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

Re: [Boost-users] in IOS, thread_info object is being destructed before the thread finishes executing (Andy Weinstein)
by Andy Weinstein 06 Feb '13

06 Feb '13

I have solved this problem. Please see my post in stack overflow at: http://stackoverflow.com/questions/14692802/boost-threads-in-ios-thread-inf… The short answer is that the IOS boost.sh script needs to be modified to use BOOST_SP_USE_SPINLOCK instead of BOOST_SP_USE_PTHREADS. But there are a lot of details involved in getting this to work in IOS which are described in that post. Maybe also a small modification to the boost code in spinlock_gcc_arm.hpp, but I need to verify that a change I made along the way is absolutely necessary.

1 0

max value by the number of bits
by Christian Henning 06 Feb '13

06 Feb '13

Hi there, is there a general template for calculating the max value by the number of bits. GIL right now uses the following template: template <int K> struct unsigned_integral_max_value<packed_channel_value<K> > : public mpl::integral_c<typename packed_channel_value<K>::integer_t, (1<<K)-1> {}; This code gives me some compiler errors with clang 3.2. for K = 31, for instance. Ideally I would like to take K up to 64. A small fix gets me up to K = 63 template <int K> struct unsigned_integral_max_value<packed_channel_value<K> > : public mpl::integral_c<typename packed_channel_value<K>::integer_t, (uint64_t(1)<<K)-1> {}; Is there a better way in boost? Thanks, Christian

6 14

[Auto-link] Visual Studio doesn't search for the correct libraries
by Angelo Moscati 06 Feb '13

06 Feb '13

Hi, I recently moved from Boost version 1.49 to 1.52. I want to use the static library versions but the autolink keep to requeire the dynamic one. This is the command line i use to build boost on Windows: set REQUESTED_BOOST_LIBRARIES=--with-serialization --with-thread --with-date_time --with-system --with-regex --with-filesystem --with-log set ADDITIONAL_BOOST_FLAGS=define=BOOST_LOG_NO_COMPILER_TLS rem x86 .\b2 -j 8 link=static runtime-link=static runtime-debugging=on variant=debug threading=multi --abbreviate-paths architecture=x86 --stagedir=stage\x86\debug %ADDITIONAL_BOOST_FLAGS% %REQUESTED_BOOST_LIBRARIES% When i try to build my application, i get the follow errors (it builds fine with Boost 1.49): error LNK1104: cannot open file 'libboost_filesystem-vc100-mt-gd-1_52.lib'. But the library i built is libboost_filesystem-vc100-mt-sgd-1_52.lib. Can you tell me what i do wrong? Thanks, Angelo

1 2

[string.algo] find_iterator, gives compile error in Boost 1.53
by gast128 06 Feb '13

06 Feb '13

Hello all, just dl Boost 1.53 and now I get a compile error in 'find_iterator' when using a forward iterator. Probably related to the change 'increment'. It now uses begin() + 1, and the + operator is not defined for the supplied iterator.

2 3

Archive search
by Angelo Moscati 06 Feb '13

06 Feb '13

Hi, I would like to know if there is a way to search the archive before asking questions. Thanks, Angelo

2 2

[program_options] Is it possible to change semantics after add_options() ?
by lukasz kucharski 06 Feb '13

06 Feb '13

Hello, I would like to ask if it is possible to change value_semantic associated with a program_option after it has been created ? I create `program_options` instance and populate it with add_options. Like in the examples provided by the documentation. I would like to change the semantics of a particular option later on. Is this even possible, or is this functionality deliberately disabled. I figured out that one can use options_description::find(), then strip it out `const`ness and dynamic cast to `typed_value_base` and use its methods. My motivation is such a design. I have a set of applications, which have a common subset of command line options/switches. However, they differ a bit in semantics. I.e. default value. I decided to refactor common parts into some `BasicApp` class and I thought I could move common options along with it as `boost::program_options` OO design should naturally support it. And it worked like a charm with `add_options` semantics to add new options. Up to the point I needed to change existing default values. Is there a proper way to do it ? Removing `const` constraint seems a bad approach since it looks like it's left there on purpose. Or my use-case is ill or hopeless. With regards, Łukasz.

1 0

enable_shared_from_this()
by Dominique Devienne 06 Feb '13

06 Feb '13

Can someone please explain why this code: struct Foo : boost::enable_shared_from_this<Foo> {}; Foo* p_foo = new Foo; boost::shared_ptr<Foo> ptr_foo(p_foo->shared_from_this()); is throwing? msvcr100d.dll!_CxxThrowException(void * pExceptionObject, const _s__ThrowInfo * pThrowInfo) Line 157 C++ boost::throw_exception<boost::bad_weak_ptr>(const boost::bad_weak_ptr & e) Line 61 + 0x4d bytes C++ boost::detail::shared_count::shared_count(const boost::detail::weak_count & r) Line 421 + 0x23 bytes C++ boost::shared_ptr<`anonymous namespace'::Foo>::shared_ptr<`anonymous namespace'::Foo><`anonymous namespace'::Foo>(const boost::weak_ptr<`anonymous namespace'::Foo> & r) Line 216 C++ boost::enable_shared_from_this<`anonymous namespace'::Foo>::shared_from_this() Line 49 + 0xf bytes C++ > `anonymous namespace'::test_method() Line 41 + 0xf bytes C++ Since Foo embeds a weak_ptr, why can't it new the pi_ on the spot? (I'm using Boost 1.44). Thanks, --DD PS: On a related note, why doesn't boost::shared_ptr<Foo>(p_foo) (where p_foo is a raw Foo*) implicitly apply the aliasing Ctor, to share the same px, to avoid the classic bug of having several different pn on the same px, since p_foo derives from enable_shared_from_this()? Initially i thought that's what sp_enable_shared_from_this() and pe->_internal_accept_owner() were meant to do, but obviously that's not the case.

4 4

Boost statechart design question
by sidharth shanbhag 05 Feb '13

05 Feb '13

Hi, I am designing an application using Boost statechart and needed some advice. I have multiple objects of the same type/functionality, however they are not related by an inheritance heirachy. The class types have a defined initialization sequence which move from 1 state to another. A few of these state initialization steps are the same eg. Resource X should be initialized as the second step in the sequence. But X can be initialized only once in the process. I am looking at doing the following: - Creating a statechart for every object, to manage its initialization. - Each state machine will maintain the pointer of the class object it has initialized. - A map of object Name -> State Machine will be created to ensure every object is initialized correctly. - Represent resource X as a separate state, which after its initialization will transition to the next state. Is there a potentially better way to do this considering: - There can be multiple objects of different class types - New objects can be added during runtime. Thanks, Sid

2 1

How can I free memory allocated when deserializing with boost::archive::text_iarchive
by Adlai Shawareb 05 Feb '13

05 Feb '13

Thanks for your reply. Here is a code sample: std::istringstream istr; <* put data into istr here *> boost::archive::text_iarchive ia(istr_); Message* p_temp; ia & p_temp; mpBaseIncomingMessage = p_temp; // mpBaseIncomingMessage is of type (Message *) mpIncomingMessage = static_cast<SystemMessage*>(mpBaseIncomingMessage); // mpIncomingMessage is a pointer to a class derived from Message mObjectCallbackMap.process(mpIncomingMessage->mEvent); // execute a callback for this event delete mpIncomingMessage; // delete the allocated memory We then see a crash on either the " mObjectCallbackMap.process..." line or the " delete mpIncomingMessage", but only when certain actions are taken by the system. Any thoughts would be greatly appreciated. Regards, Adlai -----Original Message----- From: Boost-users [mailto:boost-users-bounces@lists.boost.org] On Behalf Of boost-users-request(a)lists.boost.org Sent: Monday, February 04, 2013 1:11 PM To: boost-users(a)lists.boost.org Subject: Boost-users Digest, Vol 3349, Issue 6 Send Boost-users mailing list submissions to boost-users(a)lists.boost.org To subscribe or unsubscribe via the World Wide Web, visit http://lists.boost.org/mailman/listinfo.cgi/boost-users or, via email, send a message with subject or body 'help' to boost-users-request(a)lists.boost.org You can reach the person managing the list at boost-users-owner(a)lists.boost.org When replying, please edit your Subject line so it is more specific than "Re: Contents of Boost-users digest..." Today's Topics: 1. Re: Crediting the Work of Others (Bill Buklis) 2. How can I free memory allocated when deserializing with boost::archive::text_iarchive (Adlai Shawareb) 3. Re: in IOS, thread_info object is being destructed before the thread finishes executing (Nat Linden) 4. Re: Undirected version of dijkstra shortest path on directed graph. (Jeremiah Willcock) 5. Re: Crediting the Work of Others (Algocoders) 6. Re: How can I free memory allocated when deserializing with boost::archive::text_iarchive (Tim Moore) 7. [mpl::integral_c] Compiler Error with Visual Studio 2010 (Christian Henning) 8. Re: [mpl::integral_c] Compiler Error with Visual Studio 2010 (Nathan Crookston) ---------------------------------------------------------------------- Message: 1 Date: Mon, 04 Feb 2013 13:18:47 -0600 From: Bill Buklis <boostusr(a)pbjzone.com> To: boost-users(a)lists.boost.org Subject: Re: [Boost-users] Crediting the Work of Others Message-ID: <51100997.1050909(a)pbjzone.com> Content-Type: text/plain; charset=ISO-8859-1; format=flowed On 2/1/2013 1:33 PM, Algocoders wrote: > Dave Abrahams <dave <at> boostpro.com> writes: > >> >> It has come to my attention that this book appears to use the work of >> others without attribution. I'd like to give the authors a chance to >> explain why that appearance is false. From what you can see in >> Amazon's limited preview of the book, one can find numerous instances >> of code lifted from Andrew Sutton's "origin" library. >> Some of the data structures have been renamed or had the case changed. >> Most of the duplication of Origin code comes in Chapter 7 in that >> book. There appear to be no references to Origin in that chapter >> (based on searching through the book in Amazon). >> >> The IndirectlyCopyable text includes unsourced quotes directly from >> http://www.open-std.org/jtc1/sc22/wg21/docs/papers/2012/n3351.pdf >> without quotation, and the concepts* diagrams are taken from the same >> document. Is there an attribution on the previous page that we can't >> see from Amazon's limited preview? > Crediting the work of others have been done in the book as mentioned > in the links below: > Preface : > http://www.algocoders.com/sites/default/files/preface_foa_revised.pdf > Acknowledgement : > http://www/algocoders.com/sites/default/files/acknowledgement_foa.pdf > Table of Contents : > http://www.algocoders.com/sites/default/files/toc_foa_revised.pdf > > Chapter 3 : > http://www.algocoders.com/sites/default/files/chapter3_foa.pdf > Chapter 11 : > http://www.algocoders.com/sites/default/files/chapter11_foa.pdf > > We have avoided repetitive mention of original sources in respective > chapters to let not the content be littered with these details again. > Please let us know if we need to attribute in respective chapters too, > we would like to work on it. > Your feedback will be really helpful to guide us as a corrective measure. > > Looking forward for suggestions, > > Cheers, > Algocoders > Only the last link for Chapter 11 actually works and there aren't any acknowledgements presented in that pdf. -- Bill ------------------------------ Message: 2 Date: Mon, 4 Feb 2013 19:46:54 +0000 From: Adlai Shawareb <adlai(a)vytronus.com> To: "boost-users(a)lists.boost.org" <boost-users(a)lists.boost.org> Subject: [Boost-users] How can I free memory allocated when deserializing with boost::archive::text_iarchive Message-ID: <842DD0CD91745A48B9355924E2CF0BEA114FEF3D(a)VYTRONUS-SBS.vytronus.local> Content-Type: text/plain; charset="us-ascii" Hi, Does anyone know how we can free memory that is dynamically allocated by boost::archive::text_iarchive when deserializing an archive? We have tried using 'delete' and free() on the pointer when we finish with it, but it results in a crash. How does Boost dynamically allocate memory, 'new', 'malloc', or other? We tried a shared_ptr, but were getting compile errors related to the include files, and have not gone back to track that down. So if there is a simple way to free the memory that would be best for our application. Thanks, Adlai Shawareb Vytronus, Inc.

2 1

Boost 1.53.0 released!
by Marshall Clow 04 Feb '13

04 Feb '13

Release 1.53.0 of the Boost C++ Libraries is now available. These open-source libraries work well with the C++ Standard Library, and are usable across a broad spectrum of applications. The Boost license encourages both commercial and non-commercial use. This release contains five new libraries and numerous enhancements and bug fixes for existing libraries. New Libraries: * Atomic: C++11-style atomic<>, from Helge Bahmann, maintained by Tim Blechmann. * Coroutine: Coroutine library, from Oliver Kowalke. * Lockfree: Lockfree data structures, from Tim Blechmann. * Multiprecision: Extended precision arithmetic types for floating point, integer and rational arithmetic from John Maddock and Christopher Kormanyos. * Odeint: Solving ordinary differential equations, from Karsten Ahnert and Mario Mulansky. For details, including download links, see http://www.boost.org/users/news/version_1.53.0 You can also download directly from SourceForge: http://sourceforge.net/projects/boost/files/boost/1.53.0/ To install this release on your system, see http://www.boost.org/doc/libs/release/more/getting_started/index.html Thanks, --The Boost release team Beman Dawes Daniel James Eric Niebler Marshall Clow Rene Rivera Vladimir Prus

2 2

[mpl::integral_c] Compiler Error with Visual Studio 2010
by Christian Henning 04 Feb '13

04 Feb '13

Hi there, shouldn't the following work: #include <boost\gil\gil_all.hpp> using namespace std; using namespace boost; int main() { typedef mpl::integral_c< unsigned char , numeric_limits< unsigned char >::max() > my_max_value_t; return 0; } Visual Studio 2010 gives me: error C2975: 'N' : invalid template argument for 'boost::mpl::integral_c', expected compile-time constant expression I'm a little surprised that this isn't working. Thanks, Christian

2 2

How can I free memory allocated when deserializing with boost::archive::text_iarchive
by Adlai Shawareb 04 Feb '13

04 Feb '13

Hi, Does anyone know how we can free memory that is dynamically allocated by boost::archive::text_iarchive when deserializing an archive? We have tried using 'delete' and free() on the pointer when we finish with it, but it results in a crash. How does Boost dynamically allocate memory, 'new', 'malloc', or other? We tried a shared_ptr, but were getting compile errors related to the include files, and have not gone back to track that down. So if there is a simple way to free the memory that would be best for our application. Thanks, Adlai Shawareb Vytronus, Inc.

2 1

A cool book for c++11 is Published now (Revised Edition)
by Algocoders . 04 Feb '13

04 Feb '13

Hi All, Revised edition of the book : "Foundation of Algorithms in C++11, Volume 1(Revised Edition): Using and Extending C++11, Boost And Beyond" : is available now. @Amazon: http://www.amazon.com/Foundation-Algorithms-Volume-Revised-Edition/dp/14819… @Pothi.com (Indian Edition): http://pothi.com/pothi/book/chandra-shekhar-kumar-foundation-algorithms-c11… This book is vital to understand and extend the C++11 Algorithms by carefully worked out synthesis of language and library features with an eye at future evolution with special emphasis to : - Template Alias - constexpr - copy_backward - requires - std::enable_if : SFINAE - Private Cast - Type Functions - Type Traits - Explicit Template Instantiations and Specializations - Trailing Return Type - auto type specifier - Intermediate Traits Idiom - Value Type Deduction Framework Table of Contents : http://algocoders.com/sites/default/files/toc_foa_revised.pdf Sample Chapter : http://algocoders.com/sites/default/files/chapter1_foa_revised.pdf For more details, please visit the link : http://www.algocoders.com/article/books Have Fun ! Cheers, Algocoders http://www.algocoders.com

4 5

Undirected version of dijkstra shortest path on directed graph.
by Jupp Tscheak 04 Feb '13

04 Feb '13

Hello BGL specialists. Is it possible to use an undirected version of dijkstra shortest path function on an /adjacency_list/ which /Directed/ template parameter is set to bidirectional? Thanks in advance :)

2 1

in IOS, thread_info object is being destructed before the thread finishes executing
by Andy Weinstein 04 Feb '13

04 Feb '13

Our project uses a few boost 1.48 libraries on several platforms, including Windows, Mac, Android, and IOS. We are able to consistently get the IOS version of the project to crash (nontrivially but reliably) when using IOS, and from our investigation we see that ~thread_data_base is being called on the thread's thread_info while its thread is still running. This seems to happen as a result of the smart pointer reaching a zero count, even though it is obviously still in scope in the thread_proxy function which creates it and runs the requested function in the thread. This seems to happen in various cases - the call stack is not identical between crashes, though there are a few variations which are common. Just to be clear - this often requires running code which is creating hundreds of threads, though there are never more than about 30 running simultaneously. I have "been lucky" and got it very very early in the run also, but that's rare. I created a version of the destructor which actually catches the code red-handed: in libs/thread/src/pthread/thread.cpp: thread_data_base::~thread_data_base() { boost::detail::thread_data_base* const thread_info=detail::get_current_thread_data(); void *void_thread_info = (void *) thread_info; void *void_this = (void *) this; // is somebody destructing the thread_data other than its own thread? // (remember that its own which should no longer point to it anyway, // because of the call to detail::set_current_thread_data(0) in thread_proxy) if (void_thread_info) { // == void_this) { __builtin_trap(); } } I should note that (as seen from the commented-out code) I had previously checked to see that void_thread_info == void_this because I was only checking for the case where the thread's current thread_info was killing itself. I have also seen cases where the value returned by get_current_thread_data is non-zero and different from "this", which is really weird. Also when I first wrote that version of the code, I wrote: if (((void*)thread_info) == ((void*)this)) and at run-time I got some very weird exception that said I something about a virtual function table or something like that - I don't remember. I decided that it was trying to call "==" for this object type and was unhappy with that, so I rewrote as above, putting the conversions to void * as separate lines of code. That in itself is quite suspicious to me. I am not one to run to rush to blame compilers, but... I should also note that when we did catch this happening the trap, we saw the destructor for ~shared_count appear twice consecutively on the stack in Xcode source. Very doubleweird. We tried to look at the disassembly, but couldn't make much out of it. Again - it looks like this is always a result of the shared_count which seems to be owned by the shared_ptr which owns the thread_info reaching zero too early. Help! Thanks, Andy Weinstein PS I cannot say for sure yet whether this problem occurs on the other platforms, though it looks like Windows is more solid that Mac. Android pretty solid also, though maybe in between Windows and Mac.

2 1

Question regarding circular buffer usage
by Salivar.William 04 Feb '13

04 Feb '13

I am using the circular buffer container to hold up to 10000 std string objects. I am finding that once the buffer has reached max capacity, memory continues to be consumed as I push new strings in. Is there something I'm missing? I stepped into the boost code and it does make a copy of the string using the std string operator so I would think it handles the string copy properly. The size of the strings are variable but I would expect the memory usage to be fairly stable when the buffer is full but that is not the case. Here's basic usage Definition: boost::circular_buffer<std::string> m_message_buffer; Initialization: m_message_buffer.set_capacity (10000); Usage: void myClass::writeToBuffer (const char *message) { m_message_buffer.push_back (string (message)); } Boost code: void push_back(param_value_type item = value_type()) { if (full()) { if (empty()) return; replace(m_last, item); increment(m_last); m_first = m_last; } else { m_alloc.construct(m_last, item); increment(m_last); ++m_size; } } void replace(pointer pos, param_value_type item) { *pos = item; #if BOOST_CB_ENABLE_DEBUG invalidate_iterators(iterator(this, pos)); #endif }

2 2

How to uninstall/reinstall boost by using ./b2 command?
by JIA Pei 04 Feb '13

04 Feb '13

Hi, all: I installed boost by using *./bootstrap.sh* *./b2 install --prefix=/usr* However, I missed two files,* patchlevel.h* and *bzlib.h*: So, afterwards, I have the following two packages installed: *libbz2-dev*and * libpython3.3-dev* . I would like to have a full boost_1_52_0 re-installed. But now, even if I have the files: *libbz2-dev: /usr/include/bzlib.h* *libpython3.3-dev: /usr/include/python3.3m/patchlevel.h* during the re-installation of boost, it still keeps reporting 2 errors: *fatal error: bzlib.h: No such file or directory* *fatal error: patchlevel.h: No such file or directory* So, any suggestions? Cheeers -- Pei JIA Email: jp4work(a)gmail.com cell: +1 604-362-5816 Welcome to Vision Open http://www.visionopen.com

1 0

Re: [Boost-users] in IOS, thread_info object is being destructed before the thread finishes executing (Andy Weinstein)
by Andy Weinstein 04 Feb '13

04 Feb '13

I should note that the boost.sh from Pete Goodliffe (and modified by others) that is commonly used to compile boost for IOS has the following note in the header: : ${EXTRA_CPPFLAGS:="-DBOOST_AC_USE_PTHREADS -DBOOST_SP_USE_PTHREADS"} # The EXTRA_CPPFLAGS definition works around a thread race issue in # shared_ptr. I encountered this historically and have not verified that # the fix is no longer required. Without using the posix thread primitives # an invalid compare-and-swap ARM instruction (non-thread-safe) was used for the # shared_ptr use count causing nasty and subtle bugs. # # Should perhaps also consider/use instead: -BOOST_SP_USE_PTHREADS I use those flags, but to no avail. I also found that there is an alternate implementation inside boost for arm processors which seems also to directly address this issue: spinlock_gcc_arm.hpp The version included with boost 1.48 uses outdated arm assembly. I took the updated version from boost 1.52, but I'm having trouble compiling it. I get the following error: predicated instructions must be in IT block I found a reference to what looks to be a similar use of this instruction here: https://zeromq.jira.com/browse/LIBZMQ-414 I was able to use the same idea to get the 1.52 code to compile by modifying the code as follows (I inserted an appropriate IT instruction) __asm__ __volatile__( "ldrex %0, [%2]; \n" "cmp %0, %1; \n" "it ne; \n" "strexne %0, %1, [%2]; \n" BOOST_SP_ARM_BARRIER : "=&r"( r ): // outputs "r"( 1 ), "r"( &v_ ): // inputs "memory", "cc" ); But in any case, there are ifdefs in this file which look for the arm architecture, which is not defined that way in my environment. After I simply edited the file so that only ARM 7 code was left, the compiler complains about the definition of BOOST_SP_ARM_BARRIER: In file included from ./boost/smart_ptr/detail/spinlock.hpp:35: ./boost/smart_ptr/detail/spinlock_gcc_arm.hpp:39:13: error: instruction requires a CPU feature not currently enabled BOOST_SP_ARM_BARRIER : ^ ./boost/smart_ptr/detail/spinlock_gcc_arm.hpp:13:32: note: expanded from macro 'BOOST_SP_ARM_BARRIER' # define BOOST_SP_ARM_BARRIER "dmb" So that's where thing stand. I am hopefully awaiting your help…. Thanks! On Feb 4, 2013, at 3:38 PM, <boost-users-request(a)lists.boost.org<mailto:boost-users-request@lists.boost.org>> wrote: Send Boost-users mailing list submissions to boost-users(a)lists.boost.org<mailto:boost-users@lists.boost.org> To subscribe or unsubscribe via the World Wide Web, visit http://lists.boost.org/mailman/listinfo.cgi/boost-users or, via email, send a message with subject or body 'help' to boost-users-request(a)lists.boost.org You can reach the person managing the list at boost-users-owner(a)lists.boost.org When replying, please edit your Subject line so it is more specific than "Re: Contents of Boost-users digest..." Today's Topics: 1. in IOS, thread_info object is being destructed before the thread finishes executing (Andy Weinstein) ---------------------------------------------------------------------- Message: 1 Date: Mon, 4 Feb 2013 12:38:29 +0000 From: Andy Weinstein <andyw(a)hyperlync.com> To: "Boost-users(a)lists.boost.org" <Boost-users(a)lists.boost.org> Cc: Moshe Rubin <moshe(a)hyperlync.com>, Ronnie Wulfsohn <roni(a)hyperlync.com> Subject: [Boost-users] in IOS, thread_info object is being destructed before the thread finishes executing Message-ID: <5D324B34-BE82-4B4C-89AD-ED8A0095129B(a)hyperlync.com> Content-Type: text/plain; charset="us-ascii" Our project uses a few boost 1.48 libraries on several platforms, including Windows, Mac, Android, and IOS. We are able to consistently get the IOS version of the project to crash (nontrivially but reliably) when using IOS, and from our investigation we see that ~thread_data_base is being called on the thread's thread_info while its thread is still running. This seems to happen as a result of the smart pointer reaching a zero count, even though it is obviously still in scope in the thread_proxy function which creates it and runs the requested function in the thread. This seems to happen in various cases - the call stack is not identical between crashes, though there are a few variations which are common. Just to be clear - this often requires running code which is creating hundreds of threads, though there are never more than about 30 running simultaneously. I have "been lucky" and got it very very early in the run also, but that's rare. I created a version of the destructor which actually catches the code red-handed: in libs/thread/src/pthread/thread.cpp: thread_data_base::~thread_data_base() { boost::detail::thread_data_base* const thread_info=detail::get_current_thread_data(); void *void_thread_info = (void *) thread_info; void *void_this = (void *) this; // is somebody destructing the thread_data other than its own thread? // (remember that its own which should no longer point to it anyway, // because of the call to detail::set_current_thread_data(0) in thread_proxy) if (void_thread_info) { // == void_this) { __builtin_trap(); } } I should note that (as seen from the commented-out code) I had previously checked to see that void_thread_info == void_this because I was only checking for the case where the thread's current thread_info was killing itself. I have also seen cases where the value returned by get_current_thread_data is non-zero and different from "this", which is really weird. Also when I first wrote that version of the code, I wrote: if (((void*)thread_info) == ((void*)this)) and at run-time I got some very weird exception that said I something about a virtual function table or something like that - I don't remember. I decided that it was trying to call "==" for this object type and was unhappy with that, so I rewrote as above, putting the conversions to void * as separate lines of code. That in itself is quite suspicious to me. I am not one to run to rush to blame compilers, but... I should also note that when we did catch this happening the trap, we saw the destructor for ~shared_count appear twice consecutively on the stack in Xcode source. Very doubleweird. We tried to look at the disassembly, but couldn't make much out of it. Again - it looks like this is always a result of the shared_count which seems to be owned by the shared_ptr which owns the thread_info reaching zero too early. Help! Thanks, Andy Weinstein PS I cannot say for sure yet whether this problem occurs on the other platforms, though it looks like Windows is more solid that Mac. Android pretty solid also, though maybe in between Windows and Mac.

1 0

Re: [Boost-users] [Graph] "Weighted Bacon number" (depth from node) with dijkstra visitor
by Tasos 04 Feb '13

04 Feb '13

Yes, Im just blind, it was so simple. Thanks a lot for your help. Best, Tasos

1 0

[Graph] "Weighted Bacon number" (depth from node) with dijkstra visitor
by The Maschine 04 Feb '13

04 Feb '13

Hi all, In order to find the depth of a source_node from other nodes in my graph, without weights, I use this simple example of the bacon BFS visitor via the tree_edge(). I also count the number of nodes from that source. //////// class bfs_vis : public boost::default_bfs_visitor { public: bfs_vis(unGraph::vertex_descriptor source, boost::shared_ptr< VertexStorage> storage) : s(source), cont(storage) { } template <typename Edge, typename Graph> void tree_edge(Edge e, const Graph& g) const { typename boost::graph_traits<Graph>::vertex_descriptor u = source(e, g), v = target(e, g); cont->depth_from->at(v) = cont->depth_from->at(u) + 1.0; } template <typename Edge, typename Graph> void discover_vertex(Edge e, Graph& g) { cont->node_count = cont->node_count + 1; } private: boost::shared_ptr<VertexStorage> cont; unGraph::vertex_descriptor s; }; //////// What would be an equivalent visitor for Dijkstra? Node count is ok but how do I calculate a "weighted bacon number" with dijkstra (boost::property<boost::edge_weight_t, float>)?? Thanks, Tasos

2 1

Re: [Boost-users] Boost-users Digest, Vol 3349, Issue 1
by Torsten＠Robitzki.de 04 Feb '13

04 Feb '13

Hi Igor, Am 04.02.2013 um 14:15 schrieb boost-users-request(a)lists.boost.org: > Send Boost-users mailing list submit >> >> s.close(); // first time to cancel outstanding handlers >> s.close(); // a second time, later in the d'tor >> But what does 'Bad file descriptor' mean in this context? Is the underlying socket is somehow broken/corrupted? If so, there must be >> a bug in my code. If not, is it save to simply ignore the error? > > > It means that tcp::socket is closed (and the underlying native > descriptor is something like -1). do you think there is something wrong with closing the socket a second time? The documentation doesn't states any precondition. kind regards Torsten

2 1

[function / bind] How to register callbacks using boost::function/bind?
by Olivier Tournaire 04 Feb '13

04 Feb '13

Hi all, I am trying to use a callback mechanism for logging in my app. I thus have a class with : class coordinates_converter { public: coordinates_converter(int id); ~coordinates_converter(); void register_error_callback (const boost::function<void(std::string message)> &callback ) { _error_callback = callback; } private: boost::function<void(std::string message)> _error_callback; } Now, if I want to register a log error callback in another object instance, how should I do this? Cannot find the right syntax ... void another_class::do() { acute_coordinates_converter* converter = new coordinates_converter(2154); converter->register_error_callback( boost::bind(&another_class::log_error, converter, _1) ); } where log_error is a Qt SIGNAL: signals: void log_error(const std::string& error); Hope you could help. Rehads, Olivier

2 2

[asio] How to handle 'Bad file descriptor'
by Torsten＠Robitzki.de 03 Feb '13

03 Feb '13

Hello, I have problems to understand, what might causing a 'Bad file descriptor' exception to be thrown in my application. The exception is most likely thrown by a call to tcp::socket::close. From the logs I have, I'm quit certain, that the last function calls on a socket are: s.shutdown(tcp::socket::shutdown_receive); s.shutdown(tcp::socket::shutdown_send); s.close(); // first time to cancel outstanding handlers s.close(); // a second time, later in the d'tor I'm not sure, whether it's the first or the second call to close(), that causes the exception. But what does 'Bad file descriptor' mean in this context? Is the underlying socket is somehow broken/corrupted? If so, there must be a bug in my code. If not, is it save to simply ignore the error? The program is not single threaded, but there is only one boost::io_service object which is run by a single dedicated thread. Kind reagards Torsten P.S: if someone interested in the code: https://github.com/TorstenRobitzki/Sioux/blob/master/source/server/connecti…; it's open source

2 1

[program options] count problem
by Olivier Tournaire 03 Feb '13

03 Feb '13

Hi all, I am having troubles with program options. I try to test if at least some of the available options are passed on the command line, but fail for some reason using count("option_name"). Here is my code: int main(int argc, char** argv) { string hierarchy; vector<string> files, directories; double vertical_shift; bool show_progress, single_file, single_dir; int target_EPSG, nb_threads; po::options_description desc("Allowed options"); desc.add_options() ("help ,h" , "Produce help message") ("hierarchy ", po::value<string>(&hierarchy) , "process a hierarchy) ("file ", po::value<vector<string> >(&files) , "process the specified files") ("directory ", po::value<vector<string> >(&directories) , "process the specified directories"); po::variables_map vm; try { po::store(po::parse_command_line(argc, argv, desc), vm); if (vm.count("help")) { cout << desc << "\n"; return 1; } po::notify(vm); } catch (const po::error &e) { cerr << "Exception caught ! --> " << e.what() << endl; cout << desc << endl; return 1; } if (vm.count("hierarchy")) { path input_path(hierarchy); // DO STUFF } else if (vm.count("files")) { for(unsigned int i=0;i<files.size();++i) { path input_path(files[i]); // DO STUFF } } else if (vm.count("directories")) { for(unsigned int i=0;i<directories.size();++i) { path input_path(directories[i]); // DO STUFF } } else { cout << "You must choose an option between --hierarchy --files --directories!" << endl; cout << desc << endl; return 1; } I always fall in the "else". What am I doing wrong? For instance with this command line : ./acute_coordinates_transform --hierarchy ~/work/data/ À other question: is there à way to specify that 1 of the 3 options is required (and only 1)? Hope you cloud help. Regards, Olivier

2 4

SIGSEGV in libboost_serialization.so.1.40.0 type_unregister
by Prasanna .S 03 Feb '13

03 Feb '13

Hey guys, I'm getting a seg fault, after the program successfully runs, and is trying to exit. Stack trace is below. Program received signal SIGSEGV, Segmentation fault. 0x00007ffff709fa2f in boost::serialization::detail::extended_type_info_typeid_0::is_less_than(boost::serialization::extended_type_info const&) const () from /usr/lib/libboost_serialization.so.1.40.0 (gdb) bt #0 0x00007ffff709fa2f in boost::serialization::detail::extended_type_info_typeid_0::is_less_than(boost::serialization::extended_type_info const&) const () from /usr/lib/libboost_serialization.so.1.40.0 #1 0x00007ffff709faed in boost::serialization::detail::extended_type_info_typeid_0::type_unregister() () from /usr/lib/libboost_serialization.so.1.40.0 #2 0x000000000047985b in boost::serialization::detail::singleton_wrapper<boost::serialization::extended_type_info_typeid<DeltaStore::Location> >::~singleton_wrapper() () #3 0x00007ffff6567262 in exit () from /lib/libc.so.6 #4 0x00007ffff654cc54 in __libc_start_main () from /lib/libc.so.6 #5 0x0000000000411f59 in _start () I do `apt-get install libboost-all-dev` on a ubuntu 10.04 box; Any help is much appreciated. Thanks, -Prasanna $ g++ -v Using built-in specs. Target: x86_64-linux-gnu Configured with: ../src/configure -v --with-pkgversion='Ubuntu 4.4.3-4ubuntu5' --with-bugurl=file:///usr/share/doc/gcc-4.4/README.Bugs --enable-languages=c,c++,fortran,objc,obj-c++ --prefix=/usr --enable-shared --enable-multiarch --enable-linker-build-id --with-system-zlib --libexecdir=/usr/lib --without-included-gettext --enable-threads=posix --with-gxx-include-dir=/usr/include/c++/4.4 --program-suffix=-4.4 --enable-nls --enable-clocale=gnu --enable-libstdcxx-debug --enable-plugin --enable-objc-gc --disable-werror --with-arch-32=i486 --with-tune=generic --enable-checking=release --build=x86_64-linux-gnu --host=x86_64-linux-gnu --target=x86_64-linux-gnu Thread model: posix gcc version 4.4.3 (Ubuntu 4.4.3-4ubuntu5)

1 0

[qi] parser stocks inside an endless loop
by Boost_mailinglist＠gmx.de 03 Feb '13

03 Feb '13

Hello, the attached example will stock inside an endless loop if I use /*loop*/ Variable = *char_("a-zA-Z_0-9")>> ... but works if I use /*works*/ Variable = &char_("a-zA-Z")>> *char_("a-zA-Z_0-9")>> ... or /*works also*/ Variable = +char_("a-zA-Z_0-9")>> ... . Why does it happen? And how can I prevent this behaviour? Best regards

2 2

how to use boost::shared_ptr?
by young 01 Feb '13

01 Feb '13

I have a class member (boost::shared_ptr): class XYZ { private: boost::shared_ptr<int> m_pBuff; // other member ... public: void func(int n) { m_pBuff.reset(new int[n]); // or m_pBuff = boost::shared_ptr<int>(new int[n]) ...; } // other functions ... } The func(n) will be called many times inside / outside the class. Can I use shared_ptr here like: m_pBuff.reset(new int[n]); or m_pBuff = boost::shared_ptr<int>(new int[n]); I tried both and get error said heap corruption. -- View this message in context: http://boost.2283326.n4.nabble.com/how-to-use-boost-shared-ptr-tp4642120.ht… Sent from the Boost - Users mailing list archive at Nabble.com.

4 6

Re: [Boost-users] [Graph] visitor dijkstra_shortest_paths
by The Maschine 01 Feb '13

01 Feb '13

Thanks a lot, even though I read half of the BGL book I didnt get this difference. I cant say Im good with techniques like that. Best, Tasos > > ---------------------------------------------------------------------- > > Message: 1 > Date: Thu, 31 Jan 2013 11:12:29 -0500 (EST) > From: Jeremiah Willcock <jewillco(a)osl.iu.edu> > To: boost-users(a)lists.boost.org > Subject: Re: [Boost-users] [Graph] visitor dijkstra_shortest_paths > Message-ID: <alpine.LRH.2.03.1301311111140.64740(a)cs.indiana.edu> > Content-Type: text/plain; charset="iso-8859-1"; Format="flowed" > > On Thu, 31 Jan 2013, The Maschine wrote: > > > > > Hi all, > > > > > > Can someone explain me why I get an error with the first one and not the > second? > > > > > > 1) > (error)?boost::dijkstra_shortest_paths(m_ugraph,*vertex_iterator_begin,? ? > ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? > > ?boost::distance_map(boost::make_iterator_property_map(dist_map.begin(), > boost::get(boost::vertex_index, > > m_ugraph))), > > > > boost::visitor(dijkstra_vis( *vertex_iterator_begin, superStorage ) )); > > The syntax for named parameters in Boost.Graph requires periods, so all > named arguments show up as a single argument to C++. > > > 2) > (good)?boost::dijkstra_shortest_paths(m_ugraph,*vertex_iterator_begin,? ? ? > ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? > > ?boost::distance_map(boost::make_iterator_property_map(dist_map.begin(), > boost::get(boost::vertex_index, > > m_ugraph))).visitor(dijkstra_vis( *vertex_iterator_begin, superStorage > ))?); > > > > > > > > Why I need to "." the visitor on the distance_map? > > That's how it was implemented (before Boost.Parameter or variadic > templates existed). > > > > > > > BFS is ok with it: > > > > (good) boost::breadth_first_search(m_ugraph, *vertex_iterator_begin , > boost::visitor(bfs_vis( > > *vertex_iterator_begin, superStorage ) )); > > You only have one named argument there; if you had more, you would need to > use a period between those as well. > > -- Jeremiah Willcock > >

1 0

[Graph] visitor dijkstra_shortest_paths
by The Maschine 31 Jan '13

31 Jan '13

Hi all, Can someone explain me why I get an error with the first one and not the second? 1) (error) boost::dijkstra_shortest_paths(m_ugraph,*vertex_iterator_begin, boost::distance_map(boost::make_iterator_property_map(dist_map.begin(), boost::get(boost::vertex_index, m_ugraph))), boost::visitor(dijkstra_vis( *vertex_iterator_begin, superStorage ) )); 2) (good) boost::dijkstra_shortest_paths(m_ugraph,*vertex_iterator_begin, boost::distance_map(boost:: make_iterator_property_map(dist_map.begin(), boost::get(boost::vertex_index, m_ugraph))).visitor(dijkstra_vis( *vertex_iterator_begin, superStorage )) ); Why I need to "." the visitor on the distance_map? BFS is ok with it: (good) boost::breadth_first_search(m_ugraph, *vertex_iterator_begin , boost ::visitor(bfs_vis( *vertex_iterator_begin, superStorage ) )); Best, Tasos

2 1

[Fusion] Why restrict 'copy' to equal sizes?
by TONGARI 31 Jan '13

31 Jan '13

Hi, As the counterpart of STL copy, I expected the src just need to be <= dst, but it requires equal sizes in Fusion. What's the rationale? Is it OK to loosen the restriction? (it just compiles if I change '==' to '<=') Thanks.

2 3

How to build boost out-of-tree?
by Keitel, Tino (ALC NetworX GmbH) 31 Jan '13

31 Jan '13

Hi, I want to build boost out-of-tree. Usually, this is done like this: $ mkdir /path/to/build $ cd /path/to/build $ /path/to/source/configure $ make The actual source is not modified, all output is written in the build directory. The best I could find was http://www.boost.org/doc/libs/1_52_0/more/getting_started/unix-variants.htm…, but the bootstrap.sh step will still modify the source directory. How can I do a real out-of-tree build of boost? Regards, Tino

1 0

boost linkage problem
by Dave Dyer 30 Jan '13

30 Jan '13

I'm trying to construct a small library of wrappers for boost-using functions. In my test environment that links with this library (a windows executable) everything is fine. The working environment doesn't name any libraries explicitly. In another environment (the one I'm actually interested in) I get a large number of linkage errors, apparently seeking libstdc++ A small sample: 1>sharedmem-debug-x64.lib(sharedmemory.obj) : warning LNK4217: locally defined symbol ??1?$basic_string@DU?$char_traits@D@std@@V?$allocator@D@2@@std@@QEAA@XZ (public: __cdecl std::basic_string<char,struct std::char_traits<char>,class std::allocator<char> >::~basic_string<char,struct std::char_traits<char>,class std::allocator<char> >(void)) imported in function "public: static bool __cdecl boost::interprocess::shared_memory_object::remove(char const *)" (?remove@shared_memory_object@interprocess@boost@@SA_NPEBD@Z) 1>sharedmem-debug-x64.lib(sharedmemory.obj) : warning LNK4217: locally defined symbol ?c_str@?$basic_string@DU?$char_traits@D@std@@V?$allocator@D@2@@std@@QEBAPEBDXZ (public: char const * __cdecl std::basic_string<char,struct std::char_traits<char>,class std::allocator<char> >::c_str(void)const ) imported in function "public: static bool __cdecl boost::interprocess::shared_memory_object::remove(charconst *)" (?remove@shared_memory_object@interprocess@boost@@SA_NPEBD@Z) The non-working link is a complex environment, but I'm stumped why it doesn't link, or how to get it to link.

1 0

[phoenix] segmentation fault with valarray
by Philipp Schwaha 30 Jan '13

30 Jan '13

Hi all! I have worked with phoenix together with valarray and am currently stumbling over a case of segmentation fault. I construct the following phoenix expression boost::phoenix::at_c<0>(arg1) += boost::phoenix::at_c<1>(arg1) / val(mass) * arg2 I then feed a std::tuple<std::valarray<double>, std::valarray<double>> to it. It compiles nicely with gcc (4.6 & 4.7) using boost 1.52 as well as trunk. The resulting executable then proceeds to segfault, however. I have put together a small example (c++11) demonstrating my problem. The valarray's memory should already be reserved from original initialization. Performing the same operations manually with just the valarray not in phoenix seems to work fine. Am I missing something or doing something stupid here? thanks for any help! cheers Philipp

3 4

[threadpool] from your experience, is it safe to use threadpool with recent boost release?
by Frédéric Bron 30 Jan '13

30 Jan '13

Dear community, I found the threadpool library very useful (http://threadpool.sourceforge.net/) Apparently, it was designed for boost 1.37.0 and seemed to target inclusion in boost but probably did never go to a review because its author had not enough time to meet all requirements for boost inclusion, in particular, documentation. However, it is ready to be used with boost as is can just be untared in the boost include directory. I have tried to use it with boost 1.51.0 and it seems to work nicely. Before using it for long term projects, I was wondering if some of you use this library with success together with a rather recent boost release? Do you experience any issue or everything is just fine. Any chance that the latest changes in the thread library (http://thread.gmane.org/gmane.comp.lib.boost.devel/237484) would impact threadpool? Regards, Frédéric

4 6

hey
by Hossein Haeri 30 Jan '13

30 Jan '13

hey have a look http://www.com-im5.net/finance/ Hossein

1 0

[boost 1.47] boost::interprocess::scoped_lock<boost::interprocess::named_recursive_mutex> and fork() (linux)
by Frank Bergemann 30 Jan '13

30 Jan '13

Hello, i have a problem with boost::interprocess::scoped_lock<boost::interprocess::named_recursive_mutex> and using that one recursively in a unix (linux) fork() process environment. I.e. I create the boost::interprocess::named_recursive_mutex in a root process, then fork a number of child processes and those synchronize each other via boost::interprocess::scoped_lock<boost::interprocess::named_recursive_mutex>. When i use the named_recursive_mutex in a recursive way, the 2nd instance of the scoped_lock<> of it hangs (for one of the children, the first one, which tries to do a recursive lock.) I didn't have an issue without recursions (before i just used named_mutex - but i need to protect an application level region on a higher level now). And i didn't have a problem with a test program that does NOT fork(), which i start #10 instances on linux in parallel, and which also uses nested scoped_lock<>'s for each of them. Does anybody know about issue of boost::interprocess::named_recursive_mutex or boost::interprocess::scoped_lock<boost::interprocess::named_recursive_mutex> for unix/linux fork()? Do i have to create my named_recursive_mutex AFTER the fork() for each process instance? (i use m_mutex = boost::interprocess::named_recursive_mutex(boost::interprocess::open_or_create, _shmId.c_str())). I read about this here: http://www.linuxprogrammingblog.com/threads-and-fork-think-twice-before-usi… It talks about need to call |pthread_mutex_init()|. Other pages seems to indicate, that there is only a problem, when i lock a mutex and then fork(), inheriting the lock to parent and child process. - many thanks! best regards, Frank

1 0

how to create smart pointer array?
by young 30 Jan '13

30 Jan '13

What is wrong with: boost::scoped_ptr<byte> buff[10]; for(int i=0; i<10; i++) { int size = 2*i + 100; buff[i] = boost::scoped_ptr<byte>(new byte[size]); } ... -- View this message in context: http://boost.2283326.n4.nabble.com/how-to-create-smart-pointer-array-tp4642… Sent from the Boost - Users mailing list archive at Nabble.com.

3 2

Simulating for loop iteration with iota
by Algocoders . 30 Jan '13

30 Jan '13

We have written a free chapter on "Simulating for loop iteration with iota" @ http://www.algocoders.com/sites/default/files/chapter1_c%2B%2B11algorithms_… HTH, Cheers, Algocoders http://www.algocoders.com/

1 0

Building boost on Windows
by John Lilley 30 Jan '13

30 Jan '13

I am trying to figure out which libraries to build on Windows. I really only need "polygon" and "geometry". But what libraries these dependent on that also need to be built? I've tried to query which will be built and I'm not getting the expected results, for example: $ ./b2 --show-libraries --build-type=complete The following libraries require building: - chrono - context - date_time - exception - filesystem - graph - graph_parallel - iostreams - locale - math - mpi - program_options - python - random - regex - serialization - signals - system - test - thread - timer - wave Hmmmm... where is polygon and geometry? $ ./b2 --show-libraries --build-type=complete --with-geometry error: wrong library name 'geometry' in the --with-<library> option. Help please?

2 3

Custom library names
by John Lilley 30 Jan '13

30 Jan '13

How does one build boost with custom library/DLL names? There are two reason for this -- Our software simultaneously runs 32-bit and 64-bit versions of itself, and things are easier if the library names reflect this (e.g. inject _x64_ and _x86_ into the names). -- If we have customized Boost in any way, we want to give it a unique name so that it cannot possibly be incorrectly loaded by other code. Thanks, John Lilley

2 2

Mac OSX build options
by Tim Burgess 30 Jan '13

30 Jan '13

Hi, I'm trying to create an application for the Mac that is compatible with OSX 10.6 through 10.8, with both 32 and 64-bit support. How do I configure my Boost 1.50.0 to generate the appropriate binaries (I use serialisation, System and FileSystem), please? Best wishes. Tim Burgess Raised Bar Ltd. Tel: +44 (0)1827 719822 Web: http://www.raisedbar.net We can supply Roland and Avid products with specialist pre and post-sales support for visually-impaired users - just drop us a line for a quote. Don't forget to sign our on-line petition to lobby for improved access to music education and music technology at: http://www.raisedbar.net/petition.htm

1 0

Build options on Mac OSX 10;.8
by Tim Burgess 30 Jan '13

30 Jan '13

Hi, I'm trying to create an application for the Mac that is compatible with OSX 10.6 through 10.8, with both 32 and 64-bit support. How do I configure my Boost 1.50.0 to generate the appropriate binaries (I use serialisation, System and FileSystem), please? Best wishes. Tim Burgess Raised Bar Ltd. Tel: +44 (0)1827 719822 Web: http://www.raisedbar.net We can supply Roland and Avid products with specialist pre and post-sales support for visually-impaired users - just drop us a line for a quote. Don't forget to sign our on-line petition to lobby for improved access to music education and music technology at: http://www.raisedbar.net/petition.htm

1 0

C++11 Algorithms Book
by Algocoders . 30 Jan '13

30 Jan '13

The book "C++11 Algorithms Volume 1 : Using and Extending C++11, Boost and Beyond <http://www.algocoders.com/>" is published. *Table of Contents: http://www.algocoders.com/sites/default/files/toc_c%2B%2B11algorithms_vol1.… *<http://www.algocoders.com/sites/default/files/toc_c%2B%2B11algorithms_vol1.…> *Sample Chapter: http://www.algocoders.com/sites/default/files/chapter1_c%2B%2B11algorithms_… *<http://www.algocoders.com/sites/default/files/chapter1_c%2B%2B11algorithms_…> **<http://www.algocoders.com/sites/default/files/toc_c%2B%2B11algorithms_vol1.…> @Amazon : www.amazon.com/11-Algorithms-Volume-Extending-Beyond/dp/1482083396/<http://www.amazon.com/11-Algorithms-Volume-Extending-Beyond/dp/1482083396/> @Pothi.com (Indian Edition) : http://pothi.com/pothi/book/chandra-shekhar-kumar-c11-algorithms-volume-1<http://pothi.com/pothi/book/chandra-shekhar-kumar-c11-algorithms-volume-1> ** *Description* This book is vital to understand *algorithms* newly introduced in *C++11*with the help of practical examples illustrating *concepts*, *variations*, *customizations* and *correctness* with deep insight into internals with primary focus on *effective usage*. This book can be read by anyone having some experience in any higher level programming. Beginners in C++ will be able to learn basic concepts of C++11 algorithms with practical examples. Intermediate programmers in C++ will learn foundational aspect of C++11 algorithms in a pragmatic way. Expert programmers(aka C++ hackers) can enjoy interesting variations leading to future of C++11 algorithms(aka C++1y), Boost and beyond. This book(*Volume 1*) illustrates following algorithms: - *Numeric Algorithms* - Simulating for-loop iteration with iota - Customizing iota - Return Type of iota - Compile Time iota - Interesting variations of iota - *Quantifier Algorithms* - Universal Quantifier(*Predicate Satisfiability For All*) - Non-Existential Quantifier(*Predicate Satisfiability For None*) - Existential Quantifier(*Predicate Satisfiability For Some*) - Unique Quantifier(*Predicate Satisfiability For One*) - *Partition Algorithms* - Predicate Based Rearrangements - Partition Structure Validation - Bisection Algorithm - Group Partitions *Recommended Approach* Though this book can be read without reference to any other source, still we recommend our readers to keep a copy of the famous book *The C++ Standard Library, Second Edition : A Tutorial and Reference by Nicolai M. Josuttis* handy for gentle introduction to C++11 algorithms followed by diving into respective sections of our book for detailed information. In-depth treatment of foundational aspect of C++11 algorithms is covered in another book published by us Foundation of Algorithms in C++11, Volume 1(Revised Edition) : Using and Extending C++11, Boost and Beyond.<http://www.amazon.com/Foundation-Algorithms-Volume-Revised-Edition/dp/14819…> For more details, please visit the link : http://www.algocoders.com/article/books Enjoy Reading ! Cheers, Algocoders http://www.algocoders.com paranoids(a)algocoders.com

1 0

[Fusion] Associative Sequence not so associative
by TONGARI 30 Jan '13

30 Jan '13

Hi, I'd expect the associative sequence could be constructed and assigned with another sequence in different order. But it doesn't work, the src sequence must in the same order as the dst sequence defined. E.g. map<pair<K1, int>, pair<K2, int> > cannot take make_map<K2, K1>(1, 2). This is a bit disappointing :-( Does I miss something?

2 3

[Range][Iterator] iterate over an enumeration?
by John M. Dlugosz 29 Jan '13

29 Jan '13

I'd like to make it easy to do a BOOST_FOREACH for every value in an enumeration. I tried enum E { E_First, ⋯ , E_Last }; boost::counting_range(E_First, E_Last) but got compiler template barf. Can someone suggest a good way to do this?

2 1

serialization
by Larry E. Ramey 29 Jan '13

29 Jan '13

I'm having some trouble with the portable_binary_archive. When we serialze our classes via the text archive, they go in and out just fine. My simple test cases work great with the portable_binary_archive. However, in our production code, I get an input stream error exception tossed when I try to read a class that has STL elements in it. I have tried reordering includes and making sure the boost/serialization/list.hpp or whatever comes in front of the STL list class. Heck, I even put all those includes as the very 1st line of my test program to make sure they were included 1st. I am at a loss as the exception really doesn't tell me much, and the only thing I could figure out is that reordering the includes did in fact fix a few of these exceptions. (I still have one left... very annoying) Any ideas? I'd post the code, but it is very intertwined and difficult to distill a clean example. (and my stupid trail test does in fact work) Larry

2 1

regular expression to extract numbers
by Neil Sutton 29 Jan '13

29 Jan '13

Hi, I am new to using boost and am trying to learn basics so I can use regex features in my code. In my test program, I want a function to simply extract some numbers that follow after a certain string pattern and then to read those numbers into a vector. This is an outline of what i would like to do: // function definition #ifndef findnumbers.h #define findnumbers.h #include <string> #include <fstream> #include <vector> #include <boost/regex.hpp> #endif void findnumbers (ifstream& afile) { // assume file is opened in main - i have passed a reference boost::regex expression ( ....having trouble with this); // search for matches - there will be a fixed number of numbers e.g 10 std::string const_iterator start, end; start = afile.begin( ); end = afile.end( ); std::vector<int> results; boost::match_results<std::vector<int>::const_iterator> what; for(int count = 0; count !=10; count++) { boost::regex_search(start,end,what,expression) { //if a number is found place in the vector results.push_back(count); } So the first problem I have is defining a correct expression. The pattern has the format "sometext>NUMBERS<sometext" newline. The NUMBERS may be one or two digits combined and the pattern is repeated 10 times. I have searched the archives and tried to build something based on the examples I have read, but cannot seem to find a working solution. Can anyone help please in defining an expression? I do not need a solution to my function overall as I should be able to write it correctly once I know how to define the pattern. Kind regards

3 4

[GIL] Why does 'get_color' take a tag object instead of its type only?
by TONGARI 29 Jan '13

29 Jan '13

Hi there, I'm curious of why gil::get_color(pixel, tag) is defined that way. Any need to take the tag value? If not, I'd prefer gil::get_color<Tag>(pixel) or just gil::get<Tag>(pixel) or even Tag::get(pixel). And if possible, I hope GIL would just make pixel a Fusion sequence. Thanks.

3 6

stack_allocator
by Andrew Hundt 29 Jan '13

29 Jan '13

Is there some version of a stack allocator in the current 1.52 version of boost, perhaps in some detail namespace? I searched the source and the archives and the only one I could find disappeared after 1.39: http://www.boost.org/doc/libs/1_39_0/boost/signals2/detail/stack_allocator.… Cheers! Andrew Hundt

2 2

Which Parallel Boost Graph Library should I use?
by Christopher Cowing-Zitron 29 Jan '13

29 Jan '13

Hello, There is more than one PBGL server, one hosted by Boost, and the other by the Open Systems Laboratory (OSL). It appears that each has code not included in the other, whether I look at the release or sandbox SVN versions of the library. I have three criteria as to which version(s) I should use, which are all equally important to my final choice. First, in general, which is more reliable, the Boost or the OSL version? Both Boost and the OSL know their stuff, so it's not that I question the authors' skill, but rather I don't know which version has received more testing and validation. Second, which is more up-to-date, both in terms of which is updated more frequently, and which is more algorithmically and mathematically current? While I know programmers do love their graph algorithms, I wonder if the actual computer scientists and mathematicians at the OSL are more on the cutting edge. Finally, whether I use the Boost or the OSL server, I am more interested in the sandbox version than the release, as the development version has a lot of good stuff which isn't present in the release. That being said, in my experience there is quite a bit of variability amongst sandbox libraries as to their readiness for use, and I am not sure whether either PBGL sandbox is functional. Note, if there is some feature in one of the sandboxes that I am quite interested in, but is not yet at production quality, I wouldn't mind fixing it up for the community. Answers to any or all of the above questions would be most appreciated. Have a good day, -- Chris

2 1

narrowing conversion warning in boost's regex_traits_defaults.hpp when trying to initialize wregex object
by rsk82 28 Jan '13

28 Jan '13

The first line compiles ok. The second also produces working EXE, but in addition I get this large error: In file included from d:\boost/boost/regex/v4/regex_traits.hpp:35:0, from d:\boost/boost/regex/regex_traits.hpp:27, from d:\boost/boost/regex/v4/regex.hpp:39, from d:\boost/boost/regex.hpp:31, from test.cpp:4: d:\boost/boost/regex/v4/regex_traits_defaults.hpp: In instantiation of 'const charT* boost::re_detail::get_escape_R_string() [with charT = wchar_t]': d:\boost/boost/regex/v4/basic_regex_parser.hpp:804:54: required from 'bool boost::re_detail::basic_regex_parser<charT, traits>::parse_extended_escape() [with charT = wchar_t; traits = boost::regex_traits<wchar_t>]' d:\boost/boost/regex/v4/basic_regex_parser.hpp:308:36: required from 'bool boost::re_detail::basic_regex_parser<charT, traits>::parse_extended() [with charT = wchar_t; traits = boost::regex_traits<wchar_t>]' d:\boost/boost/regex/v4/basic_regex_parser.hpp:134:10: required from 'void boost::re_detail::basic_regex_parser<charT, traits>::parse(const charT*, const charT*, unsigned int) [with charT = wchar_t; traits = boost::regex_traits<wchar_t>]' d:\boost/boost/regex/v4/basic_regex.hpp:215:7: required from 'void boost::re_detail::basic_regex_implementation<charT, traits>::assign(const charT*, const charT*, boost::re_detail::basic_regex_implementation<charT, traits>::flag_type) [with charT = wchar_t; traits = boost::regex_traits<wchar_t>; boost::re_detail::basic_regex_implementation<charT, traits>::flag_type = unsigned int]' d:\boost/boost/regex/v4/basic_regex.hpp:661:4: required from 'boost::basic_regex<charT, traits>& boost::basic_regex<charT, traits>::do_assign(const charT*, const charT*, boost::basic_regex<charT, traits>::flag_type) [with charT = wchar_t; traits = boost::regex_traits<wchar_t>; boost::basic_regex<charT, traits>::flag_type = unsigned int]' d:\boost/boost/regex/v4/basic_regex.hpp:382:33: required from 'boost::basic_regex<charT, traits>& boost::basic_regex<charT, traits>::assign(const charT*, const charT*, boost::basic_regex<charT, traits>::flag_type) [with charT = wchar_t; traits = boost::regex_traits<wchar_t>; boost::basic_regex<charT, traits> = boost::basic_regex<wchar_t, boost::regex_traits<wchar_t> >; boost::basic_regex<charT, traits>::flag_type = unsigned int]' d:\boost/boost/regex/v4/basic_regex.hpp:367:48: required from 'boost::basic_regex<charT, traits>& boost::basic_regex<charT, traits>::assign(const charT*, boost::basic_regex<charT, traits>::flag_type) [with charT = wchar_t; traits = boost::regex_traits<wchar_t>; boost::basic_regex<charT, traits> = boost::basic_regex<wchar_t, boost::regex_traits<wchar_t> >; boost::basic_regex<charT, traits>::flag_type = unsigned int]' d:\boost/boost/regex/v4/basic_regex.hpp:336:7: required from 'boost::basic_regex<charT, traits>::basic_regex(const charT*, boost::basic_regex<charT, traits>::flag_type) [with charT = wchar_t; traits = boost::regex_traits<wchar_t>; boost::basic_regex<charT, traits>::flag_type = unsigned int]' test.cpp:15:37: required from here d:\boost/boost/regex/v4/regex_traits_defaults.hpp:329:73: warning: narrowing conversion of ''\37777777605'' from 'char' to 'const wchar_t' inside { } [-Wnarrowing] d:\boost/boost/regex/v4/regex_traits_defaults.hpp:331:64: warning: narrowing conversion of ''\37777777605'' from 'char' to 'const wchar_t' inside { } [-Wnarrowing] code : #include <iostream> #include <boost/regex.hpp> int main() { boost::regex narrow_str_regex("a+"); boost::wregex wide_str_regex(L"a+"); } boost: http://sourceforge.net/projects/boost/files/boost/1.52.0/boost_1_52_0.7z compiler: http://sourceforge.net/projects/mingwbuilds/files/host-windows/releases/4.7… ( and I know that this is the bug: https://svn.boost.org/trac/boost/ticket/7263 ) so, any tricks to get around this (some compiler switches?, I dunno) ? Or can I just ignore this error ? (Yes I need regex matching on wide strings.) -- View this message in context: http://boost.2283326.n4.nabble.com/narrowing-conversion-warning-in-boost-s-… Sent from the Boost - Users mailing list archive at Nabble.com.

3 3

[1.53.0] Beta release candidates 2 now available for testing
by Marshall Clow 28 Jan '13

28 Jan '13

Updated release candidate files for 1.53.0 beta 1 are available at http://boost.cowic.de/rc/ As always, the release managers would appreciate it if you download the candidate of your choice and give building it a try. Please report both success and failure, and anything else that is noteworthy. This helps ensure the candidates build OK before we push them out to SourceForge. The files (and associated md5s) are: 579b680bf55f53cd6d50adc391cb6c23 *boost_1_53_0_beta1_rc2.tar.gz 02098bb6da0e9c10712ae63c16e6aeea *boost_1_53_0_beta1_rc2.tar.bz2 998e2616a66c50e5693f563afb899002 *boost_1_53_0_beta1_rc2.zip e4fde6a869a3e21f8751ba8b7fd471a8 *boost_1_53_0_beta1_rc2.7z There were three changes between rc1 and rc2: * Fixes for compilation errors in Boost.Serialization * Documentation updates for Boost.SmartPtr * Updates to Boost.Threads Thanks, --The release managers

2 3

Re: [Boost-users] boost random in 2 dimensions
by Robert Patterson 28 Jan '13

28 Jan '13

@Gabriel >From your answer I understand that, the uniform distributed values are >generated seperately for x and y axis. Then both should be combined to >form a point over a plane. @Norbert i think this could do uniform distribution only over one co-ordinates. I am assuming you want to sample a point uniformly at random from a rectangular subset of the plane with the sides of the rectangle parallel to the axes. If not, the problem is a little more complex, but still soluble. However, for a square of size 1 Norbert is correct. The extension to a general rectangle is simple - just use two uniform_real_distributions (this will also give you floating point numbers as per your first post). Calls to two different uniform_real_distributions (or successive calls to one object) using the same engine object will give a point uniformly distributed in the rectangle. You are correct to point out that the individual coordinates are also uniformly distributed, but since they are independent they also yield a uniformly distributed point on the rectangle. Robert

2 1

Assertion `invariant_()' failed. in boost multi index
by Sunil S Nandihalli 28 Jan '13

28 Jan '13

Hi Everybody, Boost multi index is a very nice library. But I have a situation where the invariant_() assertion fails. One of the indices is ordered based on a score which is computed with a function. The score of the same element can change when computed at different times. but however, the scores change, score decay over time, such that the previously computed order will always be right. While I don't know if this is the issue, I wanted to make sure that this kind of scoring functions does not cause a problem because of some caching that boost may be doing internally .. and also, the return value of the scoring function is a double which I can imagine can introduce some comparison issues. I am digging into the code to make sure that that is not the problem .. any pointers from the community is greatly appreciated. Thanks, Sunil.

2 1

[spirit/qi] Error handler not called
by Szymon Gatner 28 Jan '13

28 Jan '13

Hi, I am trying to develop a JSON parser using spirit/qi so far it is not going bad but I am having serious problem with on_error and error handler. This is my code: namespace qi = boost::spirit::qi; namespace ascii = boost::spirit::ascii; namespace phoenix = boost::phoenix; template <typename Iterator> struct error_handler { template <typename, typename, typename, typename> struct result { typedef void type; }; void operator()(Iterator first, Iterator last, Iterator err_pos, const boost::spirit::info& info) const { std::cout << "never called"; } }; template <class Iterator> struct JsonGrammar : public qi::grammar<Iterator, Value(), ascii::space_type> { JsonGrammar() : JsonGrammar::base_type(json, "json"), error(error_handler<Iterator>()) { using qi::double_; using qi::lexeme; using qi::char_; using qi::int_; using qi::on_error; using qi::fail; using phoenix::construct; using phoenix::val; using namespace qi::labels; qi::on_error<qi::fail>(json, error(_1, _2, _3, _4)); string %= lexeme['"' >> +(char_ - '"') >> '"']; number = int_; member = string >> ':' >> value; array = '[' >> value % ',' >> ']'; object = '{' >> member >> *(',' >> member) >> '}'; value = array | object | number | string | qi::bool_ | "null"; json = array | object; json.name("json"); } phoenix::function<error_handler<Iterator> > const error; qi::rule<Iterator, std::string(), ascii::space_type> string; qi::rule<Iterator, int(), ascii::space_type> number; qi::rule<Iterator, Value(), ascii::space_type> json; qi::rule<Iterator, Array(), ascii::space_type> array; qi::rule<Iterator, Value(), ascii::space_type> value; qi::rule<Iterator, Object(), ascii::space_type> object; qi::rule<Iterator, std::pair<std::string, Value>(), ascii::space_type> member; }; now when i try to parse std::string json = "[1, 2, 3, 4, t]"; parising result is false. but error handelr is never called. What am I doing wrong? Regards, Simon -- Szymon Gatner The Lordz Games Studio www.thelordzgamesstudio.com

3 4

boost random in 2 dimensions
by Danny 28 Jan '13

28 Jan '13

I am trying to use a boost random generator to make random points uniformly distributed over a plane surface. In the the following link, the code for doing it in a single dimension is available: http://stackoverflow.com/questions/2254909/boost-random-number-generator Here they use boost::uniform_int<> to generate numbers as int in a single dimension. But in my case I wish to generate numbers as float in two dimensions over a plane. Is there any distribution type available to use it in two dimensions? (I have seen boost::uniform_on_sphere, but it is for spherical surfaces.) Is it possible to to this using boost library? -- View this message in context: http://boost.2283326.n4.nabble.com/boost-random-in-2-dimensions-tp4641761.h… Sent from the Boost - Users mailing list archive at Nabble.com.

3 5

[ICL] interval_map::find returns only const iterator
by CarlosRoVe 27 Jan '13

27 Jan '13

Hi, I am using boost v.1.49.0 and I have a question about ICL. We are building a system that stores information associated with time intervals. For this we are using the interval_map container, and every element that we insert in the map contains a right_open interval as the key, and an instance of the structure that holds the information as the value. Some of the features of the system are: 1. Minimum interval length is one second. 2. We have seen that the default behavior for interval_map is aggregate on overlap, but in our case we enforce that overlapping time intervals are not allowed. To achieve this, every time we try to insert an element we perform a check if the time interval overlaps with any of the existing intervals in the map, and if so, we ignore the element. 3. We should be able to modify the data associated with a time interval. We have seen that the find method of the interval_map container returns only a const iterator, but we would like to have a version of find that returns a non-const iterator as well. We want to be able to search for an element whose time interval contains a certain timestamp, and change its content. With the given requirements, find should return one element at most. For us it would be handy if ICL could also provide a version of find that returns a non-const iterator, but maybe I this would break the philosophy behind the design of ICL. We would be glad to hear ideas suggesting what we could do, or alternatives ways of dealing with other problem. Thank you very much. -- View this message in context: http://boost.2283326.n4.nabble.com/ICL-interval-map-find-returns-only-const… Sent from the Boost - Users mailing list archive at Nabble.com.

2 1

boost::asio async
by huvcbo 26 Jan '13

26 Jan '13

Hello, I have a problem I hope I can get some help with. I use boost http-server3, I have a number of tcp/ip-client started and connects to a server, every 15 seconds sending tcp/ip-client a keepalive message. When an HTTP client connects get tcp/ip-client from a container, and from the http message, a new message must be sent via the tcp/ip-client, when the reply is received how can I return it to the http-session? When i send the message to the tcp/ip-client the http-session goes out of scope. Both http-server and tcp/ip-client use async_write and read. Appriciate any help. /Christer

1 0

[accumulators] accumulator "adapters"?
by Pete Bartlett 25 Jan '13

25 Jan '13

Dear all, The accumulators library is a very nice way to get a set of statistics for a stream of numbers, doubles, say. Now let’s suppose I’m receiving a stream of boost::optional<double>s – e.g. let’s suppose I’m receiving simulation results and some simulations fail. It would be neat if I could somehow write and “adaptor accumulator” that took this stream of optionals and then, if they are not empty, pass them to an underlying accumulator. In rough terms: template<typename UnderlyingAccumulator , typename Sample > struct accumulator_adaptor { template<typename Args> void operator()( Args const& args) { Sample const& s = args[sample]; if( s ) { m_underlying_accumulator( *s ) ; //doesn’t compile UnderlyingAccumulator expects an Args, not a Sample } } private: UnderlyingAccumulator m_underlying_accumulator; }; (we might also define another accumulator that just counts the number of “bad” samples. Then by adding an appropriate feature adaptor, one could write something like accumulator_set< boost::optional<double> , tag::accumulator_adaptor< mean > , tag::bad_samples_count > For the data 1.0,2.0,boost::optional<double>(),4.0 This would return a mean of 7/3 and a bad_samples_count of 1. Is it possible for me to implement such an adaptor that would work for an arbitrary underlying accumulator? Thanks for your interest, Pete

3 2

Does Boost have polymorphic functors as named-operators?
by TONGARI 25 Jan '13

25 Jan '13

Hi all, Is there something like -- functional::identity, functional::plus (not std::plus<T>), functional::assign...etc -- in Boost? I know I can use Phoenix _1, _1 + _2, _1 = _2..., but I'd expect something more basic. Sometimes I just need something like that but not willing to have all the ET things... Maybe they're just provided within Phoenix or Proto? Thanks for any help.

1 0

HTTP server3 example question
by huvcbo 25 Jan '13

25 Jan '13

Hi all, I have problem with HTTP Server3 example, in connection::handle_write i have a dedline_timer with a async_wait. but before the handle_timed execute the connection_ptr goes out of scope, why? Appriciate any help. Christer The boost HTTP Server 3 in file connection.cpp in server.hpp boost::shared_ptr<connection> connection_ptr; server.cpp void server::handle_accept(const boost::system::error_code& e) { if (!e) { new_connection_->start(); new_connection_.reset(new connection(io_service_)); acceptor_.async_accept(new_connection_->socket(), boost::bind(&server::handle_accept, this, boost::asio::placeholders::error)); } } in connection.cpp void connection::handle_write(const boost::system::error_code& e) { if (!e) { const boost::shared_ptr<boost::asio::deadline_timer> t(new boost::asio::deadline_timer(*io_service_, boost::posix_time::seconds(15))); } } void connection::handle_timed( const boost::system::error_code& ec, const boost::shared_ptr<boost::asio::deadline_timer> t) { syslog(LOG_INFO, "handle_timed: %s", ec.message().c_str()); }

2 1

boost::bind/boost::function and fast delegates
by Littlefield, Tyler 24 Jan '13

24 Jan '13

hello all: We have a project we're working on that will be distributed, and I didn't want to have to set boost as a dependency if we're only using boost::bind and boost::function. Given this, I had a couple steps. First I thought I could possibly just copy out the code and modify it a bit, but it uses a ton of boost::preprocessor, so that'd require copying that out too. I'd rather not imbed half of boost in the project. My other solution (and one I really like) was to write my own delegates. I set to work on this and it's coming out good, but I have a couple questions. I know you can assign function pointers to boost::function objects, though I've always done it through bind. I'm curious how this works though: I have a normal function pointer set up and working right now, but how does boost::function take other pointers (say to static methods)? Also I'm curious what boost::bind generates. I'd like to create my own version since I love using it to pass in arbitrary values/using placeholders, but the boost::function has to be able to take a newly constructed function object and bind to that from what I understand. The new function object somehow has it's arguments stored on the object so that the () operator can just pass those in. Any help here would be appreciated. Thanks, -- Take care, Ty http://tds-solutions.net The aspen project: a barebones light-weight mud engine: http://code.google.com/p/aspenmud He that will not reason is a bigot; he that cannot reason is a fool; he that dares not reason is a slave.

6 6

Re: [Boost-users] intrusive_ptr vs shared-ptr
by John M. Dlugosz 24 Jan '13

24 Jan '13

Gottlob Frege<gottlobfrege(a)gmail.com> wrote: > Yes, I was wondering if the magic call inside of smart_ptr's constructor, > that handles the enabled_shared_from_this, wouldn't be better as a free > function, like smart_pointer_being_created(), which gives you an extension > point to go look up whether the object already had a smart pointer. It > could default back to the enabled_shared_from_this trick if you don't > override it for your types, etc. > > Qt's smart pointer has extra tracking like this - typically only turned on > in debugging. The issue boils down to non-intrusiveness. I used a base class 'can_handle' for types I planned on using smart pointers on, much as you inherit from enable_shared_from_this but without the need for CRTP. The free function that looks up the 'lifetime object' for any given pointer (which may or may not already exist in the system) can be overloaded for different ways of doing it for different classes, and inheriting from that base allows it to match the intrusive form. Now as it turns out I never needed to make shared_ptr's to types that I did not define (or trivially derive from), so it never bothered me; but the intent was that a non-intrusive general solution would use a map or something. Only later did I figure out the catch: knowing when the map can be cleaned up of counts to deleted objects. Certainly it can be done: the logic in the smart pointer "I hit zero, free it" code can call another free function to report it; the deleter can capture the need to supplement the normal destructor with additional cleanup; etc. Certainly with the maturity of current compilers there is no reason not to support "lost and found" in an efficient manner. If non-intrusive, it only pays for it when you construct a smart pointer from a raw pointer (potential "find"), and for creating fresh objects a make_shared bypasses that step and knows that it does not exist already in the system anyway. Or, just as you have shared_from_this available now only if enabled, you can have lost-and-found available only if enabled using the intrusive mechanism. Basically, shared_from_this would always work whether it was "in the system" already or not, and constructing a shared_ptr from a raw ptr that has such a base class would call shared_from_this on it rather than new'ing up the counter object. Easy. Hmm, a non-standard shared_from_this mechanism that works this way could still interoperate with code that expects standard shared_ptr class to be used, if you call it that way yourself when you need to "find". Standard code never will, and constructing a shared_ptr the normal way still works as it always did. Now I have noticed that the implementations for shared_from_this interact with internals of weak_ptr. But a portable implementation doesn't have to. If I had time to write for magazines (and such magazines still existed) I'd write that up for an issue. —John

2 1

(no subject)
by khaled hamidouche 24 Jan '13

24 Jan '13

http://www.lightbox.sk/partneri/wp-content/plugins/yahool321.php

1 0

[1.53.0] Beta1 release available
by Marshall Clow 24 Jan '13

24 Jan '13

Boost release 1.53.0 beta 1 is now available from SourceForge. See https://sourceforge.net/projects/boost/files/boost/1.53.0.beta.1/ This release contains numerous enhancements and bug fixes for existing libraries. For details of what's in the release, see http://www.boost.org/users/history/version_1_53_0.html. Please download the beta, give it a try, and report any problems you encounter. Thank you, -- The Boost Release Team Beman Dawes, Eric Niebler, Rene Rivera, Daniel James, Vladimir Prus and Marshall Clow.

3 9

[MultiIndex] question on replace() documentation
by John M. Dlugosz 24 Jan '13

24 Jan '13

“|replace| performs this substitution in such a manner that: … Iterator and reference validity are preserved.” I suppose "Iterator validity is preserved" means that no iterators are invalidated by this operation. Does that refer to the iterator passed as the first argument, too? That is, the iterator that referred to the replaced element now refers to the new element? And other indexes' iterators referring to the same element (e.g. project()) too? That might still be "valid" in the STL sense but might not continue to work if the position changed (e.g. it is now past the end of the range you are following). That's presuming the changed (but still "valid") iterator follows the new sort order rather than remembering the original order. Can someone clarify? My original reason for writing was to as what is meant by "reference validity is preserved"? Does that mean that the address of the items stored in the collection are not changed, and never move around as a result of this operation? —John

2 1

Re: [Boost-users] [Graph] betweenness_centrality - output completed percentage?
by The Maschine 24 Jan '13

24 Jan '13

Hi Nick, Thanks a lot for both comments! We currently use an "approximation/sub sampling" algorithm done 10years ago by a colleague but its take 10 times more than BGLs and Im not sure its statistically verified. The paper seems really good. About the brandes_betweenness_centrality_impl(), I suppose you are implying that I have to change the boost.graph header file directly? Right? Best, Tasos > > Message: 5 > Date: Wed, 23 Jan 2013 20:54:16 -0500 > From: Nick Edmonds <ngedmond(a)cs.indiana.edu> > To: boost-users(a)lists.boost.org > Subject: Re: [Boost-users] [Graph] betweenness_centrality - output > completed percentage? > Message-ID: <F264944D-B0A4-4FB6-BDF0-635E9DCCC199(a)cs.indiana.edu> > Content-Type: text/plain; charset="us-ascii" > > Hi Tasos, > > The outer loop of betweenness centrality is a loop over all the vertices > in the graph with each used as the source of a shortest paths computation. > If you want to see how far along the algorithm is you can just instrument > the "for (boost::tie(s, s_end) = vertices(g); s != s_end; ++s)" in > brandes_betweenness_centrality_impl(). Note that you don't actually need > to count shortest paths from *every* vertex in the graph to get a good > approximation of betweenness centrality, check out this paper [1]. > Sampling the vertices in the graph and counting shortest paths from a > subset of them should significantly speed up betweenness centrality if a > (relatively good) approximation will work for your application. > > Cheers, > Nick > > [1] http://www.cc.gatech.edu/~bader/papers/ApproxBC.html > > On Jan 22, 2013, at 8:04 AM, Tasos wrote: > > > Hello all, > > > > I calculate the betweenness-centrality of a large graph > > (V:150k, E:500k, boost::adjacency_list<boost::vecS, boost::vecS, > > boost::undirectedS,) with boost::brandes_betweenness_centrality > > and I can find how to get a indication of the 'completed' > > part of the algorithm. > > > > For example this graph took 1h:38m (single thread 10.8 iMac 3.4ghz) > > to complete but in order to be able to know the execution time > > of larger graph im a bit lost... > > > > On "one to all" BFS you know how many vertices you > > have left but this algorithm seems to me lacking a custom > > visitor or something in order to output some data. > > > > Any ideas as Im really new to boost.graph? > > > > Thanks a lot, > > Tasos > > > > _______________________________________________ > > Boost-users mailing list > > Boost-users(a)lists.boost.org > > http://lists.boost.org/mailman/listinfo.cgi/boost-users > >

1 0

Re: [Boost-users] Range library suggestion
by John M. Dlugosz 24 Jan '13

24 Jan '13

On 1/24/2013 3:59 AM, boost-users-request-at-lists.boost.org |Boost/Allow to home| wrote: > >> >I just added my second metafunction of the form is_range_of_X<T> for use >> >with enable_if, and although the first one took some fiddleing to pass >> >both compilers I was using, the second was a clone of the first with the >> >name and type changed. So it really ought to be a general purpose >> >is_range_of<Whatever, T> metafunction. > I assume it it's simply something like > is_same<typename mpl::eval_if< is_range<T>, range_value<T>, void>::type, > Whatever> I ended up with: template <typename Flag, typename T> struct is_range_of_MYTHING2 : boost::mpl::false_ { }; template <typename T> struct is_range_of_MYTHING2<boost::mpl::true_, T> { typedef typename boost::range_value<T>::type val_type; // chokes here if T is not a range typedef typename is_convertible<val_type, MYTHING>::type type; }; template <typename T> struct is_range_of_MYTHING : is_range_of_MYTHING2<typename boost::has_range_iterator<T>::type, T> { }; I'm familiar enough with type_traits, but not with all of Boost MPL. Is eval_if a lazy-evaluation for short circuiting the AND of the two tests? ---John

1 0

boost asio disign
by Christer Borgqvist 24 Jan '13

24 Jan '13

Hello, how can I design an application that will have a http server like HTTP Server 3 example, but also x number of tcp clients connecting to a server, client must connect to the server before the http server is active, if client is idle with a timer events send an 'I'm here' in every 10 seconds, i tink i have to put the tcp client on a list or ?. Data is received by the http server get the client from list, if non available wait for x amount of time, if time elapsed return incomming data, else, parsing on incomming data, send to tcp server, parse response data put the client back on the list, send response on http, when sending data via tcp client reset the timer. I have the http server working but how do I do with the tcp client, design the threads, should there be only one io_service?, I dont now, im lost! Appreciate all the help. /Christer

3 3

[Graph] betweenness_centrality - output completed percentage?
by Tasos 24 Jan '13

24 Jan '13

Hello all, I calculate the betweenness-centrality of a large graph (V:150k, E:500k, boost::adjacency_list<boost::vecS, boost::vecS, boost::undirectedS,) with boost::brandes_betweenness_centrality and I can find how to get a indication of the 'completed' part of the algorithm. For example this graph took 1h:38m (single thread 10.8 iMac 3.4ghz) to complete but in order to be able to know the execution time of larger graph im a bit lost... On "one to all" BFS you know how many vertices you have left but this algorithm seems to me lacking a custom visitor or something in order to output some data. Any ideas as Im really new to boost.graph? Thanks a lot, Tasos

2 1

Boost.Process and STARTUPINFOEX
by Nelson Teixeira 23 Jan '13

23 Jan '13

Hello, First, if this is not the correct place to ask, please point me an appropriate forum/maillist about Boost.Process as I couldn't find any in their site. Thanks. I'm trying to use Boost.Process library in windows XP. I installed Boost libraries version 1.52, and used Boost.Process from this page: http://www.highscore.de/boost/process0.5/index.html however, when I made the includes in my test program, I get this error: C:\boost/boost/process/windows/executor.hpp:120:5: error: 'STARTUPINFOEX' does not name a type C:\boost/boost/process/windows/executor.hpp: In constructor 'boost::process::windows::executor::executor()': C:\boost/boost/process/windows/executor.hpp:24:24: error: 'EXTENDED_STARTUPINFO_PRESENT' was not declared in this scope C:\boost/boost/process/windows/executor.hpp:30:23: error: 'startup_info_ex' was not declared in this scope C:\boost/boost/process/windows/executor.hpp:34:9: error: 'STARTUPINFOEX' was not declared in this scope I can't find this STARTUPINFOEX. At most I have STARTUPINFO structure. Do I have to install anything else ? Can anyone help please ?/ Thanks -Nelson ------------------------------------------------- Blog: Além do Óbvio http://www.alemdoobvio.com

4 3

intrusive_ptr vs shared-ptr
by ilya shlyakhter 23 Jan '13

23 Jan '13

Dear Boosters, The documentation of intrusive_ptr says "As a general rule, if it isn't obvious whether intrusive_ptr better fits your needs than shared_ptr, try a shared_ptr-based design first." but does not explain why. If the classes involved are all my own and deriving them all from a base class that implements reference-counting is not a problem, what are some reasons to prefer shared_ptr to intrusive_ptr? Thanks, Ilya

5 8

Range library suggestion
by John M. Dlugosz 23 Jan '13

23 Jan '13

I just added my second metafunction of the form is_range_of_X<T> for use with enable_if, and although the first one took some fiddleing to pass both compilers I was using, the second was a clone of the first with the name and type changed. So it really ought to be a general purpose is_range_of<Whatever, T> metafunction. I would suggest that such a thing be added to the Range library. It is meant for overloaded forms that operate on ranges, and you don't want to accept just any ol' thing because, perhaps, you want other overloads or to get better error messages for bad arguments. A typical example would be something like: void insert (const C& x); template <typename R> typename enable_if<typename is_range_of<C,R>::type, void>::type insert (const R& range) { ⋯ } In my case, I check whether R is indeed a Range (something that appears at first to be missing from the library too, but has_range_iterator is probably what is really meant by that) and the underlying element is convertible to the desired type. In a general-purpose library you might need to put more thought into what tests are so common as to be worth reusing at that level. I think having convertable-to-C (where you just give C) and another one which lets you pass in another metafunction for matching the element would allow reuse of experly written common code that takes care of nuances across compilers, e.g. choking on value_type if the outer is NOT a range, so you need to ensure short-circuit evaluation of the meta-expression. —John

2 1

[mpl] Suggestion on handle protocol messages
by Daniele Barzotti 23 Jan '13

23 Jan '13

Dear, I need to handle a protocol defined with Protocol Buffers. My messages are defined as: enum EMessages { E_MSG_METHOD_CONNECT = 0x8001, E_MSG_EVENT_CONNECT = 0xA001, ... } struct MsgHeader { long sessionRef; long transactionId; long status; } struct MSG_METHOD_CONNECT { Messages opCode; MsgHeader header; .. other fields .. } Now, I defined an interface and a template class to add a level of indirection.. class IMessage { virtual INT getOpCode() = 0; virtual STRING getName() = 0; virtual size_t getSize() = 0; .... } template<class MESSAGE> class IMessageImpl : public IMessage { protected: MESSAGE m_Message; ///< The Message Implementation public: virtual MESSAGE& getMessage() = 0; }; Where I'll use IMessageImpl<MSG_METHOD_CONNECT> MsgConnect; Now, the problem in this situation is to build the correct message class on the receiving part, knowing only the Message opCode. I would like to manage the messages with a type list or map, but I'm a newbie in MPL... Is there a way to define something like this: boost::mpl::map<int, MESSAGE> where the key is the Message opCode and the MESSAGE is the message type? The idea was born reading this article: http://blog.florindinu.ro/posts/message-dispatching-in-c-part-1-type-lists.… Thanks in advance, Daniele.

1 0

Boost and OMPL : access graph vertex
by Dan 23 Jan '13

23 Jan '13

Hello, I'm using OMPL (the Open Motion Planning Library) that makes use of Boost. The problem I've got is that I need to access the content of the vertexes of a graph which is returned by an OMPL function. In particular, the OMPL function is const Graph& g = planner->as<ompl::geometric::PRM>()->getRoadmap(); Now, given the graph g, and a vertex v, I need to read the state contained in g[v] and copy it to an OMPL variable. I'd need something like const ompl::base::State* state = FUNCTION_GET_STATE(g,v) ; This way I would be able to extract from variable state the components x,y,z I'm interested in. Is there any way to implement the FUNCTION_GET_STATE ? Thank you! Dan -- View this message in context: http://boost.2283326.n4.nabble.com/Boost-and-OMPL-access-graph-vertex-tp464… Sent from the Boost - Users mailing list archive at Nabble.com.

1 0

Re: [Boost-users] boost::bind/boost::function and fast, delegates
by John M. Dlugosz 23 Jan '13

23 Jan '13

On 1/22/2013 4:34 AM, boost-users-request-at-lists.boost.org |Boost/Allow to home| wrote: > And if you still wish to write such a lib(s) take a look at this article: > > http://www.codeproject.com/Articles/7150/Member-Function-Pointers-and-the-F… Coincedentely, that article references my project that I cited on another thread (shared pointers): > For Microsoft's (and Intel's) ugly, non-standard 12 byte |virtual_inheritance| pointers, > yet another trick is used, based on an idea invented by John Dlugosz. To pass bound member functions to legacy code (including OS API) that expects a genuine function pointer, I generated a thunk at run-time and used a compiler trick to get the information I needed regarding the 'this' adjustment. It is certainly the _fastest_ possible; but I did not elaborate on a full "bind" feature like Boost (and std) eventually came up with. In fact, I simplified my own code so that all the arguments to pass were just members of one structure, so I only worried about the case of binding to a member function with one argument. It didn't automatically capture smart-pointer objects for the binding, but I handled that separately since it was already being done for "windows" owned by the OS's reference to raw pointers; I had my own fake +1 count as part of the object, and Window objects were my primary interest. As I recall, the C++11 lambda spec provides for a primitive C function pointer to be coughed up, but I've not yet used a compiler that can do that.

1 0

Re: [Boost-users] intrusive_ptr vs shared-ptr
by John M. Dlugosz 23 Jan '13

23 Jan '13

> Well I wouldnt say these items are 'marginal', because most come from real life > scenarios. You can work around things but it is always that u have to remember that > there were some boundary cases with shared_ptr. It took me some time to convince my > colleagues to swap from raw pointers to shared pointers and having some boundary cases > doesn't help then. I still prefer shared pointers, but I am not sure if all in all they > would be better than e.g. intrusive_ptr's with ref counting (even without lacking the > weak_ptr feature). Note that i look upon this from a client perspective (i.e. large data > acquisition application), not from a library builder perspective. I really miss having a "lost and found" feature, which (in the late 90's) I wrote about as being one of the six essential features of my smart pointer design. That is, given a raw pointer p that is already known to the smart pointer system, creating another smart pointer from p will work in harmony with them, NOT break it. I've repeatedly come across code that hooks up objects with other objects in the constructor and I've had to work-around that in various ways, usually by adding a Create wrapper that finishes the work, and sometimes fiddling more than I would have wanted to with the design. I've also found cases where it is really annoying not knowing whether another smart pointer has already been created yet for the object. Anyway, for intrusive smart pointers, the "lost and found" feature is very easy and efficient to implement. So perhaps it should be part of Boost intrusive_ptr as an extension to the standard, or another class can be added with a different feature set. Anyway, if you want a class like that, feel free to grab my old code from the "Classics" or "Repertoire" project parked on github. —John

3 2

[Program Options] terminate parsing after first positional argument
by Stuart Pook 22 Jan '13

22 Jan '13

Hello, How can I tell Boost Program Options to stop parsing the arguments as soon as it finds a positional argument? I'd like to be able to parse the argument list "-v chmod -R -v 600 foo" as having one argument for my program (-v) and the rest as a std::vector of std::string (including the -v). In the attached program I have tried 2 different approaches. The first uses multitoken with the arguments "--command date -u -v", it fails because "-u" isn't known. The second uses collect_unrecognized with "chmod -R -v 600", it doesn't work because -v is interpreted by boost and doesn't end up in the vector. As someone said back in 2005, perhaps I'm trying to write xargs: http://thread.gmane.org/gmane.comp.lib.boost.user/9529/focus=9641 I know that I can ask my users to write a.out -v -- chmod -R -v 600 foo but I don't want to do that. I want them to write "a.out -v chmod -R -v 600 foo". The first positional argument should should tell boost to stop processing. Another slightly different question. If I have a multitoken option that takes doubles how can I get it to accept negative values? If I give my program "--floats 1 8 -4 6" where --floats starts a multitoken option that accepts double, I get an exception saying that "-4" is an unknown option. thanks Stuart This is what my test program gives me with boost 1.50: argv: dd --command date -u -v unrecognised option '-u' Allowed options: -v [ --verbose ] verbose --command arg command to run 2nd try argv: dd chmod -R -v 600 I'm verbose v: chmod -R 600 3rd try argv: dd --floats 1 8 -4 6 unrecognised option '-4' Allowed options: -v [ --verbose ] verbose -f [ --floats ] arg numbers and my program (formatted to fit in 80 characters) #include <boost/program_options.hpp> #include <boost/range/algorithm.hpp> #include <boost/program_options/parsers.hpp> #include <boost/program_options/options_description.hpp> #include <iostream> namespace po = boost::program_options; #define nelements(a) (sizeof (a) / sizeof (a)[0]) void print(char const * * b, char const * * e) { std::cout << "argv: "; std::copy(b, e, std::ostream_iterator<char const *>(std::cout, " ")); std::cout << std::endl; } int main(int argc, char ** argv) { { bool verbose; std::vector<std::string> command; po::options_description desc("Allowed options"); desc.add_options() ("verbose,v", po::bool_switch(&verbose)->default_value(false), "verbose") ("command", po::value<std::vector<std::string> >(&command) ->multitoken(), "command to run") ; char const * args[] = { "dd", "--command", "date", "-u", "-v", 0 }; print(args, args + nelements(args) - 1); po::variables_map vm; try { po::store(po::command_line_parser(nelements(args) - 1, args) .options(desc).run(), vm); po::notify(vm); if (verbose) std::cout << "I'm verbose\n"; std::cout << "command: "; boost::copy(command, std::ostream_iterator<std::string>(std::cout, " ")); std::cout << std::endl; } catch (boost::program_options::unknown_option & x) { std::cerr << x.what() << "\n" << desc << "\n"; } } std::cout << "2nd try\n"; { bool verbose; po::options_description desc("Allowed options"); desc.add_options() ("verbose,v", po::bool_switch(&verbose)->default_value(false), "verbose") ; char const * args[] = { "dd", "chmod", "-R", "-v", "600", 0 }; print(args, args + nelements(args) - 1); po::variables_map vm; po::parsed_options parsed = po::command_line_parser(nelements(args) - 1, args). options(desc).allow_unregistered().run(); try { po::store(parsed, vm); } catch (boost::program_options::unknown_option & x) { std::cerr << x.what() << "\n" << desc << "\n"; exit(2); } po::notify(vm); if (verbose) std::cout << "I'm verbose\n"; std::cout << "v: "; std::vector<std::string> v = po::collect_unrecognized( parsed.options, po::include_positional); boost::copy(v, std::ostream_iterator<std::string>(std::cout, " ")); std::cout << std::endl; } std::cout << "3rd try\n"; { bool verbose; std::vector<double> f; po::options_description all("Allowed options"); all.add_options() ("verbose,v", po::bool_switch(&verbose)->default_value(false), "verbose") ("floats,f", po::value<std::vector<double> >(&f)->multitoken(), "numbers") ; char const * args[] = { "dd", "--floats", "1", "8", "-4", "6", 0 }; print(args, args + nelements(args) - 1); po::variables_map vm; try { po::store(po::command_line_parser(nelements(args) - 1, args) .options(all).run(), vm); po::notify(vm); if (verbose) std::cout << "I'm verbose\n"; std::cout << "floats: "; boost::copy(f, std::ostream_iterator<double>(std::cout, " ")); std::cout << std::endl; } catch (boost::program_options::unknown_option & x) { std::cerr << x.what() << "\n" << all << "\n"; } } }

6 16

boost::program_options - undocumented "*" feature
by ST 21 Jan '13

21 Jan '13

I recently discovered an undocumented feature of `boost::program_options`, namely that it accepts "*" as a special wildcard that allows declaration of a group of options with the same prefix, like this: configOptions.add_options() ("item_*", value<int>(), "items"); This declaration worked as expected and recognized `item_1`, `item_3`, etc, while rejecting unknown options. Well now the question is, how can I write a custom validate function that will populate a map with options key and its value, like this: map<string, int> itemsMap; options_description items("items options"); items.add_options() ("item_*",value<map<string, int>>(&itemsMap)->multitoken(), "items") ; My question is - how do I get the key of the option being validated from within validate() function? template <typename T> void validate(boost::any& v, const std::vector<std::string>& values, map<string, T> *, int) Thank you, Simone

1 0

Support for windows NTLM Authentication Protocol
by Gheorghe Marinca 21 Jan '13

21 Jan '13

Hi, I use asio and need to connect to a proxy server that might require NTLM Authentication Protocol. Is there a way to use this using boost ? Thanks, -Ghita

1 0

Re: [Boost-users] [MultiIndex] more on key extractor issues
by John M. Dlugosz 21 Jan '13

21 Jan '13

On 1/20/2013 11:00 AM, boost-users-request-at-lists.boost.org |Boost/Allow to home| wrote: > From: Joaquin M Lopez Munoz<joaquin(a)tid.es> > Thanks for your elucidation. I'm much more comfortable with the class now. Like most people in this situation, having found the explanation I'm much to busy trying to employ it in real-world projects and don't have time to edit and expand the docs. But, a pointer to your responses on an archiver might be a great quick addition to the documentation page. It occures to me that if the docs were a Wiki then more might get done, including fixing typos and putting this kind of commentary on the Discussion page. —John

1 0

[MSM] Get state name as string
by Stefan Wiehler 20 Jan '13

20 Jan '13

Hello, For debugging purposes, I want to get the current state name as a string. For this purpose I found the class get_state_name in the tools section of the backend. Unfortunately there is no example on how to use this method in the documentation nor anywhere else; and the extensive use of templates makes it very difficult for me to infer this. I am especially confused about the argument boost::msm::wrap<StateType> const& of the call operator. What do I have to provide her? Could anybody please post a short example how to this? Kind regards, Stefan Wiehler

2 1

Re: [Boost-users] [MultiIndex] more on key extractor issues
by John M. Dlugosz 20 Jan '13

20 Jan '13

> Have you taken a look at the table in: > > http://www.boost.org/libs/multi_index/doc/tutorial/ > key_extraction.html#advanced_key_extractors > > Regardless of whether the indexed-by member is const or not, when using > pointers to const elements the proper way of declaring the extractors is: > > > member< > ItemBase, > const unsigned long, > &ItemBase::some_member > > > > member< > ItemBase, > const S2, > &ItemBase::another_member > > > > This should clear problems away both in VS and Clang (please report > otherwise.) > > Joaqu?n M L?pez Mu?oz > Telef?nica Digital > > I read that part in detail (including the typo, T:gg instead of T::gg), going over all the MI docs again after my postings. I agree this seems to address that use-case, but I really don't understand the table and section, and certainly had no clue what it was getting at when I read it without this issue specifically in mind. "The column "applicable to |const| elements?" states whether the corresponding key extractor can be used when passed constant elements (this relates to the elements specified in the first column, not the referenced |T| objects)." What does that mean? I was actually looking at this in regards to figuring out what a non-const member function extractor was good for. "this also implies that |multi_index_container|s of elements of |T| cannot be sorted by |T::g| or |T::gg|," if you are not using it to sort by, what is the point of having the key extractor? Anyway, I found that it works correctly on both compilers if the middle argument (the key type) is declared const for member and not for const_mem_fun. I suppose that is due to differences in the metaprogramming, but really they should be treated the same. Const on the first parameter, T, doesn't seem to matter if it's declared const or not. ---John

2 1

Re: [Boost-users] [MultiIndex] is ordered_non_unique stable?
by John M. Dlugosz 20 Jan '13

20 Jan '13

> The only problem with this is that the behavior is *undocmented*. If you want > to stay legal, use hinted insertion with end() (which will *documentedly* > work for all the ordered_non_unique indices of your container, not only the one > you happen to be doing for the insertion.) > > Joaqu?n M L?pez Mu?oz > Telef?nica Digital > > > That is very interesting. The wording is kind of strange though: "If succesful, insertion happens as close as possible to the location just prior to |position|." I wonder if stability was the intention? Thank you very much for your analysis of the code. ---John

2 1

Help a novice today getting started with regex.hpp
by Neil Sutton 20 Jan '13

20 Jan '13

Please can someone tell me the best place to start studying how to write and use boost::regex.hpp? Is there any documentation or tutorial that is easier for a novice to understand than the standard distribution documentation?

4 3

[Container] --begin() on empty deque crashes
by TONGARI 20 Jan '13

20 Jan '13

Hi there, The code below would simply crash on g++ 4.7.1/MinGW with default optimization level, it doesn't shown with -O/1/2/3 ---------------------------------------- boost::container::deque<int> d; --d.begin(); ---------------------------------------- I took a look into the impl, seems that it dereferences from (this->m_node - 1) which is invalid at the time. I guess it's the problem. Thoughts?

3 5

[asio] How do you fake the time for boost timers?
by Josh Quigley 20 Jan '13

20 Jan '13

(For those interested this question is also at http://stackoverflow.com/questions/14191855/how-do-you-fake-the-time-for-bo… ) Hi all, If possible, how do you fake the time for the purpose of triggering boost timers in a unit test? For example, is it possible to achieve something like the following: #include <iostream>#include <boost/asio.hpp>#include <boost/date_time/posix_time/posix_time.hpp> void print(const boost::system::error_code& /*e*/){ std::cout << "Hello, world!\n";} int main(){ boost::asio::io_service io; // Possibly another class needed here, or a way of setting the clock to be fake boost::asio::deadline_timer t(io, boost::posix_time::hours(24)); t.async_wait(&print); io.poll(); // Nothing should happen - no handlers ready // PSEUDO-CODE below of what I'd like to happen, jump ahead 24 hours io.set_time(io.get_time() + boost::posix_time::hours(24)); io.poll(); // The timer should go off return 0;} I'm aware I could wrap the io_service and deadline_timer in a wrapper and write my own test implementation, thus providing the real and test version to the production code I'm trying to test. What I'm hoping is the boost framework allows a way to create an io service that is related to a fake clock rather than the system time, meaning I don't need to create my own wrappers. I should also point out that the docs indicate providing something like my own WaitableTimerService orTimerService may be the way to go, but I can't see which one (if any) is responsible for providing the clock. Any tips appreciated! Josh.

2 2

[MultiIndex] more on key extractor issues
by John M. Dlugosz 19 Jan '13

19 Jan '13

I think my problem stems from the use of 'const' in the underlying element with pointer traversal. Consider: typedef multi_index_container< shared_ptr< const ItemBase >, indexed_by< ordered_non_unique< member< const ItemBase, // <<--- here const unsigned long, // <<--- here &ItemBase::some_member > >, // ordered_non_unique ordered_unique< tag<S2>, member< ItemBase, const S2, &ItemBase::another_member > > // ordered_unique > // indexed_by > ItemList_type_base; I added the two 'const' occurrences on the lines marked with "here", and it behaved better. I think it worked with the second index originally because one of those const's was originally there. The member in question was originally const, so I had included const as part of the type there. Basically, what is happening on Windows VS10 is: template<class Class,typename Type,Type Class::*PtrToMember> struct non_const_member_base { typedef Type result_type; template<typename ChainedPtr> #if !defined(BOOST_NO_SFINAE) typename disable_if< is_convertible<const ChainedPtr&,const Class&>,Type&>::type #else Type& #endif operator()(const ChainedPtr& x)const { return operator()(*x); // <<-- HERE } const Type& operator()(const Class& x,int=0)const { return x.*PtrToMember; } Type& operator()(Class& x)const { return x.*PtrToMember; } }; The ChainedPointer form unpeels a const-qualified type with (*x), and that matches the middle form, not the third. But the signature on the first form is looking for Type, which is exactly what I put in the member<...> index directive. So, I suggest that a quick note concerning constant types inside chained pointer following could be added to the tutorial section. Meanwhile, a stranger problem occured on the Mac, using the Clang compiler. I got an error saying that none of the operator() candidates were viable for overloading, with the error indicating that the type of thing being passed in was the shared_ptr<const ItemBase>. It said the first form was ignored because of instantiation failure, but that is exactly contrary to the disable_if. I'm suspecting that the is_convertable itself somehow choked rather than producing false_type...? Naturally none of the other overloads matched, as a shared_ptr<T> cannot be passed where a T is expected, and that's exactly when the first form would be switched in! —John

2 1

[MultiIndex] is ordered_non_unique stable?
by John M. Dlugosz 19 Jan '13

19 Jan '13

Will equivalent elements in a ordered_non_unique index be "stable" and maintain their relative ordering based on the order in which they were inserted? Or do I need to use a complex key, adding my own sequence number, to ensure the desired ordering? —John

2 1

Re: [Boost-users] [serialization] Serializing derived
by Brown, Matt 18 Jan '13

18 Jan '13

Hi Daniel, You may have already considered this option, but what if you were to require a template argument for which the user could pass a serialization behavior class ... much like an allocator for STL containers. I.E. YOUR CODE: template <typename T, typename serialize_T> struct container { void serialize(unsigned version) {serialize_T::do(data, version);} virtual ~container() = default; T data; }; USER'S CODE: struct my_serialize { static void do(SOME_TYPE data, unsigned version) {<user does what's needed here>} }; typedef container<SOME_TYPE, my_serialize> my_container; Maybe of some help to you? -- Matt ________________________________ This e-mail contains privileged and confidential information intended for the use of the addressees named above. If you are not the intended recipient of this e-mail, you are hereby notified that you must not disseminate, copy or take any action in respect of any information contained in it. If you have received this e-mail in error, please notify the sender immediately by e-mail and immediately destroy this e-mail and its attachments.

1 0

[Phoenix][Range] What's wrong here...
by Robert Jones 18 Jan '13

18 Jan '13

Hi All Can anyone tell me what's wrong with this? The error message is so huge I'm a bit overwhelmed! I suspect type deduction is failing, but none of my efforts to correct it have worked. Thx, Rob. #include <boost/phoenix/bind.hpp> #include <boost/phoenix/bind/bind_function_object.hpp> #include <boost/phoenix/core.hpp> #include <boost/phoenix/operator.hpp> #include <boost/range/adaptor/adjacent_filtered.hpp> #include <boost/range/algorithm/for_each.hpp> void f( const double & ) { } template <class Compare> void sorted( Compare cmp ) { using boost::adaptors::adjacent_filtered; using boost::phoenix::arg_names::_1; using boost::phoenix::arg_names::_2; std::vector<double> input = { 0.0, 1.0, 2.0, 3.0, 4.0 }; // Ok for_each( input | adjacent_filtered( cmp ) , f ); // Not Ok for_each( input | adjacent_filtered( bind( cmp, _1, _2 ) ) , f ); } int main( ) { sorted( std::less<double>( ) ); }

4 7

[serialization] Serializing derived template class via base pointer
by Daniel Mitchell 18 Jan '13

18 Jan '13

Hi everyone, has any progress has been made on a general solution for this problem? Given classes like these, struct base { template<typename Archive> void serialize(Archive& ar, unsigned version) { } virtual ~base() = default; }; template<typename T> struct derived : base { template<typename Archive> void serialize(Archive& ar, unsigned version) { ar & data; } T data; }; is it possible to serialize the derived type (with T unknown) via a base pointer? Sticking a call to register_type<derived> in derived::serialize doesn't seem to do it.

3 5

[MultiIndex] strange problem with simple use of member key extractor
by John M. Dlugosz 18 Jan '13

18 Jan '13

namespace bmi = boost::multi_index; .... bmi::ordered_non_unique< bmi::member< ItemBase, unsigned, &ItemBase::my_member > >, // ordered_non_unique .... I'm getting a compiler error on VS10 from template<class Class,typename Type,Type Class::*PtrToMember> struct non_const_member_base { .... Type& operator()(const ChainedPtr& x)const { return operator()(*x); } boost/multi_index/member.hpp(105): error C2440: 'return' : cannot convert from 'const unsigned int' to 'unsigned int &' It also tells me that the ChainedPtr is a shared_ptr to a const ItemBase, and I can see that _none_ of the operator()'s in that class return a non-reference. But if the compiler is already thinking ref-becomes-lvalue and leaves that out, the next form const Type& operator()(const Class& x,int=0)const { return x.*PtrToMember; } is a perfect match to *x and so I expect the return value to be in fact const (because the object being dereferencd was const) Any idea what's going on here, and what I might do about it? Our project is using Boost 1.49, though I didn't see anything in the release notes about changes to MultiIndex in later releases. —John

2 1

[BGL] Limited depth search without O(number_of_verices) complexity?
by Ilja Honkonen 18 Jan '13

18 Jan '13

Hello Is it possible to do a limited depth depth first search without using O(N) additional memory? The following code seems to visit only a part of the graph but it still requires O(N) memory for the external color map. Could that be avoided with an internal color map? I didn't manage to get depth_first_visit(...) to use an internal color map. Would the solution work also with other graph types besides <vecS, vecS, ...>? #include <iostream> #include <boost/graph/adjacency_list.hpp> #include <boost/graph/depth_first_search.hpp> using namespace std; using namespace boost; struct custom_dfs_visitor : public default_dfs_visitor { template < typename Vertex, typename Graph > void discover_vertex(const Vertex& v, const Graph& g) const { cout << v << endl; } }; struct Terminator { template<class Vertex, class Graph> bool operator()(const Vertex& v, const Graph& g) { return v > 2; } }; int main() { typedef adjacency_list< vecS, vecS, undirectedS > Graph_T; Graph_T g(6); add_edge(0, 1, g); add_edge(1, 2, g); add_edge(2, 3, g); add_edge(3, 4, g); add_edge(4, 5, g); vector<default_color_type> color_map(num_vertices(g)); depth_first_visit( g, vertex(0, g), custom_dfs_visitor(), make_iterator_property_map( color_map.begin(), get(vertex_index, g), color_map[0] ), Terminator() ); cout << "size: " << color_map.size() << endl; return 0; } Thanks Ilja

3 6

[MultiIndex] stable? Please help
by John M. Dlugosz 17 Jan '13

17 Jan '13

earlier I posted a question as to whether a non-unique index in MultiIndex is "stable" with regards to insertion order. Can anyone speak to this?

1 0

boost::range RangeConcept problem.
by Hurcan Solter 17 Jan '13

17 Jan '13

The following snippet is the mock-up of the problem I am having in order to be able to reproduce. -------------------------------------------------------------------------------- #include <iostream> #include <boost/range.hpp> #include <boost/range/adaptors.hpp> #include <boost/range/algorithm.hpp> #include <boost/bind.hpp> #include <boost/assign.hpp> using namespace boost::assign; struct wxSize { wxSize(int height=0,int width=0):m_height(height),m_width(width) { } int GetWidth() const {return m_width;} int m_height; int m_width; }; class wxWindowBase { public: wxSize GetTextExtent(const std::string val) { return wxSize(0,val.length()); } void DoLayout(std::vector<std::string> m_FilteredItems) { using boost::adaptors::transformed; wxSize longest = *boost::max_element(m_FilteredItems | transformed(boost::bind(&wxWindowBase::GetTextExtent,this,_1)), boost::bind(std::less<int>(),boost::bind(&wxSize::GetWidth,_1),boost::bind(&wxSize::GetWidth,_2))); std::cout <<longest.GetWidth(); } }; int main() { std::vector<std::string> input; input += "o","two","threexxx","four"; wxWindowBase window; window.DoLayout(input); } ----------------------------------------------------------------------------------------- The above code fails to compile on GCC4.7.2 and MSVC10 although it compiles and runs fine on clang3.3 >From what I understand ,transformed returns a Single Pass Range and max_element expects a ForwardRange and that's the gist of the problem.But it works for something like max_element(someint_vector | transformed(doubleit()),std::less<int>()); without problems. Is there a elegant way to achieve what I am trying to without using a intermediate container? Thanks in advance..

3 3

[thread][move] compiling move(thread) with Android NDK GCC4.7
by Igor R 17 Jan '13

17 Jan '13

Hello, I'm building my project with Android NDK toolchain (Boost1.51, GCC4.6 or GCC4.7), and I get the following error when trying to move thread. Is there any workaround for this issue? //... vector_.push_back(boost::move(thread(&func, arg))); //... try.cpp:86:62: error: no matching function for call to 'move(boost::thread)' try.cpp:86:62: note: candidates are: In file included from boost/thread/detail/move.hpp:20:0, from boost/thread/locks.hpp:11, from boost/thread/pthread/mutex.hpp:12, from boost/thread/mutex.hpp:16, from boost/thread/pthread/thread_data.hpp:13, from boost/thread/thread.hpp:17, from boost/thread.hpp:13, from try.h:3, from try.cpp:6: boost/move/move.hpp:324:97: note: template<class T> typename boost::move_detail::disable_if<boost::has_move_emulation_enabled_aux<T>, T&>::type boost::move(T&) boost/move/move.hpp:324:97: note: template argument deduction/substitution failed: boost/move/move.hpp: In substitution of 'template<class T> typename boost::move_detail::disable_if<boost::has_move_emulation_enabled_aux<T>, T&>::type boost::move(T&) [with T = boost::thread]': try.cpp:86:62: required from here boost/move/move.hpp:324:97: error: no type named 'type' in 'struct boost::move_detail::disable_if<boost::has_move_emulation_enabled_aux<boost::thread>, boost::thread&>' boost/move/move.hpp:330:96: note: template<class T> typename boost::move_detail::enable_if<boost::has_move_emulation_enabled<T>, boost::rv<T>&>::type boost::move(T&) boost/move/move.hpp:330:96: note: template argument deduction/substitution failed: boost/move/move.hpp: In substitution of 'template<class T> typename boost::move_detail::enable_if<boost::has_move_emulation_enabled<T>, boost::rv<T>&>::type boost::move(T&) [with T = boost::thread]': try.cpp:86:62: required from here boost/move/move.hpp:330:96: error: no type named 'type' in 'struct boost::move_detail::enable_if<boost::has_move_emulation_enabled<boost::thread>, boost::rv<boost::thread>&>' boost/move/move.hpp:336:96: note: template<class T> typename boost::move_detail::enable_if<boost::has_move_emulation_enabled<T>, boost::rv<T>&>::type boost::move(boost::rv<T>&) boost/move/move.hpp:336:96: note: template argument deduction/substitution failed: try.cpp:86:62: note: 'boost::thread' is not derived from 'boost::rv<T>' In file included from boost/thread/detail/move.hpp:20:0, from boost/thread/locks.hpp:11, from boost/thread/pthread/mutex.hpp:12, from boost/thread/mutex.hpp:16, from boost/thread/pthread/thread_data.hpp:13, from boost/thread/thread.hpp:17, from boost/thread.hpp:13, from try.h:3, from try.cpp:6: boost/move/move.hpp:943:6: note: template<class I, class O> O boost::move(I, I, O) boost/move/move.hpp:943:6: note: template argument deduction/substitution failed: try.cpp:86:62: note: candidate expects 3 arguments, 1 provided In file included from boost/thread/locks.hpp:11:0, from boost/thread/pthread/mutex.hpp:12, from boost/thread/mutex.hpp:16, from boost/thread/pthread/thread_data.hpp:13, from boost/thread/thread.hpp:17, from boost/thread.hpp:13, from try.h:3, from try.cpp:6: boost/thread/detail/move.hpp:55:124: note: typename boost::enable_if<boost::is_convertible<T&, boost::detail::thread_move_t<T> >, boost::detail::thread_move_t<T> >::type boost::move(T&) [with T = boost::thread; typename boost::enable_if<boost::is_convertible<T&, boost::detail::thread_move_t<T> >, boost::detail::thread_move_t<T> >::type = boost::detail::thread_move_t<boost::thread>] boost/thread/detail/move.hpp:55:124: note: no known conversion for argument 1 from 'boost::thread' to 'boost::thread&' In file included from boost/thread/locks.hpp:11:0, from boost/thread/pthread/mutex.hpp:12, from boost/thread/mutex.hpp:16, from boost/thread/pthread/thread_data.hpp:13, from boost/thread/thread.hpp:17, from boost/thread.hpp:13, from try.h:3, from try.cpp:6: boost/thread/detail/move.hpp:62:37: note: template<class T> boost::detail::thread_move_t<T> boost::move(boost::detail::thread_move_t<T>) boost/thread/detail/move.hpp:62:37: note: template argument deduction/substitution failed: try.cpp:86:62: note: 'boost::thread' is not derived from 'boost::detail::thread_move_t<T>'

3 11

FW: ASIO io_service performance under single or multiple io_services questions
by Bryan Fok 17 Jan '13

17 Jan '13

To asio users, I am confused by the results from my experiments with asio recently. Firstly, am I correct that, when OS I/O buffer is ready to be read, io_service will only pick an available thread from its thread pool to notify the socket's complete handler? For example, if I have a single io_service with 1 thread, and 100 udp sockets executed async_read_from. Hypothetically, if they all simultaneously ready to read from OS kernel buffer, would the io_service notify each socket's complete handler one per time because I only allocated 1 thread to it? Secondly, is it possible to assign workload to different cpu cores through asio? I did an experiment by creating 2 io_services in a program, and each associated with 50 udp sockets for receive same data at the same rate. My theory was that, 2 cores from my 8 cores i7 cpu would be utilize with similar rate, but the result was, only the first core has been fully utilized. Could it be because I/O operation is control by OS only? Thirdly, at what situations should I use multiple threads in a io_service? I increased 50 threads for both io_service in my last experiment setup, and perform the same test. My theory was, when more threads allocated in a io_service , this should help io_service to find a free thread to trigger sockets' complete handler concurrently, therefore, sockets didn't have to wait too long to make the next async read call, and minimize the risk of buffer overflow. The result was completely opposite; the cpu usage has obviously increased, but performance was dramatically decreased because many packet loss was detected. Would it be because of context switching? I thought doesn't matter how many thread were allocated to a io_service, same number of context switching are still required (switching new context from the next complete handler to core to execute after the previous one completed), am I right? B.R Bryan

1 0

[asio][chrono] using steady_timer portably
by Igor R 17 Jan '13

17 Jan '13

Hello, There's some use of asio::steady_timer in my code (1.51): asio::steady_timer timer_; //....... timer_.expires_from_now(boost::chrono::milliseconds(100)); //...... However, the above appears to be non-portable, as under c++11 steady_timer uses std::chrono, while boost::chrono doesn't seem to "forward" its types to std::chrono. I know I can use asio::basic_waitable_timer<boost::chrono::steady_clock> explicitly, but maybe there's some more straightforward solution (eg. boost::chrono configuration)? Thanks.

2 2

Why is boost::format::operator% not declared const?
by Chris Stankevitz 17 Jan '13

17 Jan '13

Hello, I use boost::format to convert a float to a string: void CalledThousandsOfTimesPerSecond(float f) { GetStream() << boost::format("%0.3f") % f << GetSomeOtherContent(); } I normally would deal with something like this by adding a static const variable: void CalledThousandsOfTimesPerSecondFromManyThreads(float f) { static const boost::format Formatter("%0.3f"); GetStream() << Formatter % f << GetSomeOtherContent(); } But this does not work. boost::format::operator% is not declared const. Question 1: Why is boost::format::operator% declared mutable? What is is mutating? Question 2: How do I not spend overhead constructing the same boost::format thousands of times per second? Possible answers to question 2: void CalledThousandsOfTimesPerSecond(float f) { static boost::format Formatter("%0.3f"); // MUTABLE -- this might not be thread safe??? GetStream() << Formatter % f << GetSomeOtherContent(); } void CalledThousandsOfTimesPerSecond(float f) { static const boost::format Formatter1("%0.3f"); boost::format Formatter2(Formatter1); // lame but I guess it works GetStream() << Formatter2 % f << GetSomeOtherContent(); } Thank you, Chris

5 5