- Boost-users - lists.boost.org

Re: [Boost-users] [Boost.Test] Problem with vc9 runtime
by Tan, Tom (Shanghai) 16 Jun '08

16 Jun '08

Greg Ravikovich <grisharav <at> gmail.com> writes: > > > Any workarounds? I see leaks report from Vc9 too, but since I am using Visual Leak Detector(http://dmoulding.googlepages.com/vld) at the same time and it reports no leak. I just ignore the Vc9 warning. I trust VLD because, in a previous release of Boost(1.34?), VC8 reported a leak in boost.thread, which actually was not(the memory was dynamically allocated and not reclaimed intentionally, but not increasing throughout the lifetime, thus not leaking ). Anyway, the author of boost.thread later patched the code to removed the warning. Since the boost.thread was re-written by another writer in 1.35, I amsuspecting this is the same case as before. However, I have no time to check it. Did you use boost.thread in your code?

1 0

Visual C++ 2008 - Regex- Build error
by Fox Fossati 15 Jun '08

15 Jun '08

After reading the start guide i tryied to install the regex from the command prompt of Visual C++ Express: D:\programmi\boost\boost_1_35_0\libs\regex\build>nmake -fvc8.mak now i have a new vc80 directory an then: D:\programmi\boost\boost_1_35_0\libs\regex\build>nmake -fvc8.mak install with this error: Building Boost.Regex without ICU / Unicode support: Hint: set ICU_PATH on the nmake command line to point to your ICU installation if you have one. copy vc80\libboost_regex-vc80-mt-s-1_35.lib "..\..\VC\lib" Impossibile trovare il percorso specificato. 0 file copiati. NMAKE : fatal error U1077: 'copy' : return code '0x1' Stop. The traduction is: impossible finding the specifiyed directory, 0 files copyed. Have i made a mistake? Thanks and regards! ___________________________________ Scopri il Blog di Yahoo! Mail: trucchi, novità, consigli... e la tua opinione! http://www.ymailblogit.com/blog/

1 0

Re: [Boost-users] [boost.thread] thread safe circular_buffer implementation
by Igor R 15 Jun '08

15 Jun '08

> But I can't say the same things for non-const function which returns reference. <...> And also iterators are same as the index operator, not thread safe. It seems that the only way to use such a container in the thread-safe manner, is to perform all the necessary locking at some higher-level of your design. Besides the problem you mentioned, you've got another issue: usually you wish to get not just "physical" thread-safety (meaning that your container remains in a consistent state), but a "logical" one. Consider the following code: if (myVector_.size()) { Element element = myVector_[0]; } It doens't make sense that size() and operator[]() themselves are "thread-safe", as between these calls the size might change. This means that the user of your container has to lock it anyway. If so, your internal locking appears to be useless overhead.

1 0

Error when using the split algorithm
by Sean Farrow 15 Jun '08

15 Jun '08

Hi: I have the following line of code: boost::algorithm::split(SplitString, temp, "-"); I get the following error when trying to compile using visual c++ 2005: error C2064: term does not evaluate to a function taking 1 arguments finder.hpp 583 Can anyone tell me what I have done wrong here. Any help apreciated. Sean. __________ Information from ESET NOD32 Antivirus, version of virus signature database 3187 (20080615) __________ The message was checked by ESET NOD32 Antivirus. http://www.eset.com

1 0

Boost spirit
by Reena Dass 15 Jun '08

15 Jun '08

I'm using boost spirit to create a URI parser. I get a syntax error for the following when I use the grammar straight off from RFC 2396. server = [[userinfo >> ch_p('@')] hostport] ; And when I change the rule and compile it, then the syntax error goes away but I get the following error. I've enclosed Grammar.h for reference Thanks, Reena d:\Work\Mainline\Generation_3\root\usr\include\NTx86debug\boost/spirit/c ore/scanner/scanner.hpp(146) : error C2064: term does not evaluate to a function taking 2 arguments d:\Work\Mainline\Generation_3\root\usr\include\NTx86debug\boost/spirit/c ore/scanner/scanner.hpp(161) : see reference to function template instantiation 'void boost::spirit::attributed_action_policy<boost::spirit::nil_t>::call<Acto rT,IteratorT>(const ActorT &,boost::spirit::nil_t,const IteratorT &,const IteratorT &)' being compiled with [ ActorT=boost::spirit::sequence<boost::spirit::chlit<char>,boost::spirit: :rule<scanner_t>>, IteratorT=iterator_t ] d:\Work\Mainline\Generation_3\root\usr\include\NTx86debug\boost/spirit/c ore/composite/actions.hpp(109) : see reference to function template instantiation 'void boost::spirit::action_policy::do_action<ActionT,boost::spirit::match<boo st::spirit::nil_t>::return_t,iterator_t>(const ActorT &,AttrT &,const IteratorT &,const IteratorT &) const' being compiled with [ ActionT=boost::spirit::sequence<boost::spirit::chlit<char>,boost::spirit ::rule<scanner_t>>, ActorT=boost::spirit::sequence<boost::spirit::chlit<char>,boost::spirit: :rule<scanner_t>>, AttrT=boost::spirit::match<boost::spirit::nil_t>::return_t, IteratorT=iterator_t ] d:\Work\Mainline\Generation_3\root\usr\include\NTx86debug\boost/spirit/c ore/non_terminal/impl/rule.ipp(233) : see reference to function template instantiation 'boost::spirit::match<boost::spirit::nil_t> boost::spirit::action<ParserT,ActionT>::parse<ScannerT>(const ScannerT &) const' being compiled with [ ParserT=boost::spirit::rule<scanner_t>, ActionT=boost::spirit::sequence<boost::spirit::chlit<char>,boost::spirit ::rule<scanner_t>>, ScannerT=boost::spirit::scanner<const char *,boost::spirit::scanner_policies_t> ] d:\Work\Mainline\Generation_3\root\usr\include\NTx86debug\boost/spirit/c ore/non_terminal/impl/rule.ipp(232) : while compiling class template member function 'boost::spirit::match<boost::spirit::nil_t> boost::spirit::impl::concrete_parser<ParserT,ScannerT,AttrT>::do_parse_v irtual(const ScannerT &) const' with [ ParserT=boost::spirit::action<boost::spirit::rule<scanner_t>,boost::spir it::sequence<boost::spirit::chlit<char>,boost::spirit::rule<scanner_t>>> , ScannerT=boost::spirit::scanner<const char *,boost::spirit::scanner_policies_t>, AttrT=boost::spirit::nil_t ] d:\Work\Mainline\Generation_3\root\usr\include\NTx86debug\boost/spirit/c ore/non_terminal/rule.hpp(135) : see reference to class template instantiation 'boost::spirit::impl::concrete_parser<ParserT,ScannerT,AttrT>' being compiled with [ ParserT=boost::spirit::action<boost::spirit::rule<scanner_t>,boost::spir it::sequence<boost::spirit::chlit<char>,boost::spirit::rule<scanner_t>>> , ScannerT=boost::spirit::scanner<const char *,boost::spirit::scanner_policies_t>, AttrT=boost::spirit::nil_t ] d:\work\mainline\generation_3\sources_rdo\urilib\Grammar.h(60) : see reference to function template instantiation 'boost::spirit::rule<T0> &boost::spirit::rule<T0>::operator =<boost::spirit::action<ParserT,ActionT>>(const boost::spirit::action<ParserT,ActionT> &)' being compiled with [ T0=scanner_t, ParserT=boost::spirit::rule<scanner_t>, ActionT=boost::spirit::sequence<boost::spirit::chlit<char>,boost::spirit ::rule<scanner_t>> ] d:\work\mainline\generation_3\sources_rdo\urilib\Grammar.h(29) : while compiling class template member function 'urilib::Grammar::definition<ScannerT>::definition(const urilib::Grammar &)' with [ ScannerT=scanner_t ] d:\Work\Mainline\Generation_3\root\usr\include\NTx86debug\boost/spirit/c ore/non_terminal/impl/grammar.ipp(262) : see reference to class template instantiation 'urilib::Grammar::definition<ScannerT>' being compiled with [ ScannerT=scanner_t ] d:\Work\Mainline\Generation_3\root\usr\include\NTx86debug\boost/spirit/c ore/non_terminal/impl/grammar.ipp(281) : see reference to function template instantiation 'void boost::spirit::impl::call_helper<0>::do_<result_t,definition_t,ScannerT> (RT &,DefinitionT &,const ScannerT &)' being compiled with [ ScannerT=scanner_t, RT=result_t, DefinitionT=definition_t ] d:\Work\Mainline\Generation_3\root\usr\include\NTx86debug\boost/spirit/c ore/non_terminal/grammar.hpp(53) : see reference to function template instantiation 'boost::spirit::match<boost::spirit::nil_t> boost::spirit::impl::grammar_parser_parse<0,urilib::Grammar,boost::spiri t::parser_context<>,ScannerT>(const boost::spirit::grammar<DerivedT> *,const ScannerT &)' being compiled with [ ScannerT=scanner_t, DerivedT=urilib::Grammar ] d:\Work\Mainline\Generation_3\root\usr\include\NTx86debug\boost/spirit/c ore/non_terminal/grammar.hpp(63) : see reference to function template instantiation 'boost::spirit::match<boost::spirit::nil_t> boost::spirit::grammar<DerivedT>::parse_main<ScannerT>(const ScannerT &) const' being compiled with [ DerivedT=urilib::Grammar, ScannerT=scanner_t ] d:\Work\Mainline\Generation_3\root\usr\include\NTx86debug\boost/spirit/c ore/scanner/impl/skipper.ipp(131) : see reference to function template instantiation 'boost::spirit::match<boost::spirit::nil_t> boost::spirit::grammar<DerivedT>::parse<scanner_t>(const ScannerT &) const' being compiled with [ DerivedT=urilib::Grammar, ScannerT=scanner_t ] d:\Work\Mainline\Generation_3\root\usr\include\NTx86debug\boost/spirit/c ore/scanner/impl/skipper.ipp(153) : see reference to function template instantiation 'boost::spirit::parse_info<> boost::spirit::impl::phrase_parser<boost::spirit::space_parser>::parse<I teratorT,DerivedT>(const IteratorT &,const IteratorT &,const ParserT &,const boost::spirit::space_parser &)' being compiled with [ IteratorT=const char *, DerivedT=urilib::Grammar, ParserT=urilib::Grammar ] d:\Work\Mainline\Generation_3\root\usr\include\NTx86debug\boost/spirit/c ore/scanner/impl/skipper.ipp(171) : see reference to function template instantiation 'boost::spirit::parse_info<> boost::spirit::parse<const CharT*,DerivedT,boost::spirit::space_parser>(const IteratorT &,const IteratorT &,const boost::spirit::parser<DerivedT> &,const boost::spirit::parser<boost::spirit::space_parser> &)' being compiled with [ CharT=char, DerivedT=urilib::Grammar, IteratorT=const char * ] .\URI.cpp(23) : see reference to function template instantiation 'boost::spirit::parse_info<> boost::spirit::parse<char,DerivedT,boost::spirit::space_parser>(const CharT *,const boost::spirit::parser<DerivedT> &,const boost::spirit::parser<boost::spirit::space_parser> &)' being compiled with [ DerivedT=urilib::Grammar, CharT=char ] d:\Work\Mainline\Generation_3\root\usr\include\NTx86debug\boost/spirit/c ore/scanner/scanner.hpp(146) : error C2064: term does not evaluate to a function taking 2 arguments d:\Work\Mainline\Generation_3\root\usr\include\NTx86debug\boost/spirit/c ore/scanner/scanner.hpp(161) : see reference to function template instantiation 'void boost::spirit::attributed_action_policy<boost::spirit::nil_t>::call<Acto rT,IteratorT>(const ActorT &,boost::spirit::nil_t,const IteratorT &,const IteratorT &)' being compiled with [ ActorT=boost::spirit::rule<scanner_t>, IteratorT=iterator_t ] d:\Work\Mainline\Generation_3\root\usr\include\NTx86debug\boost/spirit/c ore/composite/actions.hpp(109) : see reference to function template instantiation 'void boost::spirit::action_policy::do_action<ActionT,boost::spirit::match<boo st::spirit::nil_t>::return_t,iterator_t>(const ActorT &,AttrT &,const IteratorT &,const IteratorT &) const' being compiled with [ ActionT=boost::spirit::rule<scanner_t>, ActorT=boost::spirit::rule<scanner_t>, AttrT=boost::spirit::match<boost::spirit::nil_t>::return_t, IteratorT=iterator_t ] d:\Work\Mainline\Generation_3\root\usr\include\NTx86debug\boost/spirit/c ore/non_terminal/impl/rule.ipp(233) : see reference to function template instantiation 'boost::spirit::match<boost::spirit::nil_t> boost::spirit::action<ParserT,ActionT>::parse<ScannerT>(const ScannerT &) const' being compiled with [ ParserT=boost::spirit::rule<scanner_t>, ActionT=boost::spirit::rule<scanner_t>, ScannerT=boost::spirit::scanner<const char *,boost::spirit::scanner_policies_t> ] d:\Work\Mainline\Generation_3\root\usr\include\NTx86debug\boost/spirit/c ore/non_terminal/impl/rule.ipp(232) : while compiling class template member function 'boost::spirit::match<boost::spirit::nil_t> boost::spirit::impl::concrete_parser<ParserT,ScannerT,AttrT>::do_parse_v irtual(const ScannerT &) const' with [ ParserT=boost::spirit::action<boost::spirit::rule<scanner_t>,boost::spir it::rule<scanner_t>>, ScannerT=boost::spirit::scanner<const char *,boost::spirit::scanner_policies_t>, AttrT=boost::spirit::nil_t ] d:\Work\Mainline\Generation_3\root\usr\include\NTx86debug\boost/spirit/c ore/non_terminal/rule.hpp(135) : see reference to class template instantiation 'boost::spirit::impl::concrete_parser<ParserT,ScannerT,AttrT>' being compiled with [ ParserT=boost::spirit::action<boost::spirit::rule<scanner_t>,boost::spir it::rule<scanner_t>>, ScannerT=boost::spirit::scanner<const char *,boost::spirit::scanner_policies_t>, AttrT=boost::spirit::nil_t ] d:\work\mainline\generation_3\sources_rdo\urilib\Grammar.h(68) : see reference to function template instantiation 'boost::spirit::rule<T0> &boost::spirit::rule<T0>::operator =<boost::spirit::action<ParserT,ActionT>>(const boost::spirit::action<ParserT,ActionT> &)' being compiled with [ T0=scanner_t, ParserT=boost::spirit::rule<scanner_t>, ActionT=boost::spirit::rule<scanner_t> ]

2 1

[boost.thread] thread safe circular_buffer implementation
by Selçuk Giray Özdamar 15 Jun '08

15 Jun '08

Hello, I wrote my template based CircularBuffer class. Here the primary aim is to make everything thread safe. Multiple concurrent reads are possible, but only exclusive write is possible. Here we have index operator overloaded. As usual index operator has 2 versions, const and non-const versions. T& operator [] (unsigned index); // return as reference T operator [] (unsigned index) const; // return by value In contrast, standart library container classes return by const reference in const methods. Here I'm returning by value. Below you can see my complete implementation. There may be possible potential risks here. For example, in overloaded index operator (const function) it returns by value, so read lock assures it's safety. But I can't say the same things for non-const function which returns reference. If you write something like this: T& _refT = mCircularBuffer[0]; It returns as reference, and some other time you are free to change the value of _refT (after leaving the index operator scope). So write lock is unlocked, and as a consequence it's not thread safe so. And also iterators are same as the index operator, not thread safe. I'm looking for much safer implementation. Thanks, regards... template <class T> class CircularBuffer { typedef boost::shared_mutex ReadWriteMutex; typedef boost::shared_lock<boost::shared_mutex> ReadLock; typedef boost::unique_lock<boost::shared_mutex> WriteLock; public: typedef typename boost::circular_buffer<T>::const_iterator const_iterator; typedef typename boost::circular_buffer<T>::iterator iterator; explicit CircularBuffer(unsigned capacity = 10) : m_capacity(capacity) , m_buffer(capacity) { } ~CircularBuffer() { } CircularBuffer(const CircularBuffer& other) : m_capacity(other.m_capacity) , m_buffer(other.m_buffer) { } CircularBuffer& operator= (const CircularBuffer& other) { CircularBuffer temp(other); WriteLock w_lock(rw_mutex); using std::swap; swap(m_capacity, temp.m_capacity); swap(m_buffer, temp.m_buffer); return *this; } iterator begin() { WriteLock w_lock(rw_mutex); return m_buffer.begin(); } iterator end() { WriteLock w_lock(rw_mutex); return m_buffer.end(); } const_iterator begin() const { ReadLock r_lock(rw_mutex); return m_buffer.begin(); } const_iterator end() const { ReadLock r_lock(rw_mutex); return m_buffer.end(); } T& operator [] (unsigned index) { WriteLock w_lock(rw_mutex); return m_buffer[index]; } T operator [] (unsigned index) const { ReadLock r_lock(rw_mutex); return m_buffer[index]; } T& at(unsigned index) { WriteLock w_lock(rw_mutex); return m_buffer.at(index); } T at(unsigned index) const { ReadLock r_lock(rw_mutex); return m_buffer.at(index); } T& front() { WriteLock w_lock(rw_mutex); return m_buffer.front(); } T& back() { WriteLock w_lock(rw_mutex); return m_buffer.back(); } T front() const { ReadLock r_lock(rw_mutex); return m_buffer.front(); } T back() const { ReadLock r_lock(rw_mutex); return m_buffer.back(); } size_t size() const { ReadLock r_lock(rw_mutex); return m_buffer.size(); } bool empty() const { ReadLock r_lock(rw_mutex); return m_buffer.empty(); } bool full() const { ReadLock r_lock(rw_mutex); return m_buffer.full(); } size_t capacity() const { ReadLock r_lock(rw_mutex); return m_buffer.capacity(); } void push_back(const T& item = T()) { WriteLock w_lock(rw_mutex); m_buffer.push_back(item); } void clear() { WriteLock w_lock(rw_mutex); m_buffer.clear(); } private: unsigned m_capacity; boost::circular_buffer<T> m_buffer; // rw_mutex for internal buffer mutable ReadWriteMutex rw_mutex; };

1 0

is it possible to initialize thread local storage automatically?
by Frank Birbacher 14 Jun '08

14 Jun '08

Hi! I'm looking for a method to execute some code for each thread starting. It suffices to run that code for each boost::thread. I want to initialize some thread local storage for each thread automatically. The creating(!) thread shall supply an initial value and the created(!) thread shall receive this value, essentially a copy of some thread local variable. That is the new thread shall create a copy of the same variable which the creating thread has. Is that possible using boost::thread? I hope my explaination is clear enough. Frank

2 2

Boost string algorithms split function
by Sean Farrow 14 Jun '08

14 Jun '08

Hi: I am using the split function in the boost string algorithms library. Can I use a vector of char, or do I have to use a vector of std::string. Sean. __________ Information from ESET NOD32 Antivirus, version of virus signature database 3186 (20080613) __________ The message was checked by ESET NOD32 Antivirus. http://www.eset.com

1 0

Nested std::for_each and Boost.Lambda
by Leon Mergen 14 Jun '08

14 Jun '08

Hello, I'm trying to convert the following code to a more functional counterpart using std::for_each: BOOST_FOREACH ( std::vector <keyword_matchable *>::value_type i, haystack->_children ) { std::for_each (needle->_children.begin (), needle->_children.end (), boost::lambda::bind (&keyword_matchable::_match_new, this, i, boost::lambda::_1)); } As you can see, basically this is a nested foreach loop: for each haystack->_children, we're looping for each needle->_children, and calling this->_match_new (haystack_child, needle_child). Now, while code works, as a sort of personal challenge I tried rewriting this to a purely std::for_each () based approach; however, I don't quite understand how I am supposed to use the parent loop's _1 inside a nested loop; consider this code: std::for_each(haystack->_children.begin (), haystack->_children.end (), boost::lambda::bind( boost::lambda::ll::for_each(), needle->_children.begin (), needle->_children.end (), boost::lambda::protect ( boost::lambda::bind ( &keyword_matchable::_match_new, this, boost::lambda::_1, boost::lambda::_1)))); As you can see, i have used the boost::lambda::_1 argument twice; logically, the same argument is provided twice to that function. Now, I've read something about using boost::lambda::protect to "prevent argument substitution to take place", however, I fail to understand how this exactly works. Anyone has any idea how I can call the function keyword_matchable::_match_new with both the current nesting level _1 and the parent nesting level _1 as arguments ? Thanks in advance! Regards, Leon Mergen

1 0

Boost RegEx for C++
by Per Franck 14 Jun '08

14 Jun '08

Hello All. I'm using the C++ library of boost RegEx. Long story short: I'm trying the expression (?<=<h1>).*?(?=</h1>) On the string <h1>hello there</h1> and the engine crashes (asserts in Kernel32(d).dll) Is this a known issue? here's my code snippet. std::vector<std::string> v; boost::RegEx expr( "<h1>hello there</h1>", TRUE); expr.Grep(v, "(?<=<h1>).*?(?=</h1>)" ); thanks for any help! - Per

3 2