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

Preprocessor: Expaning empty VA_ARGS
by Florian Lindner 09 Sep '16

09 Sep '16

Hello, I have this preprocessor macros: ----------- #include <boost/preprocessor/variadic/to_seq.hpp> #include <boost/preprocessor/seq/for_each_i.hpp> #include <boost/preprocessor/stringize.hpp> #include <iostream> #define TRACE(...) \ std::cout << "foo" \ BOOST_PP_SEQ_FOR_EACH_I(LOG_ARGUMENT,, BOOST_PP_VARIADIC_TO_SEQ(__VA_ARGS__)); #define LOG_ARGUMENT(r, data, i, elem) \ << std::endl << " Argument " << i << ": " << BOOST_PP_STRINGIZE(elem) << " == " << elem int main(int argc, char *argv[]) { TRACE(); return 0; } ----------- They compile when TRACE has arguments, but not without. Problem is, that LOG_ARGUMENT is called even when no args are supplied, leading to the expansion: std::cout << "foo" << std::endl << " Argument " << 0 << ": " << "" << " == " << ;; elem is empty, therefore there is no expression after the last << expected primary-expression before »;« token How can I deal with that? Thanks, Florian

3 4

[MultIndex] Iterator/Range validity when modify entries.
by Deniz Bahadir 09 Sep '16

09 Sep '16

Hi guys, While trying to implementing a new feature, I came across a question concerning iterator/range validity of boost::multi_index. To understand the question, here is my setup: <code> // Database entry. struct Entry { int key1; int key2; }; // Access-keys. struct Index1{}; struct Index2{}; using namespace boost::multi_index; using namespace boost; // The multi_index database. using Database = multi_index_container< Entry, indexed_by< hashed_non_unique< tag<Index1>, member<Entry, int, &Entry::key1> >, hashed_non_unique< tag<Index2>, member<Entry, int, &Entry::key2> > > >; Database db; /* ... Here db is getting filled. ... */ </code> I obtain a range of entries from that database with the same value for key1 through the first index and want to modify each single entry of that range. However, I will only modify key2 (which is only used as key in the other index)! My main question is now: Does modifying an entry invalidate my range although I do not modify the member that was used as key for obtaining the range? The code in question would possibly look like this: <code> auto& index1 = db.get<Index1>(); auto range = index1.equal_range(42); for (auto it = range.first; it != range.second; ++it) { index1.modify(it, [](Entry& entry) { entry.key2 = 123; // <---- Does this invalidate my range? // entry.key1 is not getting modified! }); } </code> My follow-up question is then: If the range is indeed invalidated, what would be a working solution for that problem? Incrementing the loop-iterator by hand before modifying the entry (and not letting the loop do it)? So, would the following be a valid solution? <code> auto& index1 = db.get<Index1>(); auto range = index1.equal_range(42); if (range.first != range.second) { auto it = range.first; auto end = --range.second; do { index1.modify(it++, [](Entry& entry) { entry.key2 = 123; // <---- Does this invalidate my range? // entry.key1 is not getting modified! }); while (it != end); } </code> Of course, neither do I want to iterate over an entry twice (and thereby modify it twice) nor do I want miss an entry. And one last question: Does this problem or its solution apply to all index-types or just to hashed indexes? Thanks for the clarification. Deniz PS: Oh and by the way: Boost.MultiIndex is a really cool library. :-)

3 4

[range] const-only UDT works as std algos, but not with boost range.
by John 08 Sep '16

08 Sep '16

A class that exposes a const_iterator, but does not expose a mutable iterator, works with std:: algorithms and range-for() loop, but does not work with boost range. // This works with std algorithms and range-for, but not with boost range. class C { public: typedef blah const_iterator; const_iterator begin() const; const_iterator end() const; }; const C c; for (auto x : c) { } // OK std::for_each(std::begin(c), std::end(c), fn); // OK boost::for_each (c, fn); // ERROR: fails SinglePassRange concept (GCC-6.2) The workaround is to add a non-const iterator type that is the same as the const iterator, and to add two additional non-const begin/end methods. class D { public: typedef blah const_iterator; const_iterator begin() const; const_iterator end() const; // kludge to make it work with boost. typedef const_iterator iterator; iterator begin(); iterator end(); }; const D d; boost::for_each (d, fn); // OK Is there a simpler solution?

2 3

[Boost.Asio] Suggestion to improve the docs
by dariomt＠gmail.com 07 Sep '16

07 Sep '16

Hi, In the Boost.Asio docs (and in the standalone Asio) the io_service docs has a section "Stopping the io_service from running out of work" where it explains "if the application requires that all operations and handlers be allowed to finish normally, the work object may be explicitly destroyed" and then shows some code using an std::auto_ptr<boost::asio::io_service::work> which will keep the work object on the heap. However, in the accepted answer to [1] the same behavior is accomplished with a, IMHO, more elegant solution: with a boost::optional<boost::asio::io_service::work>, which, if I'm correct, will keep the work object on the stack. To destroy the work object the optional is assigned to boost::none. I'd suggest improving the docs with the boost::optional case, which - does not use "ugly" (and deprecated) auto_ptr - keeps the boost::asio::io_service::work object on the stack Regards [1] http://stackoverflow.com/questions/15568100/confused-when-boostasioio-servi…

1 0

[boost][graph] Support for C++11 on adjacency list
by Marco Craveiro 06 Sep '16

06 Sep '16

Hi boost-users, Thanks very much for a great library - Boost in general and Boost Graph in particular . I'm trying to understand the status of a fix on Boost Graph - any pointers are greatly appreciated. My basic problem is that trying to use Boost Graph in C++ 11 (or later) mode seems to break due to a bug in adjacency list. This bug has been reported in Track [1], [2] and elsewhere [3]. Up to now I have been carrying local patches to fix this issue as per tickets [4], but to the untrained eye, it appears this issue has been fixed upstream a while ago [5]. I've read the discussion regarding Boost Graph merges [6] but - apologies - It's still not clear to me where we are. Is it possible to confirm: - that the fix is in develop and will be shipped? - if so, would it be shipped with Boost 1.62 or with a later release? I read the release notes [7] but couldn't see any references to Boost Graph. Many thanks for your time. -- Marco Craveiro MD, Domain Driven Consulting about: http://about.me/marcocraveiro blog: http://mcraveiro.blogspot.co.uk twitter: https://twitter.com/MarcoCraveiro That the Ideas are themselves manifestations (of the Idea-Idea) and that the Idea-Idea is a-kind-of Manifestation-Idea—which is a-kind-of itself, so that the system is completely self-describing— would have been appreciated by Plato as an extremely practical joke [Plato]. -- Alan Key [1] #11164 Graph adjacency_list, add_vertex compile error on clang 3.6 https://svn.boost.org/trac/boost/ticket/11164 [2] #10382 1.56.0 Graph adjacency_list has compile errors on g++ 4.6.4 https://svn.boost.org/trac/boost/ticket/10382 [3] https://lists.debian.org/debian-mentors/2016/01/msg00327.html [4] https://github.com/DomainDrivenConsulting/dogen/blob/master/patches/boost_1… [5] Fix compilation with gcc 4.4 in C++11 mode https://github.com/boostorg/graph/commit/0f4ecbcd4bd13743f74745257b95fabdbd… [6] http://lists.boost.org/Archives/boost/2016/08/230533.php [7] http://www.boost.org/users/history/version_1_62_0.html

2 2

Fwd: Boost.Range: 'Use of deleted function' GCC error when using lambdas
by Remko Tronçon 05 Sep '16

05 Sep '16

Hi, I have the following piece of code: std::vector<std::string> list1; std::vector<std::string> list2; push_back(list2, list1 | transformed([](const auto& w) { return w; }) | sliced(0, 1)); This compiles fine with CLang, but with GCC I get the error: /usr/include/boost/optional/optional.hpp:546:64: error: use of deleted function 'main(int, const char**)::<lambda(const auto:1&)>& main(int, const char**)::<lambda(const auto:1&)>::operator=(const main(int, const char**)::<lambda(const auto:1&)>&)' void assign_value ( rval_reference_type val ) { get_impl() = static_cast<rval_reference_type>(val); } ~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ main.cpp:46:19: note: a lambda closure type has a deleted copy assignment operator | transformed([](const auto& w) { return w; }) When I replace the lambda by a regular function, the code compiles; when I don't use 'sliced' (but pipe it to e.g. 'transformed' instead), the code also compiles. Is there a way to make this code compile with a lambda? thanks! Remko PS: I'm using GCC 6.2.0 with Boost 1.61.

2 1

using asio.hpp - boost::system::generic_category()
by Hermann Norpois 04 Sep '16

04 Sep '16

Hello, I just started with boost and it is obvious that I do not understand give my compiler (not mentioning code::blocks) the correct instructions. I installed boost_1_61_0 on Ubuntu 14.04 and followed the instruction. So I did (as root after sudo -i) in the directory of boost_1_61_0. I use g++4.9. It was the same with g++4.8. 1) ./bootstrap.sh prefix=/usr/local 2) ./b2 install So the libs are: /usr/local/lib The headers: /usr/include/boost Actually I was interested in some stuff with asio. So I started with this code from the tutorial. http://www.boost.org/doc/libs/1_61_0/doc/html/boost_asio/tutorial/tutdaytim… g++ -Wall main.cpp -o zclient /tmp/ccxx7PrW.o: In Funktion `__static_initialization_and_destruction_0(int, int)': main.cpp:(.text+0x52e): Nicht definierter Verweis auf `boost::system::generic_category()' main.cpp:(.text+0x53a): Nicht definierter Verweis auf `boost::system::generic_category()' main.cpp:(.text+0x546): Nicht definierter Verweis auf `boost::system::system_category()' /tmp/ccxx7PrW.o: In Funktion `boost::system::error_code::error_code()': main.cpp:(.text._ZN5boost6system10error_codeC2Ev[_ZN5boost6system10error_codeC5Ev]+0x17): Nicht definierter Verweis auf `boost::system::system_category()' /tmp/ccxx7PrW.o: In Funktion `boost::asio::error::get_system_category()': main.cpp:(.text._ZN5boost4asio5error19get_system_categoryEv[_ZN5boost4asio5error19get_system_categoryEv]+0x5): Nicht definierter Verweis auf `boost::system::system_category()' /tmp/ccxx7PrW.o: In Funktion `boost::asio::detail::posix_thread::~posix_thread()': main.cpp:(.text._ZN5boost4asio6detail12posix_threadD2Ev[_ZN5boost4asio6detail12posix_threadD5Ev]+0x26): Nicht definierter Verweis auf `pthread_detach' /tmp/ccxx7PrW.o: In Funktion `boost::asio::detail::posix_thread::join()': main.cpp:(.text._ZN5boost4asio6detail12posix_thread4joinEv[_ZN5boost4asio6detail12posix_thread4joinEv]+0x2b): Nicht definierter Verweis auf `pthread_join' /tmp/ccxx7PrW.o: In Funktion `boost::asio::detail::posix_thread::start_thread(boost::asio::detail::posix_thread::func_base*)': main.cpp:(.text._ZN5boost4asio6detail12posix_thread12start_threadEPNS2_9func_baseE[_ZN5boost4asio6detail12posix_thread12start_threadEPNS2_9func_baseE]+0x29): Nicht definierter Verweis auf `pthread_create' collect2: error: ld returned 1 exit status I have to add that the simple programm in the starting guide worked properly http://www.boost.org/doc/libs/1_61_0/more/getting_started/unix-variants.htm… via g++ -Wall main.cpp -o zu It worked in Code::Blocks as well (without any additional information). Even the second simple programm (same link) I managed to compile giving the full path: g++ -Wall main.cpp -o extractmail /usr/local/lib/libboost_regex.a At no time I had not to instruct the compiler about the headers. I was wondering. So it must have anything to do with the libs. Coming back to my asio- example. Do I need libs for it? I guess is has something to do with the expression -lboost_system. I used it without understanding it. But it did not help. Actually I would appreciate if was able to use Code::Blocks but I did not manage to give Code::Blocks the correct instructions. I found some hints, at least amounts of hints ... But it did not work at all. If somebody is able to explain it (for idiots) ... This would be great. Thanks Hermann

2 1

passing variables into thread
by Alexander Carôt 03 Sep '16

03 Sep '16

Hello all, I have a thrad function like: void sj::eventTrigger(){ boost::mutex::scoped_lock lock(eventTriggerMutex); while (eventTriggerActive){ dFC->eventTriggerLock.wait(lock); /// do something with pointer-variable } } This function is executed by a eventTriggerLock.notify_all() – call from other objects and when I need to pass certain variables into the thread I have been unsing a global pointer variable. Now my question: I wonder if there is a way to get rid of the pointer variable and directly pass variables into the thread such as: eventTriggerLock.notify_all(data1) void sj::eventTrigger(int data1){ boost::mutex::scoped_lock lock(eventTriggerMutex); while (eventTriggerActive){ dFC->eventTriggerLock.wait(lock); /// do something with data1 } } Any hint welcome – thanks in advance Alex

3 2

Priority deque?
by Paolo Bolzoni 01 Sep '16

01 Sep '16

Dear list, I was wondering if there some kind of "priority queue", that is a data structure akin the priority queue, but where you can top and pop from both sides. My use case is that I need to quickly access the maximum, but also not push new elements if they are smaller than the minimum. I am aware I can use a std::set, but in the current implementation where I always push new elements even if they are bad, using a std::set the performance is significantly worse. Yours faithfully, Paolo

3 5

iostreams: ceased devolopment?
by Christian Meesters 01 Sep '16

01 Sep '16

Hi, I noticed that there are numerous issues in the bug list, regarding boost iostreams. However, the current doc states Copyright 2004-2007 and Tickets do not receive responses. (I even read: "(I tried e-mailing Jonathan Turkanis directly, but the e-mail address on the website seems to be not used anymore)" in ticket 7574. So, my question is: I anyone maintaining this library? If not: I doubt that there is a good alternative, because iostreams just is convenient to use and a selling point to for boost. It would be a pitty. Best regards, Christian

7 15