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

improve TypeTraits
by Gero Peterhoff 04 Oct '20

04 Oct '20

Hello John, you have probably already read that I think some new TypeTraits are useful. Here are the details: a) very useful: precise type classification - this is the basis of every abstraction and should definitely be implemented - is_character: contains only characters (char32/16/8_t, wchar_t, char); is_un/signed_character - is_integer: contains only integers and no characters and no bool (u/int8..X_t); is_un/signed_integer - is_logic: contains only logical types (bool and boost::tribool, *)) - is_string: contains everything that can be used as a string, but needs these specializations: 1) is_bounded_string (similar to is_bounded_array) 2) is_unbounded_string (similar to is_unbounded_array) 3) is_static_string (for boost::basic_static_string) 4) is_string_ref (for boost::basic_string ref) 5) is_string_view (for std::basic_string_view) 6) is_string_class (for std::basic_string) - character_of: the underlying character type based on is_string b) very useful - remove_cv_ptr (similar to remove_cv_ref) - make_same_signed (similar to make_un/signed): changes the sign depending on a type - conditional_signed: changes the sign depending on a truth value - is_same_signed: same sign c) also makes sense - is_tuple - is_rational (similar to boost::is_complex) - is_binary: if base==2 (numeric_limits::radix==2) - is_decimal: if base==10 (numeric_limits::radix==10) d) - is_native: a type is natively available on the platform (e.g. int64_t: Linux32:false, Linux64:true; float80_t: x86:true, else:false) - is_vectorisable: a type can be vectorized natively on the platform - there are basic SSE/AVX/SVE/... functions However, I have a problem: boost/std::is_integral contain bool. Perhaps you can do that like this, but it leads to problems with is_arithmetic, since is_arithmetic is based on is_integral and thus contains bool. I think the point of is_arithmetic is to be able to calculate numerically with the type - so the scenarios are not useful: abs(bool), sin/cos/...(bool), etc. The same applies to characters. There are two possible solutions: - redefine is_arithmetic -> is_integer || is_floating_point - a new TypeTrait is_numeric (e.g.) -> is_integer || is_floating_point There are some differences between boost- and std-TypeTraits. Can you please include that in the documentation? Furthermore there seem to be errors with char8_t, e.g. boost::is_unsigned<char8_t> -> false, std::is_unsigned<char8_t> -> true (correct). Summary With these measures, a large number of template functions/classes can then be tailored precisely to the application and incorrect use can be prevented. regards Gero *) I've written a couple of classes for logic: bool3_t (a modern variant of boost::tribool), bool4_t (http://www.filosofia.unimi.it/dagostino/wp-content/uploads/2017/05/Belnap.p…) and bool2_t (for completeness)

1 0

Using Boost.Random with fixed-point numbers
by John W 03 Oct '20

03 Oct '20

I have a fixed-point number class (based on libfixmath) which in code largely behaves like a float due to various operator overloading. Now, I am trying to generate random numbers with them. Uniform random is easy enough — just reinterpret some random bits as my fixedpoint type. However, I'd like to have other distributions, like boost::random::normal_distribution. But from what I can tell, Boost.Random might have some baked-in assumptions about only working on float/double/long double. Is that correct? Or are there ways I can more-or-less transparently use boost::random::xxx distributions with my custom type? Basically, I want to know how much the <RealType> template argument can be abused. In std::random, there is some clear language[1], saying: "The result type generated by the generator. The effect is undefined if this is not one of float, double, or long double." and I believe boost::random aims to be in agreement with std::random, generally. But I could not find such clear language in the Boost.Random docs. Thanks -John [1] https://en.cppreference.com/w/cpp/numeric/random/uniform_real_distribution

2 1

[Boost.Compute]: double3 - typedef redefinition with different types
by Manfred Doudar 30 Sep '20

30 Sep '20

Hello List, [Posting over from the Slack channel which didn't get any attention]: Any Boost.Compute users here that might be able to help with the following code snippet - BOOST_COMPUTE_ADAPT_STRUCT(double3, double3, (x, y, z)) BOOST_COMPUTE_FUNCTION(bool, CompareX, (double3 a, double3 b), { return a.x < b.x; }); boost::compute::vector<double> points; ... auto it = boost::compute::min_element(points.begin(), points.end(), CompareX, queue); results in the kernel compilation failure - --- build log --- <kernel>:9:3: error: typedef redefinition with different types ('struct double3' vs 'double __attribute__((ext_vector_type(3)))') } double3; ^ cl_kernel.h:154:1: note: previous definition is here __NV_DECLARE_VECT_TYPES(double, double) ^ cl_kernel.h:114:61: note: expanded from macro '__NV_DECLARE_VECT_TYPES' typedef __attribute__(( ext_vector_type(3) )) CTYPE CLTYPE##3; \ ^ <scratch space>:53:1: note: expanded from here double3 ^ ^@ Thanks, /M

1 0

unable to update boost version
by Shiladitya Porey 24 Sep '20

24 Sep '20

Respected Sir/Madam, I am running a code where the minimum required version of the boost package is 66. But the version of boost in my laptop is 65. I tried the command > sudo apt install libboost-filesystem-dev libboost-log-dev But it failed to update. Then I tried to download the package from the website and followed the instructions to install it. But still the version is still 65. Could you kindly help me, please? Thanking you, Sincerely, Shiladitya Porey -- Shiladitya Porey Graduate student Laboratory of Cosmology and Elementary Particle Physics <http://astro.nsu.ru/> Novosibirsk State University <https://english.nsu.ru/>, Russia Office: Room no. : 434,Old Building Mobile: 8 913 462-44-32 https://sites.google.com/a/g.nsu.ru/shiladitya/home

3 2

[log]: Filter declaration in settings files, associativity and operator precedence
by Bill Somerville 18 Sep '20

18 Sep '20

Hi all, I am experimenting with using multiple channels with Boost.Log. I am using the boost::trivial severity levels, the common attributes, and logging using BOOST_LOG_SEV() using a mix of global and class specific loggers. I am also using boost::log::init_from_stream(std::istream&) to configure sinks from a settings file. I would like to be able to define a sink filter that allows increasing the logging verbosity on just one channel but I am struggling to write succinct filer expressions. E.g given three channels A, B, and C I would like to write: Filter="(%Channel% = \"B\" and %Severity% >= trace) or %Severity% >= warning" to increase the verbosity to trace only on channel B, but this doesn't work. I am finding I have to write: Filter="(%Channel% = \"B\" and %Severity% >= trace) or (%Channel% != \"B\" and %Severity% >= warning)" Is this expected behaviour? If so then declaring more complex filtering will become burdensome. Regards Bill Somerville.

1 0

[Urgent request] : Boost default_locale() is causing an application crash!
by kamallochan Jena 17 Sep '20

17 Sep '20

Hi All, Thank you for your kind support! In one of our applications, we are using Boost library and recently found an application crash because of the locale object return from below highlighted code from Boost source code (path.cpp). Boost library source code (File name: path.cpp and line No: 865) seems to return NULL locale object [image: image.png] As per call stack, Application is found to be crashed when std::use_facet() tried to Getfacet id from the passed locale object : [image: image.png] Please Note: At above line 934, path_locale() calls to the above default_locale() which return an locale object with NULL value. *Can someone please answer the below queries:* 1>> In which scenario the code at std::locale(global_loc, new windows_file_codecvt);", mentioned above will return a locale object with NULL values ?? 2>> How to enable BOOST_ASSERT_MSG, so below code will execute ?? Example: BOOST_ASSERT_MSG(&path_locale(), "boost::filesystem::path locale initialization error"); 3>> In which conditions a locale object can be null ?? Thank you in Advance!!, please let me know if any further details are required on this issue. Regards, Kamal

3 4

[multiprecision] 128 bit integer support on Windows
by Neill Clift 15 Sep '20

15 Sep '20

Hi, I managed to get boost multi-precision to work with 128 bit integers on Windows using clang. I had to change this in clang.hpp: #if defined(__SIZEOF_INT128__) && !defined(BOOST_NVCC_CXX03) && !defined(_MSC_VER) <--- remove this last thing # define BOOST_HAS_INT128 #endif Clang-cl defines _MSC_VER for compat I guess. I had to pull in the library clang_rt.builtins-x86_64.lib for two routines needed for division and remainder. For my particular application I only saw a 10% improvement in performance. I am wondering if there are other things that need to be changed to get better performance? My program deals with integers of various sizes so it might be an artifact of the sizes of integers I mostly do operations on. I was expecting a bigger return. I am seeing the most expensive routine being _udivmodti4 on profiling. Division wasn't showing up as much in the traces before I made the change. Thanks. Neill.

1 0

Can boost.asio read/write io of files asynchronously?
by lampahome 02 Sep '20

02 Sep '20

As title, I read boost.asio documents and generally said io about http/network. I don't see any info. of io of files. Does boost.asio support read/write io of files asynchronously?

5 11

Program-options and priority
by DV Henkel-Wallace 31 Aug '20

31 Aug '20

Hi. I'm finding program_options pretty handy but I have a usage question which is not clear from the documentation. Is a hierarchy of sources not possible if I use value<T>(&my_arg)? My use case is the following: I have a hierarchy of sources: command line, environment variables, user-specified config file or default config file, and lastly a system default. I just set my_arg to whatever default is appropriate and then let parse_command_line do all the work. This is convenient because I can have a module simply define an options_description for callers to use. This works fine unless the option is specified twice (say command line and then environment). The last of the two values is used rather than the first one. Is this hierarchy only possible if I use value<T>() and not value<T>(&my_var)? That will add some complexity for callers. Thanks, d

1 0

Re: [Boost-users] Beast for IMAP
by Kenneth Porter 29 Aug '20

29 Aug '20

--On Saturday, August 29, 2020 11:27 AM +0300 Zhivko Vasilev via Boost-users <boost-users(a)lists.boost.org> wrote: > And IMAP complexity by itself is a "Beast" compared with SMTP. > > Not from boost libraries, but have you look at these libraries if you need > IMAP support. > https://github.com/dinhviethoa/libetpan > https://github.com/MailCore/mailcore2 > > As far as I know, most of the "modern" email apps which haven't used > proprietary code are based on these libraries. Thanks. I'd seen those and also this one: https://www.vmime.org/ I'm coming from using Mulberry and Thunderbird, which have their own IMAP implementation, but neither of those look easy to extract as standalone libraries. In addition to the complexity of IMAP, there's the need for authentication and encryption support, so there's a stack of additional libraries like GnuTLS, cyrus-sasl, and openssl needed. Some of these are hard to build on Windows. Fortunately the new Windows Linux Subsystem allows one to run Linux-based GUI programs in a VM and display on the Windows desktop. So I may go that route.

1 0