May 2013 - Boost-users - lists.preview.boost.org

[msm] Manually setting next state
by samriti katoch 22 May '13

22 May '13

Thanks for the reply Christophe. I had considered this alternative, but adding a transition in every state to the initial state, when Error -> Ok defeats the purpose of having an orthogonal region, which I added to basically reduce transitions on an Error. -Samriti

1 0

Compile errors including boost/regex.hpp
by Jordan H. 22 May '13

22 May '13

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA512 When I try to compile a file that includes <boost/regex.hpp> I get the following error: # /home/me/Projects/myproject> make # /usr/bin/make all-am # make[1]: Entering directory `/home/me/Projects/myproject' # compiling main.cpp (g++) # In file included from /usr/local/include/boost/regex/regex_traits.hpp:27:0, # from /usr/local/include/boost/regex/v4/regex.hpp:39, # from /usr/local/include/boost/regex.hpp:31, # [ ... more traceback ... ] # # /usr/local/include/boost/regex/v4/regex_traits.hpp:87:1: error: macro "test" passed 2 arguments, but # takes just 1 # In file included from /usr/local/include/boost/regex/v4/regex.hpp:79:0, # from /usr/local/include/boost/regex.hpp:31, # [ ... more traceback ... ] # # /usr/local/include/boost/regex/v4/regex_format.hpp:849:1: error: macro "test" passed 2 arguments, but # takes just 1 # In file included from /usr/local/include/boost/regex/regex_traits.hpp:27:0, # from /usr/local/include/boost/regex/v4/regex.hpp:39, # from /usr/local/include/boost/regex.hpp:31, # [ ... more traceback ... ] # # /usr/local/include/boost/regex/v4/regex_traits.hpp:87:1: error: template declaration of ‘char (& # boost::re_detail::test)[2]’ # /usr/local/include/boost/regex/v4/regex_traits.hpp:87:1: error: expected unqualified-id before ‘...’ # token # /usr/local/include/boost/regex/v4/regex_traits.hpp:87:1: error: expected ‘)’ before ‘...’ token # /usr/local/include/boost/regex/v4/regex_traits.hpp:87:1: error: expected ‘)’ before ‘...’ token # /usr/local/include/boost/regex/v4/regex_traits.hpp:87:1: error: expected unqualified-id before ‘(’ # token # In file included from /usr/local/include/boost/regex/v4/regex.hpp:79:0, # from /usr/local/include/boost/regex.hpp:31, # [ ... more traceback ... ] # # /usr/local/include/boost/regex/v4/regex_format.hpp:849:1: error: template declaration of ‘char (& # boost::re_detail::test)[2]’ # /usr/local/include/boost/regex/v4/regex_format.hpp:849:1: error: expected unqualified-id before ‘...’ # token # /usr/local/include/boost/regex/v4/regex_format.hpp:849:1: error: expected ‘)’ before ‘...’ token # /usr/local/include/boost/regex/v4/regex_format.hpp:849:1: error: expected ‘)’ before ‘...’ token # /usr/local/include/boost/regex/v4/regex_format.hpp:849:1: error: expected unqualified-id before ‘(’ # token # make[1]: *** [myproject-main.o] Error 1 # make[1]: Leaving directory `/home/me/Projects/myproject' # make: *** [all] Error 2 # *** Failure: Exit code 2 *** This is how I'm compiling: # g++ -DHAVE_CONFIG_H -I. `mysql_config --include` `mysql_config --cflags` -std=c++0x -g -O2 -MT # myproject-main.o -MD -MP -MF .deps/myproject-main.Tpo -c -o myproject-main.o `test -f 'src/main.cpp' # || echo './'`src/main.cpp Basically, the compile errors trace back to when I include boost/regex.hpp: | #include <boost/regex.hpp> Any ideas why this occurs? Maybe it's because I'm not including -lbooost_regex as a compile option? I specified it as my "myproject_LDADD" macro in Makefile.am so I assume g++ knows when to use it. Thanks. Sorry for the lengthy email. -----BEGIN PGP SIGNATURE----- iQEcBAEBCgAGBQJRnN/nAAoJEOT99oqiSuRNim0H/1U+xgx+TCwOtDiGfC5FH6yR 1HwZYwTb0F/7CwldOBsPoDMfjSJmptH4a4s84PR0xDHFrhvgDkjDvzYkoWPLR7gf eSAMHzQG3kvJawvHiJNSu/eTZivVfkCE14+zoE+cVriCJicJUJg1g8hp5Cydk6D3 OM9WtOY/GCVHD8lK84UX5HMqnjeX4nPLEA7nJGd1u2+5PkHxpQMF2RO+EWnlNm0e eTUdZBlGLc/RGHMgl2uF1q3b+qCRrC/Wwn/4OtkDUn0Y3V879LbkYej06oB94zWD 9kKHAnTFOisQJDUmkHr9BiEGgJ9UNO2JCVkHlzrjVgfwKe1r6FIaZmTGNpIme0c= =Av+g -----END PGP SIGNATURE-----

3 6

[msm] Manually setting next state
by samriti katoch 22 May '13

22 May '13

Hi, Is there a way to manually set the next state for the state machine. I have an orthogonal region defined for handling errors, where Error is an interruptible state. ---------- ------------- | | ----------->| Error | | Ok | | | ---------- ------------- This is similar to the example here : http://svn.boost.org/svn/boost/trunk/libs/msm/doc/HTML/examples/Orthogonal-… Once the state machine recovers from the error, I would like to start from the initial state again in the main orthogonal region. Is there a way to do this ? Thanks! Samriti

2 1

[atomic] When to load/store
by Michael Powell 22 May '13

22 May '13

Hello, I have some thread synchronization to do, probably will end up using some sort of internally synchronized cancelation or other state to message between parent and child thread. If the state uses boost::atomic<bool> _canceled, for instance, how do I go about setting/getting the value? I am assuming that we do something like this: //To get the value. return _canceled.load(boost::memory_order::memory_order_acq_rel); //To set the value. _canceled.store(canceled, boost::memory_order::memory_order_acq_rel); Using acquire/release to get in, do what we need with the value, and get out, type thing. Also bearing in mind this is targeting ARM architecture (ARM5 I think). Also, how complex a value can be protected? Are we talking primitive types only? Basic data types like bool, int, float, that sort of thing? Thank you... Regards, Michael Powell

2 1

[Atomic] / PPC / error: impossible constraint in 'asm'
by nicolas.danet＠free.fr 22 May '13

22 May '13

Hello, I get the error below compiling with (Xcode 2.5 / OS X 10.4.11) only for PPC arch. boost_1_53_0/boost/atomic/detail/gcc-ppc.hpp:1204: error: impossible constraint in 'asm' The related code is: static boost::atomic<long> messageCount_(0); number_ = messageCount_.fetch_add(1, boost::memory_order_relaxed); Is this a known bug ? Is there any way to fix it ? Thanks.

1 0

[accumulator] with complex domain model
by Michael Powell 21 May '13

21 May '13

Hello, We've got a complex domain model which must be accumulated and eventually averaged together. I am already overloading the operators for that model. We could roll our own with just the operators we'll need to accommodate, like +, +=, /, etc, but would like to take a gander at Boost.Accumulator first. Would be much better, obviously, rather than having to reinvent the wheel. How doable would it be to accumulate an arbitrarily complex domain model. That's assuming that it is sensible to add, average, and so on, the model, which it is. Regards, Michael Powell

2 2

endless loop and/or crash in boost::interprocess
by Charanga 21 May '13

21 May '13

How do I debug a potential endless loop and heap corruption issue involving boost::interprocess "new managed_shared_memory" freezing forever or crashing? I posted this on stackoverflow but I am hopeful that a real expert in Boost::Interprocess may have seen this freeze-up and crash before. Outside the debugger it manifests as a freeze-up, inside the debugger it manifests as a series of exceptions inside an endless try-catch-ignore loop which is coded into the boost implementation code. http://stackoverflow.com/questions/16651878/how-do-i-debug-a-potential-endl… Warren

2 1

[SmartPtr] enable_shared_from_raw / weak_from_raw remains not associated
by Slava 20 May '13

20 May '13

I'm converting existing code with parent-child object tree to boost shared_ptr/weak_ptr with parents having [now shared] ownership over the children and children having weak reference to the parent. In the existing code a parent usually becomes the children as constructor parameters, and thus I have to extract the weak_ptr to "this" in constructor, which does not work with current boost::enable_shared_from_this implementation. So I am looking to enable_shared_from_raw alternative, which is promising, but unfortunately does not quite work for my case. Here is a small example demonstrating the issue: test.cpp: #include <boost/smart_ptr/enable_shared_from_raw.hpp> #include <iostream> struct Parent; struct Child { Child() { std::cout << "Construct Child" << std::endl; } ~Child() { std::cout << "Destruct Child" << std::endl; } boost::weak_ptr<Parent> parent; }; struct Parent : public boost::enable_shared_from_raw { Parent(boost::shared_ptr<Child> _child) { std::cout << "Construct Parent" << std::endl; this->child = _child; this->child->parent = boost::weak_from_raw(this); } ~Parent() { std::cout << "Destruct Parent" << std::endl; } boost::shared_ptr<Child> child; }; int main() { boost::shared_ptr<Child> child(new Child()); boost::shared_ptr<Parent> parent(new Parent(child)); std::cout << "child->parent is " << (child->parent.lock() ? "not null" : "null") << std::endl; parent.reset(); std::cout << "child must be alone" << std::endl; std::cout << "child->parent is " << (child->parent.lock() ? "not null" : "null") << std::endl; return 0; } `g++ -g -O0 test.cpp && valgrind a.out` produces: ==8467== Memcheck, a memory error detector ==8467== Copyright (C) 2002-2012, and GNU GPL'd, by Julian Seward et al. ==8467== Using Valgrind-3.8.1 and LibVEX; rerun with -h for copyright info ==8467== Command: a.out ==8467== Construct Child Construct Parent child->parent is null Destruct Parent child must be alone child->parent is null Destruct Child ==8467== ==8467== HEAP SUMMARY: ==8467== in use at exit: 0 bytes in 0 blocks ==8467== total heap usage: 4 allocs, 4 frees, 112 bytes allocated ==8467== ==8467== All heap blocks were freed -- no leaks are possible ==8467== ==8467== For counts of detected and suppressed errors, rerun with: -v ==8467== ERROR SUMMARY: 0 errors from 0 contexts (suppressed: 2 from 2) The problem is that the weak_ptr to parent in child does not get associated properly, when shared_ptr to parent is finally complete. Do I expect too much from enable_shared_from_raw or did I hit some bug in the implementation? I'm using boost 1.53.0-2 on Archlinux. -- Slava

2 3

[boost][thread] Future's scheduler
by Klaim - Joël Lamotte 20 May '13

20 May '13

I'm using boost.thread V4 in my projects because I would like to use futures. I understand that boost 1.53 implementation of future.then() is incomplete but I think it's ok for my current use. However I think there might be something missing. I'm in a case where the user of my library provide his own task scheduler (or executor) which means that when I have functions that return futures, they will use that scheduler. I would like to limit the boilerplate necessary for the user to schedule tasks properly when one task is finished, so I assume the user will use then() with just lambdas or function pointers. The documentation of then() says: - When the scheduler or launch policy is not provided the continuation inherits the parent's launch policy or scheduler. Which is perfect to me as I just have to associate the first future with the scheduler, the user that will use .then() will not have to specify which scheduler to use. My problem is that I can't find how to associate a scheduler to a future (without using .then()). There is apparently no way to do this in boost::future or boost::promise interfaces. Am I missing something? I see that I could use a dummy initial future just to build another future with .then() which would associate the second future with the scheduler, but this is clearly just ductapping. Joel Lamotte

1 1

64 bit or 32 bit?
by Scott Alexander 20 May '13

20 May '13

Hi, I have a Windows 8 64 bit OS and would like to use Visual Studio Express 2012 for Windows 8. Should I install a 64 bit or 32 bit version of Boost? Thanks, Scott

4 4