April 2002 - Boost-users - lists.boost.org

RE: Problem Compiling Boost Graph Library User Guide Code
by Wayne Hartell 13 Apr '02

13 Apr '02

Hi again, I managed to get past the first VC7 error by adding the following line to graph/properties.hpp which appears like it was an omission... #ifdef BOOST_NO_TEMPLATE_PARTIAL_SPECIALIZATION template <> struct property_traits<default_color_type*> { typedef default_color_type value_type; typedef value_type& reference; // Newly Added typedef std::ptrdiff_t key_type; typedef lvalue_property_map_tag category; }; but now I run into yet another compiler error... \boost_1_27_0\boost\concept_check.hpp(185): fatal error C1001: INTERNAL COMPILER ERROR (compiler file 'msc1.cpp', line 1414) Please choose the Technical Support command on the Visual C++ Help menu, or open the Technical Support help file for more information template <class TT> struct DefaultConstructibleConcept { void constraints() { TT a; // require default constructor ignore_unused_variable_warning(a); } }; template <class TT> struct AssignableConcept { void constraints() { #if !defined(_ITERATOR_) // back_insert_iterator broken for VC++ STL a = a; // require assignment operator #endif const_constraints(a); } void const_constraints(const TT& b) { // C1001 INTERNAL COMPILER ERROR #if !defined(_ITERATOR_) // back_insert_iterator broken for VC++ STL a = b; // const required for argument to assignment #endif } TT a; }; Any help would be greatly appreciated... Kind Regards Wayne Hartell -----Original Message----- From: Wayne Hartell Sent: Friday, April 12, 2002 5:54 PM To: 'Boost-Users(a)yahoogroups.com' Subject: Problem Compiling Boost Graph Library User Guide Code Hi, I am running into the following compile error when using VC7 to compile the BGL text book example on Edge Connectivity (Chap 8.2) boost_1_27_0\boost\property_map.hpp(151): error C2039: 'reference' : is not a member of 'boost::property_traits<PA>' with [ PA=boost::default_color_type * ] at line : typedef typename property_traits<PMap>::reference reference; My first step was to try to compile this using VC6 instead, but in that case I run into the following error : boost_1_27_0\boost\graph\detail\adjacency_list.hpp(1830) : fatal error C1001: INTERNAL COMPILER ERROR (compiler file 'msc1.cpp', line 1794) Please choose the Technical Support command on the Visual C++ Help menu, or open the Technical Support help file for more information at line : inline vec_adj_list_impl(vertices_size_type _num_vertices) : m_vertices(_num_vertices) { } The last error is coincidentally the same type of compiler error that I just experienced in VC7 to which I recently posted a work around for. I would be extremely grateful for any suggestions for either of these problems... Kind Regards, Wayne Hartell [Non-text portions of this message have been removed]

1 0

Problem Compiling Boost Graph Library User Guide Code
by Wayne Hartell 12 Apr '02

12 Apr '02

Hi, I am running into the following compile error when using VC7 to compile the BGL text book example on Edge Connectivity (Chap 8.2) boost_1_27_0\boost\property_map.hpp(151): error C2039: 'reference' : is not a member of 'boost::property_traits<PA>' with [ PA=boost::default_color_type * ] at line : typedef typename property_traits<PMap>::reference reference; My first step was to try to compile this using VC6 instead, but in that case I run into the following error : boost_1_27_0\boost\graph\detail\adjacency_list.hpp(1830) : fatal error C1001: INTERNAL COMPILER ERROR (compiler file 'msc1.cpp', line 1794) Please choose the Technical Support command on the Visual C++ Help menu, or open the Technical Support help file for more information at line : inline vec_adj_list_impl(vertices_size_type _num_vertices) : m_vertices(_num_vertices) { } The last error is coincidentally the same type of compiler error that I just experienced in VC7 to which I recently posted a work around for. I would be extremely grateful for any suggestions for either of these problems... Kind Regards, Wayne Hartell [Non-text portions of this message have been removed]

1 0

Regexp: std::runtime_error, bad pattern?
by vemundha 12 Apr '02

12 Apr '02

Hello, Can anybody tell me what's the problem with this warning: --- non dll-interface class 'std::runtime_error' used base for dll-interface class 'boost::pattern' --- _ Vemund Halvorsen, vemund.halvorsen(a)kongsberg.com Software Engineer RO CCIT, Kongsberg Defence and Aerospace As PO Box 1003 N-3601 Kongsberg

2 1

Error in bind1st.cpp example for boost:::function
by rnicz 12 Apr '02

12 Apr '02

There is bug in bind1st.cpp example for boost::function library. I'm using VC6SP5 and when I was trying to compile this example I got error === bind1st.cpp D:\src\boost\1.27\libs\function\example\bind1st.cpp(34) : error C2784: 'class std::mem_fun_t<_R,_Ty> __cdecl std::mem_fun(_R (__thiscall _Ty::*)(void))' : could not deduce template argument for '<Unknown>' from 'int (__thiscall X::*)(int)' Error executing cl.exe. === My first thought was that VC6 not enought standard conformant to compile it. But after some inquiry I found that line 35 f = std::bind1st(std::mem_fun(&X::foo), &x); should be f = std::bind1st(std::mem_fun1(&X::foo), &x); May be this is quite obvious, but examples shouldn't have errors. Best regards RNicz

1 0

Compiling Boost in VC++ .Net
by Wayne Hartell 12 Apr '02

12 Apr '02

Hello Boost Users, I am a newbie to the Boost Graph library and am trying to come up to speed with the use of the library as quickly as I can. Last week I thought that one of the best ways to do this would be to work through some of the example code in the Boost User Guide and Reference Manual but I was disappointed to find out that the moment I included some boost header files in my VC++ .Net project that I ran into a C1001 FATAL COMPILER ERROR. This problem doesn't occur in VC6 nor does it occur in the Borland 5.5 C++ compiler, however, my current project dictates that I use the VC++ .Net compiler. Before jumping on to this list with a posting I searched the archive and noticed that a previous posting had been lodged with the same compiler error. To cut a long story short I have been in touch with Microsoft Technical support (who were extremely helpful) and they have confirmed my issue (and that of the previous posting) to be a bug in the VC++ .Net compiler. Microsoft also gave me a work around for the problem so the main purpose of this posting is to make public the work around in order to use the BGL with the VC++ .Net compiler. As I am new to the BGL if anyone has any comments on the validity of the work around provided by Microsoft I would be grateful to hear them. I have tried compiling the code and it now compiles, however, as I am new to the library I am not yet in a position where I have tried out the library in a project to verify that it besides compiling correctly, that it also works correctly ;0) Here's the work around: composite_traits.hpp template <typename T> // line 149 struct is_array { private: static T t; public: #if _MSC_VER == 1300 // VC.Net static const bool value; #else // This is the original code block that causes the compiler error BOOST_STATIC_CONSTANT(bool, value = (::boost::type_traits::ice_and< (1 == sizeof(detail::is_array_helper(&t, t))), ::boost::type_traits::ice_not< ::boost::detail::is_reference_or_const_volatile<T>::value>::value, ::boost::type_traits::ice_not< (1 == sizeof(detail::is_function_tester(t)))>::value >::value)); #endif }; #if _MSC_VER == 1300 // VC.Net template <typename T> const bool is_array<T>::value = (::boost::type_traits::ice_and< (1 == sizeof(detail::is_array_helper(&t, t))), ::boost::type_traits::ice_not< ::boost::detail::is_reference_or_const_volatile<T>::value>::value, ::boost::type_traits::ice_not< (1 == sizeof(detail::is_function_tester(t)))>::value >::value)); #endif Kind Regards, Wayne Hartell [Non-text portions of this message have been removed]

1 0

Re: [Boost-Users] Re: Re: Newbie needs pointer to currently executing thread.
by dick.bridges＠tais.com 11 Apr '02

11 Apr '02

Thanks for the response. Because this pattern involves a condition.wait() followed by select(), I believe that only the leader should be able to set the predicate to NO_CURRENT_LEADER. Otherwise there is the [very unlikely] danger that a spurious wake up on the condition could result a second thread attempting to process the same handle/event before the first thread could deactivate it. Although I can wait until all the housekeeping for the select() has been done before unguarding the condition and continue with processing the event, that's a little more synchronous than I care for given that a simple comparison against a thread id allows me to unguard the condition much earlier. Actually, for now, that's EXACTLY what I'm going to do. Probably good enough anyway. ;) Thanks again for the response and for the info on Bill Kempf's intent to provide portable thread_id s. "Moore, Dave" <dmoore@ashtontec To: "'Boost-Users(a)yahoogroups.com'" <Boost-Users(a)yahoogroups.com> hgroup.com> cc: Subject: [Boost-Users] Re: Re: Newbie needs pointer to currently executing thread. 04/11/2002 12:48 PM Please respond to Boost-Users <dick.bridges(a)tais.com> wrote in message news:OFAB9F41E0.6C6A5A14-ON88256B97.005045D8@tais.net... > > I'm implementing the Leader/Followers thread pool pattern - reference > "Pattern-Oriented Software Architecture Volume 2" (Schmidt, et al). > > When a thread is released via notify_one(), it is [assumed to be] in the > "leader" role and stores it's id. A sanity check should be performed to > verify that there is no other thread in the leader role (i.e., > current_leader is "none") before the thread stores it's id there. > > When one of the event sources is signaled and the leader is unblocked, it > must, among other tasks, demote itself (i.e., set the current_leader value > to "none") > and promote one of the followers by notify_one() before continuing to > service the request. Because of the potential for race conditions, > (especially in the context of future program updates, modifications, etc.) > I should enforce the rule that only the leader can promote a follower. > Something along the lines of: current_leader == Thread::this(). > I disagree (about the race condition). Your condition variable is usually used in conjunction with a mutex which protects your shared state's invariants. boost::mutex m; boost::condition c; Invariant: leader != NONE Pseudo-code for leader releasing boost::mutex::scoped_lock l(m); // lock mutex if(TimeForANewLeader) { leader=NONE; c.notify_one(); // Will signal one waiter..... // finish task - MUTEX IS STILL HELD! // } l.unlock(); // Only now will a waiter return from wait(); Pseudeo-code for follower waiting for signal: boost::mutex::scoped_lock l(m); // lock mutex while(Not TimeForANewLeader) { c.wait(m); // Waiting automically releases AND reacquires // the mutex } // When we get here, we know we hold the mutex. // Follower is responsible for "self-promotion" leader=ME; // Start my work as a leader. An important note is that if we reach the while() statement and it IS time for a new leader, then we never wait, we already hold the mutex, and it is safe for us to promote ourselves anyways! Note that for the time being, you'll have to go outside of Boost.Threads for the leader=ME statement because thread objects aren't copyable so leader=boost::thread(); doesn't work to get the current thread. Bill Kempf has indicated that some form of portable thread_id s are coming in a future version. Hope this helps... Dave > Info: <http://www.boost.org> Wiki: <http://www.crystalclearsoftware.com/cgi-bin/boost_wiki/wiki.pl> Unsubscribe: <mailto:boost-users-unsubscribe@yahoogroups.com> Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/

1 0

Re: Re: Newbie needs pointer to currently executing thread.
by Moore, Dave 11 Apr '02

11 Apr '02

<dick.bridges(a)tais.com> wrote in message news:OFAB9F41E0.6C6A5A14-ON88256B97.005045D8@tais.net... > > I'm implementing the Leader/Followers thread pool pattern - reference > "Pattern-Oriented Software Architecture Volume 2" (Schmidt, et al). > > When a thread is released via notify_one(), it is [assumed to be] in the > "leader" role and stores it's id. A sanity check should be performed to > verify that there is no other thread in the leader role (i.e., > current_leader is "none") before the thread stores it's id there. > > When one of the event sources is signaled and the leader is unblocked, it > must, among other tasks, demote itself (i.e., set the current_leader value > to "none") > and promote one of the followers by notify_one() before continuing to > service the request. Because of the potential for race conditions, > (especially in the context of future program updates, modifications, etc.) > I should enforce the rule that only the leader can promote a follower. > Something along the lines of: current_leader == Thread::this(). > I disagree (about the race condition). Your condition variable is usually used in conjunction with a mutex which protects your shared state's invariants. boost::mutex m; boost::condition c; Invariant: leader != NONE Pseudo-code for leader releasing boost::mutex::scoped_lock l(m); // lock mutex if(TimeForANewLeader) { leader=NONE; c.notify_one(); // Will signal one waiter..... // finish task - MUTEX IS STILL HELD! // } l.unlock(); // Only now will a waiter return from wait(); Pseudeo-code for follower waiting for signal: boost::mutex::scoped_lock l(m); // lock mutex while(Not TimeForANewLeader) { c.wait(m); // Waiting automically releases AND reacquires // the mutex } // When we get here, we know we hold the mutex. // Follower is responsible for "self-promotion" leader=ME; // Start my work as a leader. An important note is that if we reach the while() statement and it IS time for a new leader, then we never wait, we already hold the mutex, and it is safe for us to promote ourselves anyways! Note that for the time being, you'll have to go outside of Boost.Threads for the leader=ME statement because thread objects aren't copyable so leader=boost::thread(); doesn't work to get the current thread. Bill Kempf has indicated that some form of portable thread_id s are coming in a future version. Hope this helps... Dave >

1 0

Streamed input for rational, badbit vs failbit
by aj_dicks 10 Apr '02

10 Apr '02

Is there an error in the declaration of template <typename IntType> std::istream& operator>> (std::istream& is, rational<IntType>& r) in boost.hpp? The lines if (c != '/') is.clear(std::istream::badbit); appear in the function definition, as the format is num/den, so we expect a slash character. However, is setting the stream to bad correct? Shouldn't it be failbit instead? The standard states that badbit indicates a loss of integrity in an input or output sequence (such as an irrecoverable read error from a file); failbit indicates that an input operation failed to read the expected characters, or that an output operation failed to generate the desired characters. Also, to cite precedence from the standard, it stipulates that the std::complex type's extraction operator should call is.setstate(ios::failbit) if bad input is found. What is confusing, however, is that Stroustrup's example input operation for complex sets badbit (section 21.3.5 in the third edition of his bible). So who is right out of: me, rational.hpp, the C++ standard, Stroustrup? Is it more than two of these, and if so, where is my misunderstanding?

1 0

newbie q: - make errors on Solaris (testing boost/regex)
by dz_nyc11354 10 Apr '02

10 Apr '02

I think I'm not linking the boost libary. But I checked many times and can't find any error. The test file I'm using is John Maddock's regex_split_example_1.cpp. I just renamed it to test.cpp, I also have a Makefile that look like this: *************************************************** CC = gcc MY_DIR = /home/dzhou/boost_1_27_0 MY_LIB = -L$(MY_DIR)/libs/regex/build/gcc MY_INCLUDE = -I$(MY_DIR) CFLAGS = -g -v -D_REENTRANT -Wall $(MY_INCLUDE) LIBS = $(MY_LIB)/libboost_regex.a LIBS_RPC = -lnsl RANLIB = ranlib RPCGEN_OPTS = -C .cc.o: $(CC) $(CFLAGS) -c $< PROGS = test all: ${PROGS} test: test.o ${CC} ${CFLAGS} -o test test.o ${LIBS} **************************************************** I've build the libary with make -f gcc.mak and it made the gcc dir I've tryed to play around with gcc.mak by changing g++ to gcc because I was getting errors with g++ also. #complier: CXX=gcc LINKER=gcc -shared here is the error that I'm getting: *********************************************************** gcc -g -v -D_REENTRANT -Wall -I/home/dzhou/boost_1_27_0 -c test.cc Reading specs from /usr/local/lib/gcc-lib/sparc-sun-solaris2.7/2.95.3/specs gcc version 2.95.3 20010315 (release) /usr/local/lib/gcc-lib/sparc-sun-solaris2.7/2.95.3/cpp0 -lang-c++ -v -I/home/dzhou/boost_1_27_0 -D__GNUC__=2 -D__GNUG__=2 -D__GNUC_MINOR__=95 -D__cplusplus -Dsparc -Dsun -Dunix -D__svr4__ -D__SVR4 -D__sparc__ -D__sun__ -D__unix__ -D__svr4__ -D__SVR4 -D__sparc -D__sun -D__unix -Asystem(unix) -Asystem(svr4) -D__EXCEPTIONS -g -Wall -D__GCC_NEW_VARARGS__ -Acpu(sparc) -Amachine(sparc) -D_REENTRANT test.cc /var/tmp/ccmuQP2y.ii GNU CPP version 2.95.3 20010315 (release) (sparc) #include "..." search starts here: #include <...> search starts here: /home/dzhou/boost_1_27_0 /usr/local/lib/gcc-lib/sparc-sun-solaris2.7/2.95.3/../../../../include/g++-3 /usr/local/include /usr/local/lib/gcc-lib/sparc-sun-solaris2.7/2.95.3/../../../../sparc-sun-solaris2.7/include /usr/local/lib/gcc-lib/sparc-sun-solaris2.7/2.95.3/include /usr/include End of search list. The following default directories have been omitted from the search path: End of omitted list. /usr/local/lib/gcc-lib/sparc-sun-solaris2.7/2.95.3/cc1plus /var/tmp/ccmuQP2y.ii -quiet -dumpbase test.cc -g -Wall -version -o /var/tmp/ccW67QRe.s GNU C++ version 2.95.3 20010315 (release) (sparc-sun-solaris2.7) compiled by GNU C version 2.95.2 19991024 (release). /usr/ccs/bin/as -V -Qy -s -o test.o /var/tmp/ccW67QRe.s /usr/ccs/bin/as: WorkShop Compilers 5.0 98/12/21 gcc -g -v -D_REENTRANT -Wall -I/home/dzhou/boost_1_27_0 -o test test.o -L/home/dzhou/boost_1_27_0/libs/regex/build/gcc/libboost_regex.a Reading specs from /usr/local/lib/gcc-lib/sparc-sun-solaris2.7/2.95.3/specs gcc version 2.95.3 20010315 (release) /usr/local/lib/gcc-lib/sparc-sun-solaris2.7/2.95.3/collect2 -V -Y P,/usr/ccs/lib:/usr/lib -Qy -o test /usr/local/lib/gcc-lib/sparc-sun-solaris2.7/2.95.3/crt1.o /usr/local/lib/gcc-lib/sparc-sun-solaris2.7/2.95.3/crti.o /usr/ccs/lib/values-Xa.o /usr/local/lib/gcc-lib/sparc-sun-solaris2.7/2.95.3/crtbegin.o -L/home/dzhou/boost_1_27_0/libs/regex/build/gcc/libboost_regex.a -L/usr/local/lib/gcc-lib/sparc-sun-solaris2.7/2.95.3 -L/usr/ccs/bin -L/usr/ccs/lib -L/usr/local/lib test.o -lgcc -lc -lgcc /usr/local/lib/gcc-lib/sparc-sun-solaris2.7/2.95.3/crtend.o /usr/local/lib/gcc-lib/sparc-sun-solaris2.7/2.95.3/crtn.o ld: Software Generation Utilities - Solaris-ELF (4.0) Undefined first referenced symbol in file boost::bad_expression::~bad_expression(void)test.o boost::re_detail::c_traits_base::class_map test.o cout test.o string_char_traits<char>::assign(char &, char const &)test.o ostream::write(char const *, int) test.o string_char_traits<char>::compare(char const *, char const *, unsigned int)test.o boost::bad_pattern virtual table test.o __length_error(char const *) test.o boost::bad_expression virtual table test.o istream::ipfx1(void) test.o boost::c_regex_traits<char>::transform(basic_string<char, string_char_traits<char>, __default_alloc_template<false, 0> > &, basic_string<char, string_char_traits<char>, __default_alloc_template<false, 0> > const &)test.o ostream::operator<<(ostream &(*)(ostream &))test.o boost::c_regex_traits<char>::init(void)test.o endl(ostream &) test.o boost::re_detail::c_traits_base::lower_case_map test.o string_char_traits<char>::set(char *, char const &, unsigned int)test.o string_char_traits<char>::eos(void) test.o cerr test.o boost::bad_pattern::~bad_pattern(void)test.o boost::regbase::regbase(void) test.o ios::rdbuf(void) const test.o ostream::operator<<(char const *) test.o boost::re_detail::c_traits_base::syntax_map test.o runtime_error virtual table test.o boost::c_regex_traits<char>::transform_primary(basic_string<char, string_char_traits<char>, __default_alloc_template<false, 0> > &, basic_string<char, string_char_traits<char>, __default_alloc_template<false, 0> > const &)test.o boost::c_regex_traits<char>::toi(char)test.o ostream::operator<<(unsigned int) test.o string_char_traits<char>::copy(char *, char const *, unsigned int)test.o boost::re_detail::c_traits_base::do_lookup_class(char const *)test.o boost::c_regex_traits<char>::m_free(void)test.o boost::re_detail::c_traits_base::error_string(unsigned int)test.o streambuf::sbumpc(void) test.o boost::c_regex_traits<char>::update(void)test.o boost::re_detail::c_traits_base::do_lookup_collate(basic_string<char, string_char_traits<char>, __default_alloc_template<false, 0> > &, char const *)test.o cin test.o string_char_traits<char>::length(char const *)test.o istream::isfx(void) test.o boost::bad_expression::bad_expression(basic_string<char, string_char_traits<char>, __default_alloc_template<false, 0> > const &)test.o string_char_traits<char>::move(char *, char const *, unsigned int)test.o boost::c_regex_traits<char>::toi(char const *&, char const *, int)test.o ios::setstate(int) test.o __out_of_range(char const *) test.o ld: fatal: Symbol referencing errors. No output written to test collect2: ld returned 1 exit status *** Error code 1 make: Fatal error: Command failed for target `test' ********************************************************* Just a note, I don't have any errors when I test tokenizer ex: *************************************** #include <boost/regex.hpp> #include <iostream> #include <boost/tokenizer.hpp> int main(){ using namespace std; using namespace boost; string s = "Field 1,\"putting quotes around fields, allows commas\",Field 3"; tokenizer<escaped_list_separator<char> > tok(s); for(tokenizer<escaped_list_separator<char> >::iterator beg=tok.begin(); beg!=tok.end();++beg){ cout << *beg << "\n"; } } **************************************** however, it works with g++ only and not gcc. :/ - thank for any info. - David

1 0

Re: [Boost-Users] Re: Newbie needs pointer to currently executing thread.
by dick.bridges＠tais.com 10 Apr '02

10 Apr '02

I'm implementing the Leader/Followers thread pool pattern - reference "Pattern-Oriented Software Architecture Volume 2" (Schmidt, et al). When a thread is released via notify_one(), it is [assumed to be] in the "leader" role and stores it's id. A sanity check should be performed to verify that there is no other thread in the leader role (i.e., current_leader is "none") before the thread stores it's id there. When one of the event sources is signaled and the leader is unblocked, it must, among other tasks, demote itself (i.e., set the current_leader value to "none") and promote one of the followers by notify_one() before continuing to service the request. Because of the potential for race conditions, (especially in the context of future program updates, modifications, etc.) I should enforce the rule that only the leader can promote a follower. Something along the lines of: current_leader == Thread::this(). "dmoore99atwork" <dmoore@ashtontec To: Boost-Users(a)yahoogroups.com hgroup.com> cc: Subject: [Boost-Users] Re: Newbie needs pointer to currently executing thread. 04/09/2002 06:53 PM Please respond to Boost-Users --- In Boost-Users@y..., dick.bridges@t... wrote: > Threads in a thread_group - derived pool are waiting on a condition. > Condition gets notify_one(). How do I get id/pointer for the thread that > was released? I've got a feeling I'm gonna be *REAL* embarrassed when I > find out, but I can't figure it out. > > Thanks for any help. Dick, In general, condition variables (both the Boost.Threads version and the POSIX version) don't tell the "notifier" which thread woke up. That's not necessarily a problem, though. A thread pool usually implies a set of threads that are equivalent in some way, even running the same thread function. Referring to the example at: http://www.boost.org/libs/thread/doc/condition.html#Example, Take a look the the receive() function. while (buffered == 0) buffer_not_empty.wait(lk); You might have three threads which reach this point when buffered==0 (meaning the buffer is empty). When a new sender in a fourth thread adds something to the buffer and calls notify_one(), exactly one of the three threads waiting on the condition will wake up and do some work. You won't care which thread wakes up, because any one of them will be prepared to do the necessary work. If this doesn't get you unstuck, please post more details, in particular why you feel it is important to know which thread "woke up" Regards, Dave Info: <http://www.boost.org> Wiki: <http://www.crystalclearsoftware.com/cgi-bin/boost_wiki/wiki.pl> Unsubscribe: <mailto:boost-users-unsubscribe@yahoogroups.com> Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/

2 1