- Boost - lists.preview.boost.org

[serialization] [libstdc++] [detail] utf8_codecvt_facet fixes broke serialization test_array_xml_warchive
by Beman Dawes 13 Sep '14

13 Sep '14

The specific crash message is: *** Error in `../../../bin.v2/libs/serialization/test/test_array_xml_warchive.test/clang-linux-libstdcpp/debug/test_array_xml_warchive': double free or corruption (!prev): 0x00000000015f6f90 *** It occurs for clang, gcc, and intel compilers, using libstdc++. It does not occur with clang using libc++. It does not occur with msvc 10.0, 11.0, or 12.0. None of the other libraries (filesystem, log, program_options, property_tree) that use utf8_codecvt_facet are failing on develop. This is mainly a heads up to let people know that the serialization problem in develop is being worked on, but it may be a day or two before I have a fix. --Beman

4 10

[asio] RFC on new reliable UDP library
by Niall Douglas 12 Sep '14

12 Sep '14

Dear Boost and ASIO users, CC: boost(a)lists.boost.org, asio-users(a)lists.sourceforge.net I am writing to ask for comments on the design of and interest in a generic reliable messaging library based on ASIO. Such a library could bring to ASIO very considerably improved data transfer performance over TCP, plus the ability to ignore NAT firewalls. The design imperatives for such a library are proving tricky to reach consensus upon, so perhaps the community might be able to help. Firstly, what features should such a reliable messaging library have? Here are some for you to consider: * Should it have wire transport backends with an abstraction from wire transports, so a UDT backend could be interchangeable with TCP for example with just a recompile needed? Upper level code need therefore not consider what format the wire transport takes, just that messages are reliably delivered or we are told of the connection failure. Some mechanism for dealing with out of order messages, and message expiry in case congestion control excessively delays outbound messages would also be included. You should be aware that congestion control for a reliable messaging library means one has no choice but to pace message construction (i.e. block/EWOULDBLOCK handing out new message send templates), otherwise one ends up buffering more data for sending than potentially can ever be delivered. This is because we cannot drop reliable messages, nor can we pace writes, so pacing new message construction is the only alternative. * Should it be zero copy for both receipt and send? If so, we'll have to break with the existing ASIO API somewhat - namely that MutableBufferSequence will need to now become MutableFileOffsetBufferSequence so one can supply scatter/gathers of user space buffers and kernel side buffers. We'll also have to templatise the buffer management so client code can supply the virtualisation (e.g. an iterator) which remaps inbound message fragments into apparently contiguous regions. By the time we have done these, the API looks quite different. * Should it be written using C++ 14 idioms instead of the existing 03 idioms ASIO uses? If so, basing read and write handlers on the C++ 17 proposed experimental::expected<T, E> (http://www.hyc.io/boost/expected-proposal.pdf) makes much more sense than the fixed ASIO handler prototype void (*handler)(const boost::system::error_code& ec, std::size_t bytes_transferred) - it is more powerful, more expressive, optionally integrates cleanly with no-alloc future-promise and the rest of the C++ 11/14 STL, and lets handlers use any mix of error reporting mechanism they choose. It also opens the door to a much cleaner and tighter integration of the future C++ 17 Networking TS with Concurrency TS. Equally, especially if combined with the earlier changes, it goes further and further away from the present ASIO design. * How concurrent should it be capable of? Unlike TCP where it makes not much sense to have concurrent writers, UDP allows enormous parallelism, potentially into queue depths of thousands on decent NICs, sufficiently so where a strand based design may simply not scale. We believe that with some effort a completely lock free design which can make full use of the batch datagram send and receive APIs on Linux and Windows is possible - we believe that such a design would scale close to linearly to CPU cores available. Unfortunately, as batch datagram send/receive is not supported by ASIO and on POSIX the main dispatcher appears to be locked by a mutex, this would require a custom code path and i/o service. If combined with all of the above, it starts to look like a rewrite of the core of ASIO which seems very big just to implement reliable messaging. I guess that is good for now for starting discussion. My thanks for your time in advance. Niall -- ned Productions Limited Consulting http://www.nedproductions.biz/ http://ie.linkedin.com/in/nialldouglas/

9 17

[core/noncopyable][test/boost::unit_test::singleton] massive test failures
by Joaquin M Lopez Munoz 11 Sep '14

11 Sep '14

Hi, There's a problem with noncopyable and boost::unit_test::singleton in Intel 13 for Linux in C++11 compatibility mode that's rippling through many Boost libs. To list a few: date_time: http://tinyurl.com/lkeakab geometry: http://tinyurl.com/n3rajyn regex: http://tinyurl.com/kxkqo4c tuple: http://tinyurl.com/oz3vlct The error messages go like protected function "boost::noncopyable_::noncopyable::noncopyable()" [...] is not accessible through a "boost::noncopyable_::noncopyable" pointer or object[...] protected function "boost::unit_test::singleton<Derived>::singleton()" [...] is not accessible through a "boost::unit_test::singleton<boost::unit_test::unit_test_log_t>" pointer or object[...] which are manifestations of the following compiler bug: https://software.intel.com/en-us/forums/topic/404755 The report also indicates a simple workaround consisting of replacing BOOST_CONSTEXPR noncopyable() = default; with noncopyable() {} (seemingly, this would also fix boost::unit_test::singleton automatically.) Maybe the maintainer of boost:noncopyable can try a fix for ICC 13 along these lines? This can potentially clear a ton of yellow cells. I'd volunteer a pull request but alas I don't have local access to that compiler. Thank you, Joaquín M López Muñoz Telefónica

8 38

[build] clang objective c support?
by Emil Dotchevski 11 Sep '14

11 Sep '14

Hello, It seems like building with the clang-darwin toolset silently ignores .m and .mm files What is the status of Objective C support in this toolset? Is this a configuration problem on my part or is it just not supported? -- Emil Dotchevski Reverge Studios, Inc. http://www.revergestudios.com/reblog/index.php?n=ReCode

3 6

ICL space and running complexity
by Kenneth Adam Miller 11 Sep '14

11 Sep '14

With regards to the ICL data structure, can anybody tell me what the space and time complexity is of the various interval_map types (split and regular)? http://www.boost.org/doc/libs/1_56_0/libs/icl/doc/html/index.html

2 1

[git] Problem after deleting submodule
by Edward Diener 10 Sep '14

10 Sep '14

Git is starting to get to me. I did a git pull from the latest master only to get a bunch of conflicts from the latest align submodule. I have never touched the align submodule locally so having conflict errors with it is absurd. After trying to straighten out the errors via a 'git reset --hard' for the align submodule, and still getting conflicts, I deleted the submodule locally just so I could get the latest master. Now I can't seem to get back the submodule. No matter what I do the submodule is no longer recognized at the top level. What do I do ? Yes git is powerful but it is also a VCS from hell. About the only intelligent thing I have ever heard Linux Torvalds say.

1 1

Trac spam report
by Joaquin M Lopez Munoz 09 Sep '14

09 Sep '14

I got this one: https://svn.boost.org/trac/boost/ticket/10437 Joaquín M López Muñoz Telefónica

5 6

_MSVC_VER - variadic_slot_invoker.hpp
by Anders Dalvander 09 Sep '14

09 Sep '14

There appears to be two predefined macro spelling mistakes in signals2/detail/variadic_slot_invoker.hpp. It says _MSVC_VER where it should say _MSC_VER. #if defined(_MSVC_VER) // <-- Here # pragma warning(push) # pragma warning(disable:4100) // unreferenced formal parameter #endif ... #if defined(_MSVC_VER) // <-- ... and here # pragma warning(pop) #endif Regards, Anders Dalvander

1 0

[range] Interest in range_call function?
by Arthur Rosenstein 09 Sep '14

09 Sep '14

Hi, I'm checking to see if there's interest in a range_call function that makes calling iterator-based algorithms with ranges a little less painful. Simply put, range_call takes a function and an argument list and forwards the arguments to the function while expanding ranges into begin/end iterator pairs. For example: range_call(std::sort, vec, std::greater<int>()); It can also transform the result a la Boost.Range: range_call(std::copy, range_call<return_begin_found>(std::unique, range_call<return_begin_end>(std::sort, vec) ), std::ostream_iterator<int>(std::cout, " ") ); Obviously, Boost.Range provides a cleaner interface for most standard algorithms. range_call, however, works with arbitrary functions and has a smaller surface area. Is there any interest in this (possibly as part of Boost.Range)? -- Cheers, Arthur

1 0

Re: [boost] [process] [sorting] [singularity] [compute] [others] Who needs review manager?
by Antony Polukhin 09 Sep '14

09 Sep '14

2014-09-08 22:52 GMT+04:00 Kyle Lutz <kyle.r.lutz(a)gmail.com>: > On Mon, Sep 8, 2014 at 9:03 AM, Antony Polukhin <antoshkka(a)gmail.com> > wrote: > > Hi, > > > > Library authors that have no review manager and feel ready for review > > please contact me via this mailing list. > > Hi Antony, > > The Boost.Compute library [1] is in need of a review manager. Also, > would it be possible to add it to the review queue listed at [2]? > > Thanks, > Kyle > > [1] https://github.com/kylelutz/compute > [2] http://www.boost.org/community/review_schedule.html > I took a quick glance at the library and it looks really interesting and promising. I'm not an expert in a problem domain, but at least I have some experience with CUDA. I'll need a few weeks to look through the sources and ask questions. If no one objects, after that we could discuss a review date and put it to the schedule. -- Best regards, Antony Polukhin

1 0

How to respond to mailing list
by Kirk Liberty 08 Sep '14

08 Sep '14

Hello, I'm sorry to have to ask this, but I've googled all around and can't figure it out. My email is set up to receive periodic digests of the mailing list. If I want to respond to an email thread, how would I do that? Thanks, Kirk

4 3

NTCTS Iterator [was: Interest in is_iterator type trait?]
by Beman Dawes 08 Sep '14

08 Sep '14

On Tue, Aug 19, 2014 at 1:56 PM, Peter Dimov <lists(a)pdimov.com> wrote: > Beman Dawes wrote: > > Source can be a iterator to a NTCTS or a container. >> > > An iterator to an NTCTS? That's an interesting invention. :-) > See http://beman.github.io/ntcts_iterator/ or https://github.com/Beman/ntcts_iterator --Beman

9 18

[Boost-users] Testers required
by Antony Polukhin 08 Sep '14

08 Sep '14

Hi, There is a big need in testers that have access to exotic platforms(zArch, MIPS) and compilers (Cray, IBM xlc ...). Could you please run `b2` in `boost/libs/type_index/test` and report results. This will help to test Boost.TypeIndex library and other libraries that depend on `demangle.hpp` I'll appreciate any help. Thanks in advance! -- Best regards, Antony Polukhin

2 2

Help identify the conversion bug in MSVC
by Andrzej Krzemienski 08 Sep '14

08 Sep '14

Hi, I found a bug in MSVC 10.0 that causes a bug in Boost.Optional. I will be trying to find a workaround for the compiler. I would like to ask people with the access to MSVC 11.0 and 12.0 to check if the bug also persist in these newer releases. The following code is expected to compile: struct A { A(int) {} A(A &&) {} }; struct B { operator A() { return A(1); } operator int() { return 0; } }; int main() { A t = B(); } The following code is expected *NOT *to compile: struct A { A(int) {} }; struct B { operator int() { return 0; } }; int main() { A t = B(); } Please let me know what your compiler says. Thanks, &rzej

2 2

Re: [boost] Finite state machine
by arcolight 07 Sep '14

07 Sep '14

> Hi, > I don't know if you're using msm or statechart but msm uses no dynamic > memory, except if you choose to use the event queue or deferred events. And > even this can be avoided by providing as template argument you own queue. > If you don't require these features, you can disable them. > > Cheers, > Christophe I'm using MSM. Yes I tried this, but I could not do so. Perhaps, I doing something wrong. Problem in that i've no OS, and even no full C library, I'm using newlib-nano. I was trying to disable everything, that I can, but this didn't help. I understand, that this situation is special and boostmust count on something. May be you can advise me how to use (and, may be,build?)boost with arm-none-eabi compiler.I've read docs and trying to build boost for arm-none-eabi, but compilation fails. Do I understand correctly, that MSM doesn't need to compile something - this is only .hpp files and it doesn't use libs, so I can try to use it without compilation? -- arcolight

2 1

[git][align][litre] git submodule update failure.
by Mostafa 07 Sep '14

07 Sep '14

$ git submodule update libs/align fatal: reference is not a tree: d32353ea7d39c0237202e3e460f3248f226db274 Unable to checkout 'd32353ea7d39c0237202e3e460f3248f226db274' in submodule path 'libs/align' $ git submodule update tools/litre fatal: reference is not a tree: 564d4d8d30b7e03ac5e25d78e14d2c19fa321c83 Unable to checkout '564d4d8d30b7e03ac5e25d78e14d2c19fa321c83' in submodule path 'tools/litre' According to this: http://stackoverflow.com/questions/13425298/git-submodule-update-fatal-erro… , this is most likely a maintainer's bug. From the above: "This is the most common problem with submodules. The commit that you are on in the outer repository has a reference to a commit in the submodule that someone did not push up yet. It's a dependency problem. Always push from the inside-out. This is most likely not something you did wrong, but someone else that's working in the repository. They were sloppy and forgot to issue push on the submodule and just pushed the containing repository. Stuff works on their machine because they made the change and those commits exist there." Is this true?

3 3

Re: [boost] Finite state machine
by arcolight 07 Sep '14

07 Sep '14

> Are you using Statechart or MSM? > -- Michael Caisse ciere consulting ciere.com I tried to use both. -- arcolight

1 0

Re: [boost] Finite state machine (arcolight)
by arcolight 07 Sep '14

07 Sep '14

> This is something that would be very interesting to me, but I wonder if the > memory allocation can be factored out in the current boost library by using > allocators or something. I'm not terrible familiar with the libraries > internals, but I'd be willing to take a look at my next free cycle. To be > clear, is there another state machine lib in boost besides > boost.statechart? That's the only one I'm familiar with from a user > perspective. > > -Greg Yes, custom allocator doesn't solve the problem, especially because current FSM-s includes many other libraries. In any case, I can't afford to use even custom allocator in my project, if I would use it, I must have sufficient size heap, it's risky in embedded system. I already have third-party code, that uses heap, and I couldn't do anything with it. :) Another FSM implementation - Boost.MSM(Meta state machine), I tried to use it in different ways, but i've failed. MSM based on MPL boost library. If you interested in code, I have small repo on github: https://github.com/arcolight/ecl, there are several small classes, one of them, in fsm.hpp - state machine. There are simple test with cd-player like in boost lib.

1 0

Finite state machine
by arcolight 07 Sep '14

07 Sep '14

Hi all. I have a question about participation in boost libraries. In my embedded project I'm using finite state machine and I tried to use one of the boost implementations of FSM, but they are too big and complex. Main disadvantage of them is usage of dynamic memory allocation, that I can't afford on my CPU and many dependencies, that I can't use on my CPU. CPU is ARM Cortex-M4 with 128K memory, I've no operating system in my project. I have write my own FSM implementation without dynamic memory and dependencies, it's more simple, than boost libraries, but, of course, not so powerful. I've used C++11 standard, specifically variadic templates. My question - Is there some interest in such FSM implementation for boost developers? I'll be very grateful for advice. P.S.: I'm sorry for my english, it's not my native language.

4 3

Autolinking broken in master
by Agustín K-ballo Bergé 06 Sep '14

06 Sep '14

There's a report that autolinking is broken for a fresh recursive checkout of master, due to b2 generating 1.57 libraries while boost/version.hpp still reports the version as 1.56. This commit setting BOOST_VERSION in Jamroot made it to master: https://github.com/boostorg/boost/commit/d9cf98c2074cf8a27017170375c5c825e4… While the corresponding commit for boost/version.hpp didn't: https://github.com/boostorg/config/commit/34e1c7416be2ede7d86f3dcc093a84580… Regards, -- Agustín K-ballo Bergé.- http://talesofcpp.fusionfenix.com

1 0

[iterator] Merge to master
by Andrey Semashev 06 Sep '14

06 Sep '14

Hi, There were lots of changes to Boost.Iterator develop in the last few months. The tests look green for quite some time now. Could these changes be merged to master? I need those changes to merge next/prior in Boost.Utility.

2 1

sqlpp11, 3rd iteration
by Roland Bock 04 Sep '14

04 Sep '14

Dear Boosters, Quite a bit has happened since last I reported about sqlpp11 in this forum [1,2]. I have incorporated a lot of the feedback you gave me, hopefully bringing the library closer to a reviewable state. Source: https://github.com/rbock/sqlpp11 Doku: https://github.com/rbock/sqlpp11/wiki (not at all formal yet) I am hoping for more feedback both here and live at CppCon (http://sched.co/1r4lue3) Here are some noteworthy changes: _Restructuring:_ The code has undergone a major restructuring, leading to a much simpler and thereby much more flexible way to define the structure of SQL statements. As a result, it would now be quite simple to add vendor specific features sqlpp11 connector libraries. Dominique Devienne mentioned array binding for inserts, Johan Baltié mentioned hierarchical searches (both features of Oracle). If you are interested in writing a connector and adding such features, please let me know, I'll walk you through it. _NULL handling:_ Enabled by the restructured code and spurred by the library quince by Michael Shepanski, sqlpp11 can now calculate which result fields can or cannot be NULL. Speaking of which, handling NULL for result values has been discussed a lot. The library now has compile-time configurable behavior, you can choose between an std::optional-like interface and mapping NULL to the trivial value of the result type, e.g. 0 for numbers or "" for strings, see also https://github.com/rbock/sqlpp11/wiki/NULL _Connectors:_ Matthijs wrote a new postgresql connector from scratch, see https://github.com/matthijs/sqlpp11-connector-postgresql The sqlite3 connector received a bunch of corrections, thanks to dirkvdb _Compilers:_ Known to compile with clang>=3.1 and gcc>=4.8. I currently cannot test with MSVC or Intel or others, but if you happen to have the latest top notch version of any of those, I'd be interested in what they have to say about sqlpp11. Cheers, Roland [1]: http://lists.boost.org/Archives/boost/2013/11/208388.php [2]: http://lists.boost.org/Archives/boost/2014/02/211206.php

7 43

[move][unique_ptr] c++14 unique_ptr comes to town
by Ion Gaztañaga 03 Sep '14

03 Sep '14

Hi to all, We've discussed about a boostified version of std::unique_ptr several times in later years. I decided we could not continue with this hole in the Boost toolset and implemented in Boost.Move a unique_ptr that follows the C++14 interface. Commit here: SHA-1: e1da7c5ca1162841f643331af891a9a730eba118 http://github.com/boostorg/move/commit/e1da7c5ca1162841f643331af891a9a730eb… This implementation is also usable from C++03, and apart from the general limitations of the Boost.Move machinery, it should be really uesful for C++03 users. I would like to make this implementation the official boost::unique_ptr implementation, but until the community approves this, it's in boost::movelib namespace (boost::move it's a function so I needed to tweak the namespace name). I would like to thank Howard Hinnant for his great public implementation of the emulated unique_ptr published here: http://home.roadrunner.com/~hinnant/unique_ptr03.html I've ported and refactored Howard's extensive testsuite to implement boost test cases. Really useful to detect many implementation errors and details. I've found unique_ptr much harder to write than I expected. The standard text is really detailed and helps implementors, but things like allowed conversion between pointers and deleters and several ICEs and different SFINAE parties on many compilers made the work quite entertaining. The API is taken from the draft n3797 and the implementation has some differences with the standard. The main reason for this is that I wanted to make unique_ptr as lightweight as possible avoiding external dependencies, as it's interesting to implement idioms like PIMPL. Main attributes and differences with the standard: -> Implemented in two headers: <boost/move/unique_ptr.hpp> and <boost/move/make unique.hpp>. -> <boost/move/unique_ptr.hpp>: Boost.Move is the only heavyweight dependency. Not even that, as unique_ptr.hpp only uses boost/move/utility_core.hpp, that does not depend on any external library excepct BoostConfig and StaticAssert. No TypeTraits, MPL and others. Only Boost.Config, BOOST_STATIC_ASSERT and BOOST_ASSERT are used in the implementation. No standard library headers used (no pointer_traits, memory or other utilities). -> <boost/move/make unique.hpp>: Boost.PP is used in compilers without variadic templates. This is the main reason to have a separate make_unique.hpp header and avoid pulling Boost.PP for c++03 users that only want to have C++11 unique_ptr. -> unique_ptr's "operator <" is implemented as x.get() < y.get() instead of std::less<CT>(). std::less requires <funcional>. std::common_type requires C++11 standard library or some partial emulation based on Boost.TypeOf. I don't think 99,99% of users will notice the difference. For raw pointers, in all our platforms comparisons provide strict weak ordering. If a smart pointer writer specializes std::less for its smart pointer, it should be a law against that. -> It's assignable from literal 0. I found some problems in some compilers trying to write a constructor for that. -> nullptr support is conditional on !defined(BOOST_NO_CXX11_NULLPTR) but I think really should have a Boost nullptr for C++03 compilers. Let's put Meyer's nullptr (http://en.wikibooks.org/wiki/More_C%2B%2B_Idioms/nullptr) in Boost for BOOST_NO_CXX11_NULLPTR compilers. -> unique_ptr<T[]> is convertible to other types when conversion is considered safe. This means that unique_ptr<T[]> is convertible to unique_ptr<const volatile T[]>. Conversion from T[] to U[] is considered safe if remove_cv<T> is the same type as remove_cv<U> and unique_ptr<T[]>::pointer is convertible to unique_ptr<U[]>::pointer. This is a feature that might go to C++17 (http://www.open-std.org/jtc1/sc22/wg21/docs/lwg-active.html#2118) Safe conversions in reset() and constructors from pointers are also implemented in unique_ptr<T[]>. I tested the implementation in GCC{4.4, 4.6, 4.7, 4.8} Clang3.4 and MSVC{7.1, 8.0, 9.0, 10.0, 11.0, 12.0} in 32/64 bits, debug/release and C++03/C++11 modes (where applicable). The class documentation is generated with Boost.Move documentation and the most part of it is taken from the standard. I haven't added a tutorial to Boost.Move but I plan to it based on Howard's c++03 unique_ptr tutorial. I hope you find this addition to Boost.Move useful. If you think the implementation is good enough to be the official boost::unique_ptr implementation living in Boost.SmartPtr, porting should be easy, except for the documentation part. That would require either combining somehow handwritten and Quickbook produced documentation or writing new docs for unique_ptr. Best, Ion

6 30

Default documentation settings
by Glen Fernandes 03 Sep '14

03 Sep '14

What is the value of chunk.section.depth used, for those Boost libraries whose documentation is supplied only in QBK form (not the generated HTML from the QBK)? It appears to be 1, i.e. my sections-within-sections are not broken into separate pages. If so, could we make the default <xsl:param>chunk.section.depth=2? Glen

2 4

Can't file issues on trac with Chrome
by legalize+jeeves＠mail.xmission.com 03 Sep '14

03 Sep '14

[Please do not mail me a copy of your followup] Attempting to file a bug report gives me this message: "Captcha Error Submission rejected as potential spam (StopForumSpam says this is spam (username)) Trac thinks your submission might be Spam. To prove otherwise please provide a response to the following." ...but there is no captcha displayed in Chrome. If I attempt again in IE, then the captcha is displayed and I can enter the number and proceed. -- "The Direct3D Graphics Pipeline" free book <http://tinyurl.com/d3d-pipeline> The Computer Graphics Museum <http://computergraphicsmuseum.org> The Terminals Wiki <http://terminals.classiccmp.org> Legalize Adulthood! (my blog) <http://legalizeadulthood.wordpress.com>

1 0