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

boost::multiprecision::cpp_int encoding and decoding little endian byte arrays
by Steffen Heil (Mailinglisten) 05 Sep '13

05 Sep '13

Hello I need to decode/encode cpp_int values from/into byte arrays. The encoding must be shortest little endian two's complement representation of the value. For example: 00 = 0 01 = 1 ff = -1 ff:7f = 32767 01:80 = -32767 00:80:00 = 32768 00:80 = -32768 01:80:00 = 32769 ff:7f:ff = -32769 ff:ff:ef:9c:d2:40:43:da:4f:40:6b:06:6a:cd:92:c2:93:bf:a4:5c:c3:f1:29:63:1d = 184467440737095516151844674407370955161518446744073709551615 01:00:10:63:2d:bf:bc:25:b0:bf:94:f9:95:32:6d:3d:6c:40:5b:a3:3c:0e:d6:9c:e2 = -184467440737095516151844674407370955161518446744073709551615 I found a way to archive my goal (see below), but it seems way to complex and too inefficient. For encoding I basically converting the cpp_int to a uint8_t to get 8 bit and then shift the cpp_int by 8, creating a new cpp_int... Twice (once to compute it's length, again to encode it). For decoding I already shift the byte read to the "right" position and "|=" it into the existing cpp_int. Isn't there an easier way access the binary representation of an cpp_int? Regards, Steffen --- #include <iostream> #include <boost/multiprecision/cpp_int.hpp> using namespace std; using namespace boost::multiprecision; size_t binLength( cpp_int value ) { if ( value.is_zero() ) return 1; if ( value.sign() < 0 ) value = ~ value; size_t length = 0; uint8_t lastByte; do { lastByte = value.convert_to<uint8_t>(); value >>= 8; length ++; } while ( ! value.is_zero() ); if ( lastByte >= 0x80 ) length ++; return length; } void binEncode( cpp_int value, uint8_t* output, size_t length ) { if ( value.is_zero() ) *output = 0; else if ( value.sign() > 0 ) while ( length -- > 0 ) { *( output ++ ) = value.convert_to<uint8_t>(); value >>= 8; } else { value = ~ value; while ( length -- > 0 ) { *( output ++ ) = ~ value.convert_to<uint8_t>(); value >>= 8; } } } cpp_int binDecode( uint8_t* input, size_t length ) { boost::multiprecision::cpp_int result( 0 ); int bits = - 8; while ( length -- > 1 ) result |= (boost::multiprecision::cpp_int) *( input ++ ) << ( bits += 8 ); uint8_t a = *( input ++ ); result |= (boost::multiprecision::cpp_int) a << ( bits += 8 ); if ( a >= 0x80 ) result |= (boost::multiprecision::cpp_int) - 1 << ( bits + 8 ); return result; } void test( string s ) { cpp_int x( s ); uint8_t buffer[ 100 ]; size_t length = binLength( x ), index = 0; binEncode( x, buffer, length ); cpp_int y = binDecode( buffer, length ); cout << hex << setw( 2 ) << setfill( '0' ) << (int) buffer[ 0 ]; while ( -- length > 0 ) cout << ':' << setw( 2 ) << setfill( '0' ) << (int) buffer[ ++ index ]; cout << " = " << dec << setw( 1 ) << x << endl; if ( x != y ) cerr << "failed" << endl; } int main( int argc, const char* argv[] ) { test( "0" ); test( "1" ); test( "-1" ); test( "32767" ); test( "-32767" ); test( "32768" ); test( "-32768" ); test( "32769" ); test( "-32769" ); test( "184467440737095516151844674407370955161518446744073709551615" ); test( "-184467440737095516151844674407370955161518446744073709551615" ); }

2 1

Re: [Boost-users] Why fusion::map<> types not the same?
by Sergey Zhuravlev 04 Sep '13

04 Sep '13

This is bug. You should add ticket to bugtracker. Try to use boost\fusion\result_of::as_map as workaround.

2 1

[Log] How to use a channel that is constant per file?
by Timo Schmiade 03 Sep '13

03 Sep '13

Hi all, I want to use boost's log library to output messages to the console which feature a per-message severity and a per-file channel. I already tried two approaches, none of which was satisfying: === 1. approach === The class boost::log::sources::severity_channel_logger supports both severity and channel and I could define it in each source file like so: --- 8< --- BOOST_LOG_INLINE_GLOBAL_LOGGER_DEFAULT( log, boost::log::sources::severity_channel_logger< boost::log::trivial::severity_level, std::string > ) --- >8 --- Unfortunately, using it like this requires me to redundantly add the channel to each log message, which is very inconvenient because it is the same for all logs in the file. --- 8< --- // Message.cpp BOOST_LOG_CHANNEL_SEV(log, Message, trace) << "Hello, world"; --- >8 --- === 2. approach === I defined my logger like this: --- 8< --- BOOST_LOG_INLINE_GLOBAL_LOGGER_DEFAULT( log, boost::log::sources::severity_logger< boost::log::trivial::severity_level > ) --- >8 --- Then, in each file, I manually set the "Channel" attribute for the logger, like so: --- 8< --- // Message.cpp log::get().add_attribute( "Channel", boost::log::attributes::make_constant("Message")); --- >8 --- This works, but as the "log" object seems to be stored globally, I would have to "invent" new names for the logger object in each file. This is both insecure and prevents the usage of macros (I think). So here's the question: What would be the recommended / a convenient way to add a per-file constant channel to log messages? Thanks in advance! Timo

2 2

Why fusion::map<> types not the same?
by niXman 03 Sep '13

03 Sep '13

Hi, Why 'type' is not same as 'expected' in the example below? using origin = boost::fusion::map< boost::fusion::pair<int, int> >; using expected = boost::fusion::map< boost::fusion::pair<int, int> ,boost::fusion::pair<char, char> >; using type = boost::fusion::result_of::push_back< origin ,boost::fusion::pair<char, char> >::type; static_assert(std::is_same<expected, type>::value, "error!"); The second question is, how can I get the same type as 'expected' when using 'result_of::push_back<>::type' ? -- Regards, niXman ___________________________________________________ Dual-target(32 & 64-bit) MinGW compilers for 32 and 64-bit Windows: http://sourceforge.net/projects/mingwbuilds/ ___________________________________________________ Another online IDE: http://liveworkspace.org/

2 2

[asio][ssl] why the error of " decryption failed or bad record mac" happen?
by lx 03 Sep '13

03 Sep '13

I use the asio and ssl , the error message of " decryption failed or bad record mac" happen some tiems. What are the possible causes? Thank you

1 1

[spirit] Parsing std::string phrase into attribute struct
by Michael Powell 03 Sep '13

03 Sep '13

Hello, Question re: parsing string into a Fusion-adapted struct. First, magic! Wow, I like it. Question about the immediate result, true/false. Let's say the operation fails for whatever reason. In our use-case, we'll be reading some text in from a Linux /sysfs -provided API. We'll parse the contents into a struct instance. If the operation fails for whatever reason, does not match the grammar, are any values mapped into the structure? My first guess, is possibly: a partial (though still failing) match, would partially populate the instance? I am using the employee-example as our boiler plate. http://www.boost.org/doc/libs/1_54_0/libs/spirit/example/qi/employee.cpp The outcome of which would be whether to consider the instance valid for reporting further into user-space, or what else to do with the outcome from there. Thank you... Regards, Michael Powell

2 1

Changing format of program_options
by Heikki Virtanen 02 Sep '13

02 Sep '13

Hi, I have tried to read config file using boost::program_options where I have large floating pointing numbers like [test_section] test = 1.432e43 but seems that I get bad an error message "boost::bad_any_cast: failed conversion using boost::any_cast" when I am reading the value from program_options::variables_map. Reading works perfectly when I am using numbers like [test_section] a = 13.62 but scientific format fails. Should I tell to variables_map somehow what is the format of floating point numbers or should I do something else? -Heikki

1 0

Using Boost.Thread with lowlevel system software: async interruption question
by Nikita V. Youshchenko 02 Sep '13

02 Sep '13

Hi I'm developing lowevel system software, that runs under linux and communicates with custom linux drivers for custom hardware. I'm evaluating possibility to switch from explicit use of pthread interface to Boost.Thread. After reading documentation, I can't understand if there is any solution for async thread interruption with Boost.Thread. First use case is - need to interrupt thread that is sleeping in custom dirver's ioctl() call. Drivers are well-written, in terms that any sleep inside ioctl()s is immediately interrupted on signal arrival. This perfectly couples with pthreads' PTHREAD_CANCEL_ASYNCHRONOUS cancellation type - which is based on signals. Thus, to make threads cancellable while sleeping inside ioctl(), it is enough to set cancellation type to PTHREAD_CANCEL_ASYNCHRONOUS before entering system call. Second use case is - need to interrupt thread that is running 3rd party legacy code. The only method I know to physically communicate termination request to thread running unaware code is signal. Again, this couples well with PTHREAD_CANCEL_ASYNCHRONOUS. Is it possible to solve these two use cases with Boost.Thread? Per Boost.Thread documentation, thread interrupts only in predefined interruption points. Which is similar to pthreads' PTHREAD_CANCEL_DEFERRED, and is not what I'm looking for. I imagine I could play with signal handler that does longjump() to some point where I manually raise boost::thread_interrupted exception. Will that work? Sure this will loose any destructor calls for objects created in call stack "below setjmp point" - but in first use case I can ensure that no cleanup below setjmp is required, and in second use case this looks unavoidable... Any better ideas? I do not have any portability needs in this software - drivers are linux-only anyway. So linux-only solution is perfectly ok. Nikita Yushchenko

2 1

On line alternative
by Brian Wood 01 Sep '13

01 Sep '13

I've been watching some of the videos from the last C++Now conference, including one that talked about the serialization library in Boost. The presenter mentioned some alternatives to that library, but he didn't mention the C++ Middleware Writer (CMW). The CMW has support for some Boost libs: http://webEbenezer.net/supportedTypes.html If possibe we'll add to that list in the future. -- Brian Ebenezer Enterprises http://webEbenezer.net

1 0

[chrono] float-based milliseconds?
by Michael Powell 31 Aug '13

31 Aug '13

Hello, Or any of the chrono-steady-clock-based time units... I am noticing some what appear to be rounding errors outputting resulting millisecond values using the boost::chrono::milliseconds(x). It's a design-decision, most if not all of our floating-point values are single-precision-float, don't need to be double-precision-float. When we output the milliseconds(x).count() values, however, what I am finding is that the values are sometimes inaccurate, the same value in 3.0, for instance, output as 2.99995, stuff like this. Is there a way to tell milliseconds what precision floating point it is? float or double? We are targeting ARM, BTW, without hardware floating point support, which is the main reason, I'd like to keep things single-precision float, if at all possible. Thank you... Regards, Michael Powell

3 6