December 2004 - Boost-users

building on HP-UX
by Michael Garriss 15 Dec '04

15 Dec '04

i'm trying to get a product of ours to work on HP-UX. we are using gcc. it seems to build fine but: 1) when i link to a libboost_*.a i get unresolved symbols like this: prompt> g++ -I/home/mgarriss/dev/tp-compiled/include/boost-1_31_0 test.cxx -L/home/mgarriss/dev/tp-compiled/lib/boost-1_31_0 -lboost_regex-gcc -pthread /usr/bin/ld: Unsatisfied symbols: std::allocator<char>::allocate(unsigned long, void const*)(first referenced in /home/mgarriss/dev/tp-compiled/lib/boost-1_31_0/libboost_regex-gcc.a(instances.o)) (code) std::__default_alloc_template<(bool)1, (int)0>::allocate(unsigned long)(first referenced in /home/mgarriss/dev/tp-compiled/lib/boost-1_31_0/libboost_regex-gcc.a(instances.o)) (code) std::allocator<char>::deallocate(char*, unsigned long)(first referenced in /home/mgarriss/dev/tp-compiled/lib/boost-1_31_0/libboost_regex-gcc.a(instances.o)) (code) std::basic_string<char, std::char_traits<char>, std::allocator<char> >::_Rep::_S_create(unsigned long, std::allocator<char> const&)(first referenced in /home/mgarriss/dev/tp-compiled/lib/boost-1_3 1_0/libboost_regex-gcc.a(instances.o)) (code) std::__default_alloc_template<(bool)1, (int)0>::deallocate(void*, unsigned long)(first referenced in /home/mgarriss/dev/tp-compiled/lib/boost-1_31_0/libboost_regex-gcc.a(instances.o)) (code) collect2: ld returned 1 exit status 2) when i link to a .sl (which i get by just renaming the .so, not sure if that's right), i get this after running the executable: prompt> g++ -I/home/mgarriss/dev/tp-compiled/include/boost-1_31_0 test.cxx -L/home/mgarriss/dev/tp-compiled/lib/boost-1_31_0 -lboost_regex-gcc -pthread prompt> ./a.out /usr/lib/dld.sl: Unresolved symbol: _ZNSs4_Rep9_S_createEmRKSaIcE (code) from /home/mgarriss/dev/tp-compiled/lib/boost-1_31_0/libboost_regex-gcc.sl ABORT instruction (core dumped) AH! can anyone offer any hints, leads, or advice? INFO: prompt> uname -a HP-UX unknown B.11.11 U 9000/800 180901557 unlimited-user license prompt> gcc -v Reading specs from /opt/gcc-3.3.2/lib/gcc-lib/hppa2.0w-hp-hpux11.11/3.3.2/specs Configured with: ../gcc-3.3.2/configure --prefix=/opt/gcc-3.3.2 --disable-nls --disable-libgcj --disable-shared --enable-languages=c,c++ --with-gnu-as --with-gnu-ld --with-as=/usr/local/bin/as --with-ld=/usr/bin/ld --enable-threads=posix Thread model: posix gcc version 3.3.2 Thanks in advance, Mike

3 2

[BGL] pathfinding
by Cromwell Enage 15 Dec '04

15 Dec '04

One of the core functions in my multi-state mazes app automagically creates graphs that are conceptually like the one in the attached program. In this case, the shortest path from vertex 0 to vertex 11 is also the only simple path from vertex 0 to vertex 11. Also, unlike a regular maze, there are no dead ends (only cycles), so no matter how far one deviates from this simple path, there is always a way back. To that end, I'm trying to write a function that gives users a path to the end vertex (11) from any other vertex in the maze graph. Unfortunately, this function currently works properly only when given the start vertex (0). I've tried both breadth-first and depth-first search. The cycles seem to be throwing them off in a manner I haven't figured out, yet. I'll keep investigating the problem, but if anyone else has dealt with something like this before, please let me know how I should proceed. Cromwell D. Enage __________________________________ Do you Yahoo!? Yahoo! Mail - Helps protect you from nasty viruses. http://promotions.yahoo.com/new_mail

2 2

crc.hpp with g++ 3.4.2
by Kathryn Hogg 15 Dec '04

15 Dec '04

including boost/crc.hpp when compiling with g++ 3.4.2 yields the following compiler error: /users/kjh/src/toolkits/boost/crc.hpp:464: error: ISO C++ forbids variable-size array `table_type' And the code in question is #if defined(__BORLANDC__) && defined(_M_IX86) && (__BORLANDC__ == 0x560) // for some reason Borland's command line compiler (version 0x560) // chokes over this unless we do the calculation for it: typedef value_type table_type[ 0x100 ]; #else typedef value_type table_type[ byte_combos ]; #endif Apparantly some version of g++ need to be included in the #if test as well. Pretty trivial to do but I thought I'd post here so the maintainers of the crc library can get it officially fixed. -- Kathryn http://womensfooty.com

2 1

Boost.Build: how to specify targets on command line
by Davlet Panech 14 Dec '04

14 Dec '04

Hello, I want to build one variant of a one boost DLL library using the regular bjam scripts; I tried the following tools/build/jam_src/bin.ntx86/bjam \ -q \ -d2 \ -sTOOLS=msvc \ -sBUILD="debug <runtime-link>dynamic <threading>multi" \ boost_date_time which builds *both* the DLL and the static lib, but I want only the DLL. I'm guessing I'd have to replace "boost_date_time" with some mangled "fully-qualified" target name, what is it? Another question, is it possible to "stage" individual libraries? Thanks, D.

1 1

boost/thread/xtime.hpp
by Weston Markham 14 Dec '04

14 Dec '04

The Boost.Threads library provides boost::xtime and boost::xtime_get. I have some questions and concerns about these: I read the page at http://www.cl.cam.ac.uk/~mgk25/time/c/ in order to understand what the intended semantics of boost::TIME_UTC are. I assume that this matches the possible C proposal that is referred to in the Boost documentation. I also read the "Terminology" section of the Boost.Date_Time documentation. 1. First of all, should these (boost::xtime and boost::xtime_get) be treated as deprecated, in favor of the types and functions in Boost.Date_Time? (If so, then my further questions might be merely academic.) 2. It appears that the boost::xtime type is used both for the purpose of measuring the length of an interval of time, as well as for designating individual points in time. Is this correct? 3. thread::sleep (as well as other Boost.Thread functions) delays the execution of a thread for a specified interval, which implicitly begins at (approximately) the time of the call. Is its input parameter intended to be the other endpoint of the interval, or is it intended to measure the length of the interval? Both the documentation and the implementation lead me to believe that the former is the case. However, the documentation seems to me to be slightly ambiguous, and, of course, using a function's implementation to determine its expected behavior is questionable. (A similar question applies to the TimedLock concept, although I haven't looked at its details.) 4. The only type of clock that xtime_get supports (according to the documentation) is TIME_UTC. This does not seem useful to me, because I am mainly interested in providing accuracy for the intervals being timed, as opposed to providing accuracy relative to some externally-defined epoch. 5. Does boost::thread::sleep use the boost::TIME_UTC clock? If the parameter specifies the other endpoint, then I don't see how anything else could make sense. 6. What happens when someone applies a correction to the system time on my machine, in an attempt to synchronize it with UTC? Intuitively, I expect any functions that report UTC time to report a discontinuity, because the newer estimate should always be considered more accurate, but it was not known until the adjustment was made. Consider the code that I currently write in order to "sleep 5 seconds": boost::xtime xt; boost::xtime_get(&xt, boost::TIME_UTC); xt.sec += 5; boost::thread::sleep(xt); How long should the waited interval last? In most cases, it will be just under 5 seconds. If a leap second occurs within 5 seconds after the xtime_get, the wait ought to last about 6 seconds. (My guess is that most implementations would not provide this behavior, however.) And in the event that the TIME_UTC clock does not exactly match UTC time, and is subject to adjustments, the sleep could be instantaneous, or perhaps last a minute or more. (Here in my office, my workstation and the phone system disagree by 1 minute right now.) If boost::thread::sleep is intended to wait until a specified UTC time, (this is the conclusion that I have drawn from the documentation and the current implementation) then I think this is a conceptual flaw in the library. I have never had any need to do such a thing. It is always more important to me in this context that the duration of the interval be accurate, rather than its relation to, say, the rotation of the earth. Admittedly, I will get the behavior that I want in most cases. However, it seems senseless to me to confuse the matter of timing short intervals with reasonable accuracy by introducing UTC time, particularly when practically all computing hardware already has the facilities to perform the measurement task at hand, but does not necessarily have any means of discovering the current UTC time. My concerns could be mitigated if boost::TIME_UTC were renamed so that it did not have UTC in its name. It should then be documented as indicating a request for a (monotonic, if possible) clock that is suitable for timing intervals with a resolution around 1 second. The underlying implementations should then prefer system calls that enable accurate measurement of intervals, rather than synchronization with an external time base. Plus, the documentation for boost::thread::sleep should make explicit the manner in which it will interpret its argument. A search of the Boost developers' list on Gmane reveals a "problems with boost::thread" thread that appears to mention the issue of applying adjustments to the clock used for TIME_UTC. (And appears to have good suggestions for adding types to fix the ambiguity that causes questions #2 and #3.) I only found it after typing up the bulk of this message. However, it would be good to know if the future direction of the library is to incorporate the suggestions there. Weston

2 1

Thread within thread
by Leon Mergen 14 Dec '04

14 Dec '04

Hello everyone, I'm currently attempting to create a thread within a thread, and for some reason, the constructor seems to block and things just don't work very well. Consider a class ``BackupHandler'' and a class ``BackupHandlerThread'' - the thread is created for once in the 50ms polling to an output file which might be used by other processes. Now, in BackupHandler, I do this: --- boost::thread thread ( BackupHandlerThread (this->backupFile, this->data) ); --- Which, as far as I am aware of, should first create a BackupHandlerThread object, and then "launch" the thread with this object. Now, since I print out some debug info, it seems like it reaches the BackupHandlerThread constructor once, and the destructor twice. The function BackupHandlerThread::operator() () never is reached even once, and the execution of the initial thread blocks at the creation. Also, according to gdb there is no new thread being created, and it adds any use, this is the stacktrace of the thread that should create the new thread: --- [Switching to thread 3 (Thread 16386 (LWP 24085))]#0 0x401d2b94 in __pthread_sigsuspend () from /lib/i686/libpthread.so.0 (gdb) bt #0 0x401d2b94 in __pthread_sigsuspend () from /lib/i686/libpthread.so.0 #1 0x401d29d8 in __pthread_wait_for_restart_signal () from /lib/i686/libpthread.so.0 #2 0x401d2151 in pthread_create@@GLIBC_2.1 () from /lib/i686/libpthread.so.0 #3 0x400a5fc6 in boost::thread::thread(boost::function0<void, std::allocator<boost::function_base> > const&) () from /usr/local/lib/libboost_thread-gcc-mt-1_31.so.1.31.0 #4 0x080846c2 in BackupHandler::addData(std::string) (this=0x40bd2cfc, data= {static npos = 4294967295, _M_dataplus = {<allocator<char>> = {<No data fields>}, _M_p = 0x8084f10 "U\211�\203�\030�D$\f"}, static _S_empty_rep_storage = {0, 0, 0, 0}}) at function_template.hpp:427 #5 0x0808455d in BackupHandler::addData(std::string*) (this=0xfffffffc, data=0xfffffffc) at BackupHandler.cc:36 #6 0x0805a759 in Normaliser::normalise() (this=0x80bd0d8) at Normaliser.cc:81 #7 0x08079c9d in HandleClient::operator()() (this=0x80b9d38) at HandleClient.cc:37 #8 0x08052c81 in boost::detail::function::void_function_obj_invoker0<HandleClient, void>::invoke(boost::detail::function::any_pointer) (function_obj_ptr= {obj_ptr = 0xfffffffc, const_obj_ptr = 0xfffffffc, func_ptr = 0xfffffffc, data = "�"}) at function_template.hpp:128 #9 0x400a69f4 in boost::thread_group::join_all() () from /usr/local/lib/libboost_thread-gcc-mt-1_31.so.1.31.0 #10 0x401cff60 in pthread_start_thread () from /lib/i686/libpthread.so.0 #11 0x401d00fe in pthread_start_thread_event () from /lib/i686/libpthread.so.0 #12 0x402f7327 in clone () from /lib/i686/libc.so.6 (gdb) --- Thanks in advance for any ideas! Regards, Leon Mergen

2 3

RE: [Boost-users] Re: Iterators and maps [was Re: Where should ST L suggestionsgo?]
by Foster, Gareth 14 Dec '04

14 Dec '04

"I believe that iterator theories are as central to Computer Science as theories of rings or Banach spaces are central to Mathematics." -- An Interview with A. Stepanov http://www.stlport.org/resources/StepanovUSA.html <http://www.stlport.org/resources/StepanovUSA.html> that's just the tip of the ice berg, i found that interview to be a real eye opener having been through university where only OO programming got extensive coverage. Always seemed odd to me that what is such a big thing in the C++ community seems to get such a 'bad deal'. I hear things are changing though ... </ramble> gaz -----Original Message----- From: boost-users-bounces(a)lists.boost.org [mailto:boost-users-bounces@lists.boost.org] On Behalf Of Jim Lear Sent: 14 December 2004 16:30 To: boost-users(a)lists.boost.org Subject: Re: [Boost-users] Re: Iterators and maps [was Re: Where should STL suggestionsgo?] David Abrahams wrote: Jim Lear wrote: So in the absolute most abhorently poor code example (let's just call it "meta-code"), the map operator would behave like: data_type &operator[](const iterator &i) { return i->second; } Am I just too ignorant to make any sense? :-) Maybe I'll crawl back into my hole. :-) You're making sense, but consider this: why should the map get involved in that operation at all when you can do it all with the iterator? Seems like a design flaw to get another object involved needlessly. Not only might it be a design flaw, but it may invite inefficiency. But really it's just consistency with other languages which iterate over the keys (like awk) and code clarity. Someone unfamiliar with STL would find the following obvious: for (m_type::iterator i=m.begin();...) for (mi_type::iterator j=m[i].begin();...) cout << m[i][j] << endl; But they may find for (m_type::iterator i=m.begin();...) for (mi_type::iterator j=i->second.begin();...) cout << j->second << endl; obtuse. Perhaps folks intimately familiar with iterators may find this is hard to believe. :-) An alternative question might be, why should one who uses maps and vectors also be required to be familiar with the intricacies of iterators? To me, that seems like a design flaw. I have never encountered another language that requires the user to understand iterators to the extent STL does. Ease of use and learning ought to be a priority in the design of these objects, to the extent it is efficient and safe. To use iterators in lieu of keys or indices seems to me like a great simplification for users. It completely obviates the need to understand iterators in this case (while still allowing existing iterator functionality). -- Jim Lear

1 0

Re: Re: Boost-1-32 multi_index compilation error on msvc6.
by Jia Pu 14 Dec '04

14 Dec '04

I forgot to include ordered_index.hpp. I thought including multi_index_container.hpp will give me everything I need. Thank you. > From: Joaquin M Lopez Munoz <joaquin(a)tid.es> > Subject: [Boost-users] Re: Boost-1-32 multi_index compilation error on > msvc6. > To: boost-users(a)lists.boost.org > Message-ID: <loom.20041214T083314-505(a)post.gmane.org> > Content-Type: text/plain; charset=utf-8 > > Jia Pu <very.funny.j <at> gmail.com> writes: > > > > > I am using MSVC 6 with Boost 1-32. Below is my sample code using two indeices: > > > [...] > > When I compile it, I have following errors. The actual message is much > > longer than this. I don't want paste them here to piss off every one > > on this mailing list. :) > > > > Anyway, I am not a template expert. I couldn't figure it out myself. > > Any help will be appreciated. Thanks. > > > > Hi, > > I cannot see what could be going wrong with your example. > The attached complete program compiles OK here > with MSVC 6.0 + SP5. Maybe you can post some complete code > that exhibits the problem? > > JoaquÃn M LÃ³pez MuÃ±oz > TelefÃ³nica, InvestigaciÃ³n y Desarrollo > > #include <boost/multi_index_container.hpp> > #include <boost/multi_index/key_extractors.hpp> > #include <boost/multi_index/ordered_index.hpp> > #include <string> > > struct by_id{}; > struct by_symbol{}; > > struct StringIdPair > { > StringIdPair(int _id, const std::string & _string): > id(_id), > string(_string) > {} > > int id; > std::string string; > }; > > using namespace boost::multi_index; > > typedef multi_index_container< > StringIdPair, > indexed_by< > ordered_unique< > tag<by_id>, > BOOST_MULTI_INDEX_MEMBER(StringIdPair, int, id) >, > ordered_unique< > tag<by_symbol>, > BOOST_MULTI_INDEX_MEMBER(StringIdPair, std::string, string) > > > > > StringIdTable; > > int main() > { > StringIdTable string_id_table; > > string_id_table.insert(StringIdPair(0,"zero")); > string_id_table.insert(StringIdPair(1,"one")); > string_id_table.insert(StringIdPair(2,"two")); > > return 0; > }

1 0

Re: newbie boost sample compile program
by Brian Gaber 14 Dec '04

14 Dec '04

"Martin Lemieux" < <mailto:lemieux@cae.com> lemieux<at>cae.com> wrote: > > As I suspected, those are linker errors! That means steps 1 and 2 worked. Now > (step 3), you need to specify the directory where to find the boost libraries, > as well as the lirbary file(s) for the linker to link with. At this point you > have plenty of choices: static/dynamic, multi-thread or not, debug/release. > > 3a) You might have some of the following library files in your AIX > [...]/boost/lib directory: > > boost_filesystem-AIX-mt-1_31.so > boost_filesystem-AIX-mt-1_31.lib > boost_filesystem-AIX-mt-gd-1_31.so > boost_filesystem-AIX-mt-gd-1_31.lib On AIX v5.2 here are the boost_filesystem* libraries I have under /usr/local/lib libboost_filesystem-xlc-1_32.a libboost_filesystem-xlc-1_32.so libboost_filesystem-xlc-d-1_32.a libboost_filesystem-xlc-d-1_32.so libboost_filesystem-xlc-d.a libboost_filesystem-xlc-d.so libboost_filesystem-xlc-mt-1_32.a libboost_filesystem-xlc-mt-1_32.so libboost_filesystem-xlc-mt-d-1_32.a libboost_filesystem-xlc-mt-d-1_32.so libboost_filesystem-xlc-mt-d.a libboost_filesystem-xlc-mt-d.so libboost_filesystem-xlc-mt.a libboost_filesystem-xlc-mt.so libboost_filesystem-xlc.a libboost_filesystem-xlc.so > On your compiler command line, add "-L /usr/local/include/boost-1_32/boost/lib" > (or make a symbolic link as you did for the include directory > towards /usr/local/include/boost/lib and use this one instead, as you wish). On AIX (maybe all unix's) boost libraries are found in /usr/local/lib > Then (usually at the end of the command line) add the library file you like. > May I the a static one (.lib). Your command line should look like: No boost libraries with a .lib extension > xlc++ -o simple_ls simple_ls.cpp -I /usr/local/include -L /usr/local/include/boost-1_32/boost/lib boost_filesystem-AIX-mt-1_31.lib the actual xlc++ compiler command line that worked was: xlc++ -o simple_ls simple_ls.cpp -I /usr/local/include -lboost_filesystem-xlc-1_32 -L/usr/local/lib NOTE: after the -l the three letters "lib" are removed from the front of the actual library name. > > 2b) On MSDEV6, add "c:\boost\include\boost-1_32" in your project settings. More specifically in the "C++" tab, under "Preprocessor" category (drop-down list), and at "Additional include directories:". > > > > I did this and I can now build, but I get a linker error: > > > > LINK : fatal error LNK1104: cannot open file "libboost_filesystem-vc6-sgd- > 1_32.lib" > > 3b) On MSDEV6, > > i) add "libboost_filesystem-vc6-sgd-1_32.lib" in your project settings. More specifically in the "Link" tab, under "Input" category (drop-down list), and at "Object/Library modules:". > > ii) and add "c:\boost\include\boost-1_32\boost\lib" in "Additional library path:". On my Windows 2000 the file "libboost_filesystem-vc6-sgd-1_32.lib" is found in c:\boost\lib Martin, thanks for your help. I would have never got this working without your assistane.

2 1

Bug in Boost.Range + Boost.String Algorithms
by Raul Huertas Diaz 14 Dec '04

14 Dec '04

I think I have found a bug mixing together this two libraries in Boost 1.32.0 Simply doing: #include <boost/algorithm/string.hpp> #include <boost/range.hpp> it's a bug because boost::algorithm::iterator_range<IteratorT> is redefined. I have attached a small demo program. ¿Am I wrong? ¿Is there a way to bypass this, apart from changing Boost code? Thanks in advance. P.D.: In case it is interesting, I'am working with: - Linux (Mandrake 10.0) - gcc 3.2.1 P.P.D: Boost.String Algorithms is great !!! Boost.Range is superb !! ;) -- Raúl Huertas Díaz ************************************************************************************************************************************************ *La información contenida en este mensaje de correo electrónico es confidencial y puede revestir el carácter de reservada. * *Está dirigida exclusivamente a la persona destinataria. * *El acceso o cualquier uso por parte de cualquier otra persona de este mensaje no están autorizados y pueden ser ilegales.* *Si no es Ud. la persona destinataria, le rogamos que proceda a borrarlo. * *The information in this e-mail is confidential and may be legally privileged. * *It is intended solely for the addressee. * *Access or any use by any other person to this Internet e-mail is not authorised and may be unlawful. * *If you are not the intended recipient, please delete this e-mail. * ************************************************************************************************************************************************

2 1