well-known expert (thanks Scott!) and he agreed that it seemed like a reasonable idea but, of course, he would also be surprised if others haven't come up with something similar. So I polished it up a bit and uploaded it to: https://github.com/ojeda/unique_val Below you have the introduction inlined [1]. If you have the time, please take a look (specially to the Rationale) and let me know your comments. The code itself is very short and straightforward. If you already have this class/template in Boost somewhere that I missed, please let me know! Otherwise, if you think this could be interesting to put into Boost in some existing library (not sure which) or as a new tiny library, I would be glad to make the effort to expand it, clean it up, etc. Thank you! Miguel [1] """ A single-header header-only library for representing unique values (handles, IDs, descriptors...) that default to a value on move for C++11, C++14 and C++17. It simplifies writing resource-owning classes (RAII) with move semantics support for resources with handles, IDs or descriptors that are supposed to be used as unique values, not as unique pointers. For instance, it can be used with non-pointer handlers (like the `int` file/socket descriptors in POSIX) and with opaque pointer handlers (like the `HWND` handles in the Win32 API). As such, it can be used as an alternative to `std::unique_ptr` with or without custom deleter, specially for resources that use a non-pointer type or resources that are not simply heap-allocated. It does not have any overhead compared to using the original type. It supports booleans, integers, pointers and enumerations. The default value can be different than the default-constructed value (i.e. different than 0, nullptr, etc.). """

7 13

bjam question about "requires" negation
by James E. King III 13 Oct '18

13 Oct '18

I want to run a test that should run without explicit operator conversion capability and fail to compile with explicit operation conversion capability, but I'm not sure how to negate the requires statement. I'm sure it's trivial but I found no examples. I tried a bunch of things but none worked. [ run implicit_conversion.cpp : : : [ requires NO cxx11_explicit_conversion_operators ] : implicit_conversion ] [ compile-fail implicit_conversion.cpp : [ requires cxx11_explicit_conversion_operators ] : explicit_conversion ] How do I properly implement the "NO" part? Thanks, Jim

2 1

[parameter] An offer to help maintain and improve the library
by Cromwell Enage 13 Oct '18

13 Oct '18

Hello, everyone. It took me a while to get around to this, but here goes. I've been a long-time user of Boost libraries, although life events imposed a bit of a hiatus until recently. But now I have the opportunity to help out some more, and one of the ways I want to give back to this community is by helping to maintain and improve Boost.Parameter. I participated in the review of this library before it was accepted, so I'm familiar with its capabilities and code structure. In addition to fixing any bugs and handling pull requests, I'd like to make the following improvements, which I will summarize here: * Add macros to generate Boost.Parameter-enabled function call operators. * Add parameter category qualifiers "consume", "move_from", and "forward". (Current qualifiers are "in", "out", and "in_out".) * Fix parameter::parameters::operator() so that it also accepts non-const rvalues. * Be no longer bound by BOOST_PARAMETER_MAX_ARITY on compilers that support perfect forwarding. * Enable parameter::template_keyword<> to accept function types. * Remove the necessity of wrapping boost::ref() around non-const lvalues. I can discuss each of these bullet points in separate threads if necessary. I await your feedback! Cromwell D. Enage

2 1

Boost xxhr ? High level http client library
by Damien Buhl 13 Oct '18

13 Oct '18

Dear Boost, I've built an header-only library based on Boost.Beast which allows you to perform HTTP requests. It is implememented with two backends : * Boost.Beast for native builds. * XmlHttpRequest for WebAssembly (plans to migrate to fetch API which is richer) Actually Boost.Beast is a great lib, however it's API is thought to build HTTP centric app not just to make some simple HTTP call ( as for example urdl was ). xxhr on the other hand is about making it simple to call an HTTP service. Usage example : GET <https://nxxm.github.io/xxhr/html/namespacexxhr.html#a69bafa24d7c33b004a7de8…>( "http://httpbin.org/anything"s, on_response = [](auto&& resp) { std::cout << resp.text; }); I would be really interested in a pre-review or some feedback about it's API. The implementation merits more attention and can significantly be improved but this email is to grab a first feedback from the community. Documentation : https://nxxm.github.io/xxhr/ GitHub : https://github.com/nxxm/xxhr Example code relying on it : nxxm/gh <https://github.com/nxxm/gh>: partial GitHub API implementation Thanks for your time :) -- Damien Buhl Software Developer +33 6 77 43 10 05 https://nxxm.github.io/

3 5

Test matrix file read errors?
by James E. King III 11 Oct '18

11 Oct '18

I'm trying to look at the develop test matrix for a couple repositories and I am getting an error: Error extracting file: Error code 9 - The specified zipfile was not found *(the > file exists, so this is probably an error reading the file)*. https://www.boost.org/development/tests/develop/developer/date_time.html https://www.boost.org/development/tests/develop/developer/dynamic_bitset.ht… Thanks, Jim

2 1

[concept_check] CopyConstructible Concept Does Not Accept Lvalue Reference
by Qu Xing 11 Oct '18

11 Oct '18

Boost version: 1.67.0 Compiler: GCC 7.3.0 Environment: Cygwin on Windows 10 version 1709 In the Boost Concept Check library, CopyConstructible concept requires operator & applicable to the type being checked, and the return value is required to be convertible to pointer to the type. However, pointer to reference is not allowed in C++. This concept check fails to compile if applying to a lvalue reference type. On the other hand, is_copy_constructible<T> type trait class in the Boost Type Traits library and the standard C++ library both give positive result on lvalue reference types. For example, the following code snippet will fail to compile if T is a lvalue reference: #include <type_traits> #include <boost/static_assert.hpp> #include <boost/type_traits/is_copy_constructible.hpp> #include <boost/concept_check.hpp> #include <boost/concept/assert.hpp> template <typename T> void RequireCopyConstructible() { BOOST_STATIC_ASSERT(boost::is_copy_constructible<T>::value); // PASS on lvalue reference BOOST_STATIC_ASSERT(std::is_copy_constructible<T>::value); // PASS on lvalue reference BOOST_CONCEPT_ASSERT((boost::CopyConstructible<T>)); // FAIL on lvalue reference } void TestReference() { RequireCopyConstructible<int&>(); } Diagnosis message is generated as: "error: forming pointer to reference type ‘int&’". I think the operator & applicable requirement should be removed for CopyConstructible concept. Alternatively, adding a template specialization on lvalue reference type for CopyConstructible concept may work as well.

1 0

[config] really, really sticky issue with BOOST_SYMBOL_IMPORT
by Robert Ramey 10 Oct '18

10 Oct '18

For many, many years we've built the Boost serialization library with the macros for BOOST_SYMBOL_IMPORT and BOOST_SYMBOL_EXPORT set according to the instructions included in the boost/config documentation written by John Maddock. These set attributes for creating and using DLLS for windows and (used to) resolve to nothing on other systems. So far so good. At some point things changed for gcc compilers. Now it seems that BOOST_SYMBOL_IMPORT resolves to nothing and BOOST_SYMBOL_EXPORT resolves to : __attribute__ ((visibility("default"))) and BOOST_SYMBOL_IMPORT still resolves to nothing. OK - this has never been a problem. Now we have an extremely subtle issue which appears on linux systems with gcc. It's related to sequence of shared library loading and unloading. It shows up as a bunch of test failures for the test named test_exported_dll. A very astute user has determined that this can be "fixed" by defining BOOST_SYMBOL_IMPORT as __attribute__ ((visibility("default"))) as well. Question. should config/gcc be updated in this way? Also is it possible that this "fix" is just a fluke unrelated to the real source of the problem? Note that this issue predates the recent change to make -visibility=default which has, as one might expect, caused some issues. I believe that this situation is unrelated. Any help appreciated Robert Ramey reference: https://github.com/boostorg/serialization/issues/104#issuecomment-416948110

4 3

Tools to manage multiple repositories effectively?
by James E. King III 09 Oct '18

09 Oct '18

Hello, I am looking for tools that would allow me to create a portal and/or monitor the following properties of boost repositories. I am in the CMT and at least one fix that was in develop never made it into master for the 1.68.0 release, so I want to be able to see, at a glance, information about the boost repositories I care about in one place. There are about 20 repositories I regularly participate in now so a portal to show what I need to worry about would be useful. 1. How far behind is master (with respect to develop)? This would help identify which repositories are behind develop. Useful for ensuring things get into a release. 2. How many open pull requests? 3. How many new pull requests (within the last week)? 4. How many updated pull requests (within the last week)? 5. How many open issues? 6. How many new issues (within the last week)? 7. How many updated issues (within the last week)? "week" could be configured, perhaps month, or "since I last visited". Anybody know of any good tools or perhaps an existing site that can do this for a multi-repository solution? Thanks, Jim

3 2

[system] The new Boost.System is now ready
by Peter Dimov 09 Oct '18

09 Oct '18

The new Boost.System is now complete, on the develop branch: https://www.boost.org/doc/libs/develop/libs/system/ - header-only - even more constexpr than 1.68 - identifier-based category comparison - message() into a caller-supplied buffer - failed() The plan is to merge it to master at some point if no breakage is encountered on develop, so that it can go into the next release.

5 11