April 2016 - Boost - lists.boost.org

Invoking a library's build//stage ignores --stagedir?
by Peter Dimov 14 Apr '16

14 Apr '16

Let's suppose that for some random reason I want to stage Boost.System into c:\stage with the current directory being outside the Boost root, which is at c:\projects\boost-git\boost: C:\Users\Peter Dimov>b2 -sBOOST_ROOT=c:\projects\boost-git\boost --stagedir=c:\stage c:/projects/boost-git/boost/libs/system/build//stage This doesn't work. Or rather, it works, but stages into C:\projects\boost-git\boost\libs\system\stage, ignoring the --stagedir. I have no idea why. I tried to debug with -d+7, and BOOST_STAGE_LOCATE is set correctly from --stagedir, but then Boost.System is staged to its local stage directory. Does anyone have an idea what's wrong here?

2 3

[multiprecision] Bug in double to cpp_bin_float conversion?
by Rob Conde 14 Apr '16

14 Apr '16

Consider the following code: #include <boost/multiprecision/cpp_bin_float.hpp> #include <iostream> int main() { typedef boost::multiprecision::cpp_bin_float_quad float128_t; float128_t f1(1e-9); float128_t f2("1e-9"); std::cout << std::setprecision(std::numeric_limits<float128_t>::max_digits10); std::cout << f1 << std::endl; std::cout << f2 << std::endl; return 0; } I'm getting as output: 1.000000000000000062281591457779856419e-09 9.999999999999999999999999999999999325e-10 The conversion of the double literal is not as expected. On wand box: http://melpon.org/wandbox/permlink/uSGNQZAEEytmb3ao Is this a bug? If not, what would be the correct way to get my expected behavior? Thanks, Rob Conde

3 3

[metaparse] how do I log a boost ticket?
by Aparna Kumta 13 Apr '16

13 Apr '16

I want to submit a Boost ticket at https://svn.boost.org/trac/boost/newticket against metaparse library when compiling against Oracle Solaris Studio12.5. However, I do not see 'metaparse' under the 'Component' tab. Should I be logging a bug elsewhere? I have gone ahead and submitted a PR. https://github.com/boostorg/metaparse/pull/6. Thanks, Aparna

2 4

Serialize a an image with Boost & GIL
by Visual Designer - Digital Team 13 Apr '16

13 Apr '16

Hi, I'm currently using boost serialization in order to save/load my data structure. The problem is that it contains some images and even with the ZIP compression the files are really big (4x bigger than the jpeg). So, I'm looking for a way to compress a RGB/A image (HDR or not) (without file operation) in memory. If possible without using any image compression library, we already use OpenImageIO but it is not possible for now ! Any idea ? I have found the GIL library, but can't find how to: 1) convert raw to GIL image 2) serialize the GIL image 1) deserialize the GIL image 2) convert the GIL to raw image Of course, the goal is to have a good compression without writing temporary file ! Thanks --- This email has been checked for viruses by Avast antivirus software. https://www.avast.com/antivirus

2 3

[endian] Some suggestions
by Tatsuyuki Ishi 13 Apr '16

13 Apr '16

Mr. Beman, Here are some suggestions for the endian library. 1. Bring back float: make a proper reason for abandon You told about NaN: but it shouldn't apply since you store it in char buffer. Provide a proper reason: float is still needed since this is the only library that wraps details (platform endianness detection+compiler builtin optimizations). Float points are indispensable in data formats. I have to do a reinterpret_cast now as a workaround. 2. Remove n_bits(endian buffers): completely useless This just look like an alternative to sizeof, and always /8'd when used. What's the meaning of that? 3. Minor API suggestions Move operator type to endian_buffer: it's not related to arithmetic but just used because it can act like an integer. Non-const data(): I'm tired to do reinterpret_cast. Look at C++11 vector for a reference. Move stream operators to endian_arithmetic: I first thought that was for reading binary. That's basically unused.

9 19

Re: [boost] [endian] Some suggestions
by Tatsuyuki Ishi 13 Apr '16

13 Apr '16

> > 1. Bring back float: make a proper reason for abandon > > You told about NaN: but it shouldn't apply since you store it in char > > buffer. > > > > The problem isn't when storing into the char buffer, but later on a > different machine or compiler when the char buffer has to be converted to a > float type. I know nothing other than float format differences, and my target platforms are using iec559. I think reinterpret_cast should work: char data shouldn't be changed as long as stored as char. > > In testing, I even ran into a case were on the same machine and compiler > the float data failed to round-trip correctly if the program that created > the file was compiled with a different set of optimization options that the > program the read the file. > I will be happy if you provide some example. > > > > > 2. Remove n_bits(endian buffers): completely useless > > This just look like an alternative to sizeof, and always /8'd when used. > > What's the meaning of that? > > > > Please give a specific code example from the docs of what you think should > be changed. Or submit a pull request. > Since this is breaking API, I'd like you to make the decision to change it or not. For detail, this is the endian_buffer definition: template <order Order, class T, std::size_t n_bits, align A = align::no> class endian_buffer; n_bits looks quite useless here. It's basically sizeof*8. > > > > > 3. Minor API suggestions > > Move operator type to endian_buffer: it's not related to arithmetic but > > just used because it can act like an integer. > > > > Again, what code is it you think should be changed? Operator value_type()? > Please submit a pull request. > Exactly: operator value_type(). This will provide much more convenience. An example: when you specify the size for the array: endian_buffer size; dynarray d(size.value()); vs dynarray d(size); > > > Non-const data(): I'm tired to do reinterpret_cast. Look at C++11 vector > > for a reference. > > > > What is the motivation? What are the safety implications? What are the > alternatives? The C++11 (and also C++17) addition of non-cost data() > members to some standard library containers was supported by motivation and > safety analysis. > Seems endian_buffer is targeted for using in structs. Then is there any problem to provide a wrap of reinterpret_cast? At least, it should be safe by design. > > > Move stream operators to endian_arithmetic: I first thought that was for > > reading binary. That's basically unused. > > > > I must be missing something. Why wouldn't stream operators apply to endian > buffers? A code example might help. > Quite confusing. I first thought that was binary r/w operators that reads/writes the stream and store the value. The problem is that, there's no method provided for reading from binary, so this is also a reason for the previous data() suggestion.

2 1

[iostreams] Maintainer not working
by Tatsuyuki Ishi 13 Apr '16

13 Apr '16

Seems nobody is doing code review for iostreams since February. Basically I want a PR being merged: https://github.com/boostorg/iostreams/pull/25 This patch and some others looks good to me. It fixes a quite annoying warning. Can anybody with permission merge it?

1 0

[Coroutine] What happened to WINFIBERS?
by Tobias Loew 12 Apr '16

12 Apr '16

When testing 1.61 beta I couldn't find any reference to BOOST_USE_WINFIBERS in the sources and also the WinFibers implementation-file "execution_context_winfib.ipp" was missing. Has support for windos-fibers been removed? Tobias -- View this message in context: http://boost.2283326.n4.nabble.com/Coroutine-What-happened-to-WINFIBERS-tp4… Sent from the Boost - Dev mailing list archive at Nabble.com.

4 6

Compiling boost_thread on CentOS 7 32 bit system
by Kristian 11 Apr '16

11 Apr '16

Hi, I am trying to build boost 1.60 on a CentOS 7 32 bit system. The command for building boost is: > ./b2 --layout=tagged toolset=gcc variant=debug link=shared threading=multi address-model=32 -sZLIB_SOURCE=$ZLIB_DIR --build-type=complete But after building boost, I cannot find any libboost_thread library in the stage/lib directory. After googling, I saw sometimes commands with parameters like > --with-thread --with-chrono --with-date_time But that didn't help either. After the second try, I only saw these files: > libboost_chrono-mt-d.so > libboost_chrono-mt-d.so.1.60.0 > libboost_chrono-mt.so > libboost_chrono-mt.so.1.60.0 > libboost_system-mt-d.so > libboost_system-mt-d.so.1.60.0 > libboost_system-mt.so > libboost_system-mt.so.1.60.0 Can someone tell me, what I am missing here?

2 1

[Nowide] Review Manager Needed
by Artyom Beilis 11 Apr '16

11 Apr '16

Boost.Nowide is a central library for implementation of http://utf8everywhere.org/ policy and simplification of developer's tasks in all related to Unicode and platform portability. http://cppcms.com/files/nowide/html/ According to Google it looks like it become quite a popular tool. It still looks for a review manager. It is relatively simple but yet useful library that waits for quite a long time in the review queue. Volunteer for review manager is needed. Artyom Beilis

2 2