- Boost - lists.boost.org

Boost Hana comments
by Jeremy Maitin-Shepard 21 Aug '14

21 Aug '14

First off, while I haven't thoroughly reviewed the library, I am certainly impressed by a lot of what I see. It is obvious that this library represents a lot of excellent work. Here are my miscellaneous comments: Constant type class: -------------------- Why isn't std::integral_constant an instance? There is no documentation of what the value method actually returns, just that it returns something that is a constexpr. Does value(integral<int,3>)) return int, or does it return integral<int,3> ? Perhaps the Constant type class should be "parameterized" by a type specifying the return type of value. Integral data type: ------------------- It would be better not to duplicate std::integral_constant. However, I recognize that the overloaded operators are useful and you wouldn't be able to do this with std::integral_constant (well, I suppose you still could, but as they wouldn't be found by ADL, they wouldn't be terribly useful). This rationale (assuming it is correct) should be documented though. Logical type class: ------------------- What is really gained by making this a type class at all? It seems that and_ and or_ require that all arguments be of the same "hana datatype", though actually this isn't documented anywhere and this might not even be the precise requirement (but obviously it needs to be documented very explicitly, and this probably applies to almost every method in Hana). Really we really should be able to mix multiple data types in a call to or_ or and_, though. Otherwise the usability is unreasonably limited. Of course there is then the question of return type. However, I would assume most users would be happy getting back either a Hana Integral or a bool, depending on whether it is a compile-time or runtime condition. Certainly this would be better than getting back a compiler error. Another question: how does the performance of and_ and or_ compare to the optimized verions you showed in your MPL11 talk? Pair datatype: -------------- Why duplicate std::pair? Alternatively, why duplicate list? Perhaps list should just define first and second as well? Product type class: ---------------- Why isn't a 2-element List, a 2-element std::tuple, etc. an instance? List type class: ---------------- Just as for Logical, it would be useful to be able to mix datatypes in calls to concat, zip, zip_with. The list function itself shouldn't be documented as part of the type class, since it isn't logically part of it. What about get<N>, get<N,M> accessors? I understand these can be defined in terms of the other methods, but clearly they are useful on their own.

3 4

[Concepts] Definition. Was [GSoC] [Boost.Hana] Formal review request
by Mostafa 21 Aug '14

21 Aug '14

On Thu, 31 Jul 2014 10:54:59 -0700, Robert Ramey <ramey(a)rrsd.com> wrote: > Eric Niebler-4 wrote >> On 07/29/2014 05:14 PM, Niall Douglas wrote: >>> I'm all for Concepts as in compiler enforced ones, and I'll add them >>> to AFIO when and only when C++ gets them. But for documentation they >>> don't help. >> >> Wow, I couldn't disagree more. I can't imagine how the standard >> algorithms would be specified without the use of concepts like >> "RandomAccessIterator", for instance. Clustering requirements into >> meaningful abstractions and assigning them names makes it possible to >> document library interfaces without an explosion of verbosity and >> repetition. > > +10 > > Usage of concepts is greatly: > > a) misunderstood > b) misunderestimated as to their value in design AND documentation > d) The word "concepts" is a big contributor to the problem - substitute > "type requirements" or "type constraints" for concepts. -1 to the above term substitutions for concepts. Type requirements/type constrains are not concepts. The reason concepts are misunderstood is because they have not been well defined. FWIW, here's my take on how they should be defined: Concepts are sets of types whose membership are compile-time determinable. Precisely, a concept is any set expressible in the following form: { T | T is a type and p(T ...) } where p is a (n+1)-ary compile time computable boolean function (a function in the mathematical sense, not an actual C++ function). p is said to be (or to express) the type requirements or type constrains for its associated concept. (Note: just as a set membership predicate is not its set, a type requirement is not its concept. For example, the predicate "there exists a,b and a,b are integers and b != 0 and c = a/b" does not denote the set of rational numbers, the symbol Q does.) A type T is said to model a concept C if T is a member of C. In such a case we also say that "T is a C". A concept B is said to be a refinement of a concept A if B is a subset of A. Why use concepts? Because they are helpful in documenting generic code, especially template type parameter requirements. Examples: 1) DefaultConstructible := { T | T has the method "T()" defined and accessible. } // Requirements: // T is DefaultConstructible. template <typename T> struct FooBar { ... }; std::tuple<int> is DefaultConstructible, therefore FooBar< std::tuple<int> > is well formed. Mostafa

14 57

[Container] Provide more guarantees for flat_multimap
by Vinícius dos Santos Oliveira 21 Aug '14

21 Aug '14

Currently, flat_multimap mirrors C++98's multimap concept. I'd like to propose an update to flat_multimap fulfil the C++11's multimap concept. The part I'm particularly interested are guarantees about preserving the insertion ordering for elements with equivalent keys. On ISO/IEC 14882-2011, 23.2.4 - associative container requirements, pages 741 and 742, you can see the effects for a_eq.emplace and a_eq.insert ends with "If a range containing elements equivalent to t exists in a_eq, t is inserted at the end of that range". I see no similar guarantees on the Boost's flat_multimap documentation[1]. This feature is useful, for instance, in HTTP messages, where field reordering is illegal[2] if the keys are equivalent. I did a few tests and it looks to me that flat_multimap already provides these guarantees, but before I go check the source and possibly provide patches, I need to know if there is interest in turning this accidental behaviour into intentional behaviour through documented guarantees. [1] http://www.boost.org/doc/libs/1_56_0/doc/html/boost/container/flat_multimap… [2] http://tools.ietf.org/html/rfc7230#section-3.2.2 -- Vinícius dos Santos Oliveira https://about.me/vinipsmaker

7 17

Metadata pull requests
by Daniel James 21 Aug '14

21 Aug '14

Hi, I just created about 100 pull requests to add library metadata to modules that don't already have it. Apart from them all containing the same typo, I think it went okay. I forgot to mention that the file format is documented at: http://www.boost.org/development/library_metadata.html I'll spend a bit of time improving that soon, so let me know if there's anything missing. Also, if it wasn't clear, the idea is that you can update this file if you need to change the library details. It might take a little while for the website to update, as it only happens when I run the appropriate script. thanks, Daniel

5 11

[units] Pull request and RFC: information unit definitions for Boost.Units
by Erik Erlandson 20 Aug '14

20 Aug '14

Link to PR: https://github.com/boostorg/units/pull/3 Description: Adds unit definitions for standard units of information: bit, byte, nat, hartley, shannon. Defines a new unit system boost::units::information for convenient manipulation. Also includes IEC binary prefixes: kibi, mebi, gibi, tebi, pebi, exbi Example code: https://github.com/erikerlandson/units/blob/info_units/example/information.… Unit testing code: https://github.com/erikerlandson/units/blob/info_units/test/test_informatio…

7 14

[type_traits] Interest in is_iterator type trait?
by Beman Dawes 20 Aug '14

20 Aug '14

Back in 2011, Howard Hinnant and I put together an is_iterator type trait based on code Howard had developed and used over a period of time. See http://beman.github.io/is_iterator/is_iterator.hpp for the actual code See http://beman.github.io/is_iterator/is_iterator_test.cpp for a test program. The intent was always to submit is_iterator to Boost, but we never got around to it. Now I need it in Boost.Filesystem, so I'd rather see it go in type traits than just sticking it into boost/filesystem/detail. If there is interest, and John Maddock gives the OK, I'll write docs and integrate into type_traits, then submit a pull request. --Beman

6 20

Ticket system getting spamed
by Mateusz cerkiewny 20 Aug '14

20 Aug '14

#10352 <https://svn.boost.org/trac/boost/ticket/10352> #10351 <https://svn.boost.org/trac/boost/ticket/10351> #10342 <https://svn.boost.org/trac/boost/ticket/10342> #10339 <https://svn.boost.org/trac/boost/ticket/10339> #10335 <https://svn.boost.org/trac/boost/ticket/10335> #10331 <https://svn.boost.org/trac/boost/ticket/10331> #10334 <https://svn.boost.org/trac/boost/ticket/10335> Are spamms

2 1

[log] Problem Building Boost 1.55 With GCC Using C++11 Mode Due to AVX2 Support
by Stewart, Robert 20 Aug '14

20 Aug '14

We've run into a problem building log. It doesn't appear to be a known issue and we're not sure how to work around it. We found a hack to make our builds succeed, but there may well be a better workaround and we're not certain whether there are any issues with what we've done. We tried to build Boost 1.55 with GCC 4.8.1, 4.8.3, and 4.9.1 with the following commands: ./bootstrap.sh --prefix=<prefix> --libdir=<prefix>/lib --with-toolset=gcc ./b2 --build-type=complete --layout=versioned 'cxxflags=-m64 -O2 -g -pipe' 'cflags=-m64 -O2 -g -pipe' 'linkflags=-Wl,-rpath -Wl,<prefix>/lib -L<gcc_dir>/lib64 -Wl,-rpath -Wl,<gcc_dir>/lib64 -m64' -d+2 -j3 ./b2 --build-type=complete --layout=versioned 'cxxflags=-m64 -O2 -g -pipe' 'cflags=-m64 -O2 -g -pipe' 'linkflags=-Wl,-rpath -Wl,<prefix>/lib -L<gcc_dir>/lib64 -Wl,-rpath -Wl,<gcc_dir>/lib64 -m64' -d+2 -j3 install The errors we get come from this command during the build: gcc.compile.c++ bin.v2/libs/log/build/gcc-4.9/release/link-static/log-api-unix/threading-multi/dump_avx2.o "g++11" -ftemplate-depth-128 -O3 -finline-functions -Wno-inline -Wall -pthread -fno-strict-aliasing -ftemplate-depth-1024 -m64 -O2 -g -pipe -march=core-avx2 -mavx -mavx2 -fabi-version=0 -m64 -O2 -g -pipe -DBOOST_ALL_NO_LIB=1 -DBOOST_CHRONO_STATIC_LINK=1 -DBOOST_FILESYSTEM_STATIC_LINK=1 -DBOOST_LOG_BUILDING_THE_LIB=1 -DBOOST_LOG_USE_NATIVE_SYSLOG -DBOOST_LOG_WITHOUT_EVENT_LOG -DBOOST_SPIRIT_USE_PHOENIX_V3=1 -DBOOST_SYSTEM_NO_DEPRECATED -DBOOST_SYSTEM_STATIC_LINK=1 -DBOOST_THREAD_BUILD_LIB=1 -DBOOST_THREAD_DONT_USE_CHRONO=1 -DBOOST_THREAD_POSIX -DBOOST_THREAD_USE_LIB=1 -DDATE_TIME_INLINE -DNDEBUG -I"." -c -o "bin.v2/libs/log/build/gcc-4.9/release/link-static/log-api-unix/threading-multi/dump_avx2.o" "libs/log/src/dump_avx2.cpp" There are many lines of output; I've elided similar messages to highlight the unique ones: {standard input}: Assembler messages: {standard input}:611: Error: operand type mismatch for `vpsrldq' ... {standard input}:819: Error: no such instruction: `vextracti128 $0x1,%ymm2,%xmm2' ... {standard input}:1087: Error: no such instruction: `vextracti128 $0x1,%ymm3,48(%r15)' ... {standard input}:1312: Error: operand type mismatch for `vpaddb' While there might be a way to actually get b2 to disable AVX2 support, we hacked libs/log/build/Jamfile.v2 with these commands: sed -i 's:dump_avx2::g' libs/log/build/Jamfile.v2 sed -i 's:<define>BOOST_LOG_USE_AVX2::g' libs/log/build/Jamfile.v2 Those remove dump_avx2 from BOOST_LOG_COMMON_AVX2_SRC and remove the definition of BOOST_LOG_USE_AVX2, which prevents compiling enable_avx2(). Is there a better way to solve or work around this problem? _____ Rob Stewart robert.stewart(a)sig.com Software Engineer, Middleware using std::disclaimer; Susquehanna International Group, LLP http://www.sig.com ________________________________ IMPORTANT: The information contained in this email and/or its attachments is confidential. If you are not the intended recipient, please notify the sender immediately by reply and immediately delete this message and all its attachments. Any review, use, reproduction, disclosure or dissemination of this message or any attachment by an unintended recipient is strictly prohibited. Neither this message nor any attachment is intended as or should be construed as an offer, solicitation or recommendation to buy or sell any security or other financial instrument. Neither the sender, his or her employer nor any of their respective affiliates makes any warranties as to the completeness or accuracy of any of the information contained herein or that this message or any of its attachments is free of viruses.

2 2

[git] Pushing commit(s) after pulling
by Edward Diener 20 Aug '14

20 Aug '14

I share a library with others and make some commit to a branch of that library. Then I try to 'push' the commit, only to be told that there are changes that have been made which I need to 'pull'. So I 'pull' the latest changes for that branch, and luckily there is no conflict(s) with my local commit(s). What do I do now ? I can 'push' my changes successfully but then the history of my particular changes appears to be lost. Am I supposed to revert my changes to the local repository by doing a 'reset' and then reapply those changes ? That seems onerous. What is the right way to 'push' my changes and still preserve the history of each one of them on the remote repository ?

5 9

[mpl] Proceeding with modularization
by Andrey Semashev 20 Aug '14

20 Aug '14

Hi, Now that 1.56 is out I believe we can proceed with modularization. I'd like to complete Boost.MPL decomposition into Boost.MPL.Core and Boost.MPL that I started some time ago. I can see two ways of doing this: I can keep the two parts in the separate git repositories or I can move Boost.MPL.Core to a sub- library within the same git repo. I'd prefer the second approach since this allows to create a single pull request and keeps the library history in one place, but maybe there are other considerations?

3 3