- Boost - lists.boost.org

Scoped Enum Emulation
by Joshua Boyce 26 Jan '12

26 Jan '12

I'm currently updating some of my code to take advantage of C++0x features where available, but I also want to retain backwards compatibility with C++03. I was looking at scoped enums and noticed a thread on the mailing list by Beman Dawes discussing a scoped enum emulation implementation for Boost. I see that it's currently being used by Boost, and resides in /boost/detail/scoped_enum_emulation.hpp. My question is, why is this an implementation detail? Would it be possible for it to be moved to Boost.Config with the other 'helper' macros like BOOST_CONSTEXPR? Obviously it's not hard to reimplement, but I figured that it would be useful to other library users, so it would be best (imo) if it were moved somewhere public, rather than being hidden away. Thanks.

4 10

Re: [boost] [lexical_cast] Fast conversion from boost::iterator_range<std::string::iterator>?
by Matthew Chambers 26 Jan '12

26 Jan '12

On Oct 13, 2011 Michel Morin wrote: > lexical_cast made optimizations for some types (such as `std::string`) > to achieve better performance. But, from my experience, > it seems that lexical_cast does not make any optimization for > `boost::iterator_range<std::string::iterator>`. > > Is there any plan to add optimization for such iterator ranges? > Currently, > - `lexical_cast<int>(std::string(iter_rng.begin(), iter_rng.end()))` > is faster > than `lexical_cast<int>(iter_rng)`. > - Assuming it is safe to use `std::atoi`, > `std::atoi(&iter_rng.front())` is faster > than both of the above two methods. > (`iter_rng` is an instance of `boost::iterator_range<std::string::iterator>`.) > > Regards, > Michel This post was on the boost users list but never got a reply. I figure it might be a better fit on the developer list. I too would like to see optimized conversions work on string ranges; what's the point in doing a fast conversion if you have to do a string copy first? :) -Matt

8 16

Attribute names with underscores for GCC
by Michel Morin 26 Jan '12

26 Jan '12

Recently, Petr Machata created several trac tickets for requesting the use of attribute names with underscores for GCC. Since there is no link to the GCC's documentation in the tickets, I'd like to put a link to the documentation. In the GCC's manuals --> "Extensions to the C Language Family" --> "Declaring Attributes of Functions", "Specifying Attributes of Variables", and "Specifying Attributes of Types", there is description about attributes with preceding and following `__`: You may also specify attributes with `__' preceding and following each keyword. This allows you to use them in header files without being concerned about a possible macro of the same name. For example, you may use __noreturn__ instead of noreturn. Here are links to the gcc-2.9.5's documentation: http://gcc.gnu.org/onlinedocs/gcc-2.95.3/gcc_4.html#SEC84 http://gcc.gnu.org/onlinedocs/gcc-2.95.3/gcc_4.html#SEC90 http://gcc.gnu.org/onlinedocs/gcc-2.95.3/gcc_4.html#SEC91 I also checked manuals of gcc-3.0.4 and 4.0.4, and found the same description. Hope this helps, Michel

2 2

[assert] [exception] Fast output binary file size growth
by Antony Polukhin 26 Jan '12

26 Jan '12

Greetings, When BOOST_ENABLE_ASSERT_HANDLER is defined or BOOST_THROW_EXCEPTION used, compiler generates a string with function name. If a function is templated, compiler will generate a string for each function template. That leads to a huge binary files. Consider the following example: #define BOOST_ENABLE_ASSERT_HANDLER #include <boost/assert.hpp> template <class MplVectorT> void do_something(bool v) { BOOST_ASSERT(v); } void do_something<boost::mpl::vector<int, double>>(v); void do_something<boost::mpl::vector<int, int>>(v); This two functions will generate strings like "void do_something<MplVectorT>(bool v) [with MplVectorT = boost::mpl::vector<int, double> ]" and "void do_something<MplVectorT>(bool v) [with MplVectorT = boost::mpl::vector<int, int> ]". When there is more mataprogramming, one such string can have a size of a few kilobytes. In one of my binaries, such strings take more than 50 megabytes of binary size (total size of binary is 70 megabytes). I`ve got a patch and documentation update for Boost.Assert and Boost.Exception, that allow to disable construction of such strings or to use different macro for current function name (for example macro that prints ONLY function name). Is there interest in this feature/patches? Best regards, Antony Polukhin

2 2

[filesystem] noexcept question? [was: Scoped Enum Emulation]
by Beman Dawes 26 Jan '12

26 Jan '12

On Tue, Jan 24, 2012 at 8:38 PM, Joshua Boyce <raptorfactor(a)raptorfactor.com> wrote: > While I've got your attention, I have a similar question along the same > lines. I notice in the mailing list archives that you added some noexcept > related macros to Boost.Config on trunk back in October 2011, I can't see > them in the current trunk or release though. What happened to those? They won't go in 1.49.0, but will begin to go into trunk within the next month. Likewise some other C++11 support, such as rvalue reference signatures. --Beman

2 1

[filesystem] documentation typo
by Greg Rubino 26 Jan '12

26 Jan '12

Hello, It looks like there's some copy/paste residue in http://www.boost.org/doc/libs/1_48_0/libs/filesystem/v3/doc/reference.html#… basic_filesystem_error has become filesystem_error, but in the anchor it's still called 'basic_filesystem_error', and also there are a couple of references on the page to basic_filesystem_error. This momentarily confused me when I was moving a codebase from v1 to v3 yesterday, so I thought it was worth mentioning. I've entered a trac ticket for this issue: https://svn.boost.org/trac/boost/ticket/6448 but I would be delighted to submit a patch for this since I've been wanting to get more involved in Boost, but it would be my first time doing and I'm not quite sure how to go about it. I assume that it involves checking out the trunk and modifying the necessary files, but I'm not sure after that. Greg

2 1

Boost.Application
by Renato Tegon Forti 26 Jan '12

26 Jan '12

Hi All, I have the first prototype of the "Boost.Application" ready! I would like to receive comments so I can refine the library. Please send your feedback to re.tf(a)acm.org 0) Introduction A Boost.Application allows a developer run your application as a Windows service, as a Unix/Linux daemon, or as a usual interactive application (terminal) without the need to add extra code, the library abstract OS specific API and provide a ease common interface. In case of server application, developer need only compile your code on a Windows to have a ready to play Windows Service or in a UNIX/Linux to have Daemon. CTRL-C signals and SCM (Service Control Manager) controls handlers are available for terminating, pause/resume(windows) the application constructively and many other until methods are available in application class like a single instance running limitation! 1) Download: https://sourceforge.net/projects/boostapp/ 2) Installation: The library is one: Automatic linking You should unpack the files and copy: Boost.Application_pre_alpha_1\boost\application To: boost_1_48_0\boost and, Boost.Application_pre_alpha_1\libs\application To: boost_1_48_0\libs 3) Build Then you must build, for example (in windows): bjam toolset=msvc-9.0 variant=debug threading=multi link=shared define=_BIND_TO_CURRENT_VCLIBS_VERSION 4) Documentation The project has an incomplete documentation: Boost.Application_pre_alpha_1\libs\application\doc\html\index.html 5) Use however, the tutorial explains how to use the lib, and that generally goes like this: ---------------------------------------------------------------------------- -------------------- #include <boost/application/engine.hpp> class myapp : public boost::application::application { protected: int main(const std::vector<std::string>& args) { // your application logic here! while(1) { std::cout << "Hello Boost.Application!" << std::endl; // ... if(state() == boost::application::application_stoped) { break; } } return 0; } }; BOOST_APP(myapp, "MainTutorial1", "Boost.Application Main Tutorial 1!", boost::application::default_application_type, 1, 0, 1) ---------------------------------------------------------------------------- -------------------- The folder: Boost.Application_pre_alpha_1\libs\application\example\tutorial Has all tutorial files! I tested the lib on Windows VC9 (Windows Service) and Ubuntu 11 GCC 4 (Daemon)! 6) Questions: * Is there any interest in this library? * If Yes! What are the refinements? (see todo, for some) Please send your feedback to re.tf(a)acm.org Thanks to all Renato

3 5

Boost.Application (PART II)
by Renato Tegon Forti 26 Jan '12

26 Jan '12

Hi all, I want to emphasize: 1) This is a prototype, a starting point to find out if this is an interesting lib to boost! 2) And if 1 is true! I will compile all comments and make modifications to meet "boost"! My doubt is if this lib don't violate: "The library must be generally useful and not restricted to a narrow problem domain." About comments of friends! First, thank you for the comments! They always show a new way of doing things, which I had not thought of! This is the main reason I'm doing (trying to) this lib, to have the opportunity to learn (and contribute) with you! I will try to enumerate the comments, and then build a path to follow for next version (if [1] is satisfied. Do you think I need more votes to continue the effort?) a) >>So selling it as "ready to play" UNIX/Linux daemon is a bit - erm - exaggerating. Sorry about this, I will change this description! I agree with you! Sorry! b) >> Why is the application not a functor or even a function? . Again I agree with you! This is a better way!, for sample I can handle exception in my main too! Thanks about this suggestion! >> I don't like it, though I don't have an alternative solution at hand. Somehow this feels very clumsy with that init() method. As example of a pure OO approach you could have a look at Qts QApplication. I will look QApplication! If you have time! You could describe what you think would be the right path follows in more detail, I appreciate it! c) >> The biggest concern for me is that you're using narrow strings I will provide wide strings. I don't know how yet. I will check Boost.Filesystem! Besides, there is other example to follow? For sample, I need use W version (GetModuleFileNameW) on all? And use a function like this to convert: inline void convert(const char* from, std::wstring & to, const codecvt_type& cvt) D) Other question, to handle exceptions! Any Idea? e) "Install and uninstall windows service" functionality, be a part of the application instead of a separate one and using a callback. f) * In the unknown mode, the program can run in either mode provided someone puts something in the command line. Could that rather be a callback maybe post initialize such that we tell your application which one we want. This way the developer controls his command line. After all, it might not come from the command line but instead from a configuration file. -------------------------------------------------------------------------- What else? Thanks for help!!!

1 0

boost filesystem path as utf-8?
by Emil Dotchevski 25 Jan '12

25 Jan '12

Hello, I understand that the path class .native() member function's return type differs depending on the platform (wstring on windows, for example.) Is there a way to get the path as a utf-8 string regardless of the platform? Likewise, is there a way to construct a path object from a utf-8 string regardless of the platform? Emil Dotchevski Reverge Studios, Inc. http://www.revergestudios.com/reblog/index.php?n=ReCode

4 9

[log][phoenix] Questions on porting to phoenix
by Andrey Semashev 25 Jan '12

25 Jan '12

Hi, [My apologies for the long post] I'm trying to port filters and formatters of Boost.Log to Boost.Phoenix and I'm not sure I understand what's the best way to go. Hope someone can help me out. A short introduction of the terms and what is currently available. * An attribute value is a type-erased value, something similar to boost::any but with more advanced interface to allow visitation and extraction of the stored value. * A log record is an object which contains a set of named attribute values. For simplicity, let's say: class attribute_value; typedef map< string, attribute_value > attribute_values_view; class record { public: attribute_values_view const& attribute_values() const; string const& message() const; }; * A filter is a function that accepts attribute values and returns a bool: typedef function< bool (attribute_values_view const&) > filter; The filter is assumed to pass the log record if it returns true. * A formatter is a function that accepts a log record and a stream and outputs strings or attribute values to the stream: typedef function< void (ostream&, record const&) > formatter; * I already have the code that extracts attribute values from log records. What I'm trying to achieve is something along these lines (this is not a finalized interface yet): enum my_severity { info, warning, error }; BOOST_LOG_REGISTER_ATTRIBUTE("Severity", severity, my_severity) filter flt = severity >= warning; formatter fmt = stream << "[" << severity << "] " << message; Here, "stream" and "message" are keywords provided by Boost.Log, the former is a placeholder for the stream argument of the formatter function, and the latter returns the result of the record::message() method. Now, the BOOST_LOG_REGISTER_ATTRIBUTE macro is what I'm struggling on. It should create a new keyword (severity) which should: 1. Be a POD object to be statically initialized. 2. Contain the information on the type (my_severity) and name ("Severity") of the attribute to be able to extract it from the log record. 3. Be a Phoenix object to initiate lazy expression generation. >From what I gathered from the docs and examples, the generated keyword should be a terminal, in terms of Proto. Am I correct? In order to produce the value, during the evaluation the keyword should receive either a log record or a attribute_values_view and I can't quite understand how do I do that. Basically, I don't quite understand how do I declare and implement a terminal so that it works with Phoenix. This question also applies to the "stream" and "message" keywords but I suspect the solution will be similar. Maybe there is an example of doing this in the docs or elsewhere. I'll be glad if someone provides a link or example of doing this. Thanks in advance.

5 11