- Boost - lists.boost.org

[preprocessor] pp-strings
by Lorenzo Caminiti 28 Jun '11

28 Jun '11

Hello all, As I understand it, Boost.Preprocessor was largely based on another preprocessor library from the same authors named CHAOS. Is this correct? Do you know where I can find CHAOS source and its documentation? Specifically, I am curious if CHAOS supports "pp space-separated strings" -- i.e., macros like the followings: PP_STRING_CAT( b o o s t ) // expand to `boost` PP_STRING_AT(3, b o o s t ) // expand to `s` etc... If yes, why were these pp-strings not ported from CHAOS to Boost.Preprocessor? Thanks a lot. --Lorenzo

8 12

[algorithm] to_lower_copy / to_upper_copy with no output param assumes sequence to be same as input
by Sebastian Karlsson 28 Jun '11

28 Jun '11

It would seem: std::string foo = boost::algorithm::to_lower_copy< std::string >( "bar" ); isn't valid. Instead one has to do for example: std::string foo = boost::algorithm::to_lower_copy( std::string( "bar" ) ), which while perhaps just as terse and natural creates an extra unneeded temporary. I propose the following addition, as well as one for to_upper_copy: template<typename OutputSequenceT, typename RangeT> inline OutputSequenceT to_lower_copy( const RangeT& Input, const std::locale& Loc=std::locale()) { OutputSequenceT Output( ::boost::begin( Input ), ::boost::end( Input ) ); ::boost::algorithm::detail::transform_range( ::boost::as_literal( Output ), ::boost::algorithm::detail::to_lowerF< typename range_value<OutputSequenceT>::type >(Loc)); return Output; } Or perhaps I've missed something altogether :) Kind regards, Sebastian Karlsson

3 9

filesystem3::rename error codes
by Andreas Haferburg 28 Jun '11

28 Jun '11

Greetings, I was wondering how to handle errors of filesystem3::rename(). I'd like to use the system::error_code& overload, but I can't seem to find out how to use it. The documentation at [1] points to POSIX rename(), so I'm guessing the error_code somehow corresponds to the POSIX rename() return value. I'm not a native speaker, but I find the documentation pretty vague here. The tutorial section [2] doesn't help either. For example if the target file doesn't exist, the error_code.value() is 3. But I'm not supposed to use the constant 3 in the code, am I? Are there any constants defined that correspond to these error code values? What's the proper way to do the following: TEST( BoostFilesystem3Test, testRenameFails ) { boost::system::error_code ec; boost::filesystem3::rename("c:/does/not/exist", "asdf", ec); ASSERT_EQ(3, ec.value()); } I'm going to use QFile for now, but only because their documentation is better. I would like to use boost for as much as possible, so any explanation or pointers to better documentation would be much appreciated. Thank you. Best regards Andreas [1] http://www.boost.org/doc/libs/1_46_1/libs/filesystem/v3/doc/reference.html#… [2] http://www.boost.org/doc/libs/1_46_1/libs/filesystem/v3/doc/tutorial.html#E…

1 0

Re: [boost] Is there interest in an alternative to the Singleton anti-pattern?
by Ben Robinson 28 Jun '11

28 Jun '11

> > On 24 June 2011 02:06, Ben Robinson <icaretaker(a)gmail.com> wrote: > > > If you choose the default policy of single_threaded, there is no > > thread safety, but there is also no overhead in using a mutex and using > a > > volatile pointer. > > > > Volatile in C++ doesn't really help with threading... Check out < > http://software.intel.com/en-us/blogs/2007/11/30/volatile > -almost-useless-for-multi-threaded-programming/> > for one of many articles on the subject. > -- > Nevin ":-)" Liber <mailto:nevin@eviloverlord.com> (847) 691-1404 Nevin, I greatly appreciate the input you have provided. I retract my original response, as I now better understand why volatile does not add any value to Singularity. Singularity is already acquiring a mutex object, before accessing the shared instance pointer, and that mutex acquisition is already providing memory fencing. I have simplified the implementation of Singularity and the latest version is available at: https://github.com/icaretaker/Singularity. Any additional input would be appreciated. Thank you, Ben Robinson, Ph.D.

1 0

Logging library
by Tigran Hayrapetyan 28 Jun '11

28 Jun '11

Hello dear Boost community I'm Tigran Hayrapetyan from Armenia, 22 years old. I have started learning C++ when I was 14, after 1.5 years practicing on BASIC. I want introduce to you an idea related to logging, that I came up with recently. Existing logging libraries which I used, are writing log messages linearly, just one after another. In contrast, as we know, almost any application's working structure is not linear, but hierarchical instead, i.e. functions which call other functions, and so on. When viewing large log file produced by an application, it is not very convenient to find log messages of functions, which took middle place in call hierarchy. Because between that messages there are a lot of other messages produced by lower-level functions in that hierarchy. So, generally, it takes some time to understand called functions sequence, and to find the point from which something goes wrong. After spending time looking for log messages I need in my applications, I think that it is possible to organize logging not linearly, but hierarchically, so the messages will form something like a tree. Log messages which are related to some subtask, can be grouped as child messages of the caller task. Also, in my opinion, priority of the caller task's message can be equal to maximal priority of child messages, so it will show whole status of that task (higher priority is for more serious messages). So I decided to implement such logging library. I put it in namespace "tlog" by now, for "Tree Logging". In my implementation hierarchical messages are presented in XML format, as it is some standard for tree-like data. There are two attached files. The library itself is in "tlog.hpp" file, and "main.cpp" contains a sample application which is using the library and writes log to "tlog.xml" file. They are written in MS Visual Studio 2008. Log messages are written with help of 4 macros: TLOG_INFO( Logger, Text ) : logs message with specified text to specified logger with priority "information" TLOG_WARNING( Logger, Text ) : logs message with specified text to specified logger with priority "warning" TLOG_ERROR( Logger, Text ) : logs message with specified text to specified logger with priority "error" TLOG_SCOPED_NODE( Logger, NodeName ) : opens xml element in the log file. The element will be closed at the end of scope, so there is no need to close it manually. I think that the best way to implement entering/leaving subnodes of the log file, is to use scopes of the code. Please review run result and the code, and see if it can be useful for boost. Best Regards, Tigran Hayrapetyan

1 0

Re: [boost] numeric_cast
by Tom Tan 27 Jun '11

27 Jun '11

It seems to me the outer #ifdef works for MSVC as well. What's the extra benefit of embedding an inner #pragma once in this case? 在 2011-06-24五的 04:22 -0400，boost-request(a)lists.boost.xn--org:-7z8fk634a > Message: 1 > Date: Thu, 23 Jun 2011 21:43:02 +0000 > From: "Stephan T. Lavavej" <stl(a)exchange.microsoft.com> > To: "boost(a)lists.boost.org" <boost(a)lists.boost.org> > Subject: Re: [boost] numeric_cast > Message-ID: > > Actually, the best thing to do is this: > > #ifndef MYPROJECT_MYDIRECTORY_FOOBAR_HPP > #define MYPROJECT_MYDIRECTORY_FOOBAR_HPP > > #ifdef _MSC_VER > #pragma once > #endif // _MSC_VER > > meow; > > #endif // MYPROJECT_MYDIRECTORY_FOOBAR_HPP > > This way, all compilers get the portable idempotency guard. Because it's on the outside, GCC (and possibly other compilers) will magically recognize it and avoid opening the header file twice. And the #pragma once for MSVC also instructs it to avoid opening the header file twice, which can improve build perf and can't hurt. > > Stephan T. Lavavej > Visual C++ Libraries Developer >

7 8

Re: [boost] [Boost-build] [1.47.0] Beta 1 Release candidate files available
by Matthew Chambers 27 Jun '11

27 Jun '11

On 6/21/2011 5:28 PM, KTC wrote: > On 21/06/2011 20:47, Jürgen Hunold wrote: > >>> bjam --without-python --stagedir=x86 --build-type=complete -j2 stage >>> >>> Both result in: "...failed updating 2 targets..." >>> See<http://pastebin.com/s7Cvw2Tm> for failure. >> >> Boost.Exceptions is a library on Win32. It builds >> "libs/exception/src/clone_current_exception_msvc.cpp" >> >> which states at the top: >> >> #if defined(_MSC_VER)&& defined(_M_IX86)&& !defined(_M_X64) >> >> so this is for x86 only. On x64, no code gets compiled. >> >> The msvc linker then "helpfully" silently omits the creation of an empty dll >> and the corresponding copy step(s) fail(s). >> >> The correct fix would be to build this file only on for the<address-model>32 >> case but I don't know if this can be specified with the current Boost.Build. > > There's the problem that this is also occurring when building for x86 while on a x64 system. Does "<address-model>32:<build>no" as a requirement work? -Matt

1 0

[msm]how to distinguish region ids?
by Takatoshi Kondo 27 Jun '11

27 Jun '11

Hello, On Boost.Msm 1.46.1, When I use explicit entries and entry pseudo states, I have to provide a template argument "region id". http://www.boost.org/doc/libs/1_46_1/libs/msm/doc/HTML/ch03s02.html#d0e930 But I couldn't find documents about numbering policy of region id. I also check the example below. http://www.boost.org/doc/libs/1_46_1/libs/msm/doc/HTML/examples/DirectEntry… The example passes the region id 0 and 1. I'd like to know how to decide which region is 0(or1). Am I missing something? Thanks, Takatoshi

2 5

[1.47.0] Permission to merge filesystem HP-UX issue fix?
by Beman Dawes 27 Jun '11

27 Jun '11

OK to merge: https://svn.boost.org/trac/boost/changeset/72521 Thanks, --Beman

2 3

[1.47.0] Merge requests for iostreams + quickbook
by Daniel James 27 Jun '11

27 Jun '11

Hi, Is it okay to merge a couple of changes to release? This change activates an existing workaround for Visual C++ 8 in iostreams. https://svn.boost.org/trac/boost/changeset/72726/ This change reverts the support for escaped characters in 'simple markup' in quickbook because it breaks the regex documentation. It isn't used anywhere else, and the documentation still builds, so it's safe. Will be reintroduced for quickbook 1.6. https://svn.boost.org/trac/boost/changeset/72695/

2 1