July 2003 - Boost-users - lists.preview.boost.org

RE: [Boost-Users] Using date_time::microsec_clock in windows ?
by Jim.Hyslop 16 Jul '03

16 Jul '03

Jeff Garland wrote: > It actually shouldn't be too hard. It is really a matter of > swapping out the > Windows API function calls to return the times and then adjusting > appropriately. (follow-ups should probably be diverted to the Boost developers list) I'm not sure if this has been discussed or not, but it may be harder than you imagine. The Windows API functions dealing with time have a resolution of approximately 10 milliseconds. Even the functions that return values of hectonanoseconds (100 nanoseconds) are limited to this resolution. To see what I mean, run this program: #include <windows.h> #include <iostream> #include <map> const int numElements=10000; DWORD WINAPI timeThread(void *) { // Step 1: fill in an array of FILETIME elements with the time // returned by GetSystemTimeAsFileTime. Make sure each element // is unique - don't increment index until the time we just read // is actually different from the last time we read. FILETIME ft[numElements]; memset(&ft[0], 0, sizeof(ft)); GetSystemTimeAsFileTime(&ft[0]); int index=1; while (index < numElements) { GetSystemTimeAsFileTime(&ft[index]); if (ft[index].dwLowDateTime != ft[index-1].dwLowDateTime) index++; } // Step 2: calculate the delta from one timestamp to the next. Store // this delta in a map<int, int> where the key is the delta, and the value // is the number of times that delta appears. For example, with the timestamps: // 0, 5, 11, 16, 21, 25 // we get the deltas: // 5, 6, 5, 5, 4 // the map will contain three entries: pair(4, 1 ), pair( 5, 3 ), pair( 6, 1 ) unsigned smallestDiff=(unsigned)-1, largestDiff=0; ULARGE_INTEGER endTime; ULARGE_INTEGER startTime; startTime.LowPart = ft[0].dwLowDateTime; startTime.HighPart = ft[0].dwHighDateTime; std::map<int, int> distribution; for (index=1; index<numElements; ++index) { endTime.LowPart = ft[index].dwLowDateTime; endTime.HighPart = ft[index].dwHighDateTime; unsigned difference = endTime.QuadPart - startTime.QuadPart; std::cout << difference << "\n"; if (difference < smallestDiff) smallestDiff = difference; if (difference > largestDiff) largestDiff = difference; startTime = endTime; distribution[difference]++; } // step 3: dump the stats. endTime.LowPart = ft[numElements-1].dwLowDateTime; endTime.HighPart = ft[numElements-1].dwHighDateTime; startTime.LowPart = ft[0].dwLowDateTime; startTime.HighPart = ft[0].dwHighDateTime; std::cout << numElements << " samples. Smallest delta: " << smallestDiff/10000.0 << "ms; Largest delta: " << largestDiff/10000.0 << "ms; Average: " << unsigned long (endTime.QuadPart - startTime.QuadPart) / (numElements * 10000.0) << "ms\nDistribution:" << std::endl; for (std::map<int, int>::iterator i = distribution.begin(); i != distribution.end(); ++i) { std::cout << (*i).first << ": " << (*i).second << "\n"; } return 0; } int main() { std::cout << "Note: this program will run for approximately " << numElements/100 << " seconds\nwith no output, at real-time priority. \n" "Your system may appear to hang while this program is running.\n" "Rest assured it *will* terminate." << std::endl; HANDLE threadHandle = CreateThread(NULL, 0, timeThread, 0, CREATE_SUSPENDED, 0); SetThreadPriority(threadHandle, THREAD_PRIORITY_TIME_CRITICAL); ResumeThread(threadHandle); WaitForSingleObject(threadHandle, -1); } When I ran it on NT4 SP6, the distribution was as follows: 100136: 7614 100137: 2359 200272: 1 500680: 17 500685: 7 600822: 1 (the units are hectonanoseconds, divide by 10 to get microseconds). I'm guessing the 50 and 60 ms timings are related to task switching or other O/S overhead. On a machine running Windows 2000 SP 3, the results were: 100144: 9999 -- Jim

2 1

Help with boost functional
by Carl Waldbieser 16 Jul '03

16 Jul '03

The following code produces errors when compiled on MSVC6: ************* Start test.cpp ************************ #include <iostream> #include "boost/functional.hpp" struct Foo { int operator()(int x) const {return x+1;} }; typedef class boost::const_mem_fun1_t<int,struct Foo,int> FType; int main() { Foo foo; std::cout << foo(1) << "\n"; //This works //std::cout << boost::bind2nd<FType>(boost::mem_fun(&Foo::operator ()), 7)(&foo) << "\n"; //This does not. std::cout << boost::bind2nd(boost::mem_fun(&Foo::operator ()), 7)(&foo) << "\n"; return 0; } ****************** end test.cpp ******************* The error is as follows: --------------------Configuration: mem_fun_test - Win32 Debug-------------------- Compiling... test.cpp C:\Documents and Settings\waldbcar.WALDBCAR\My Documents\My Projects\VC\test\mem_fun_test\test.cpp(20) : error C2782: 'class boost::binder2nd<Operation> __cdecl boost::bind2nd(Operation &,const Operation &)' : template parameter 'Operation' is ambig uous could be 'int' or 'class boost::const_mem_fun1_t<int,struct Foo,int>' C:\Documents and Settings\waldbcar.WALDBCAR\My Documents\My Projects\VC\test\mem_fun_test\test.cpp(20) : error C2782: 'class boost::binder2nd<Operation> __cdecl boost::bind2nd(const Operation &,const Operation &)' : template parameter 'Operation' is ambiguous could be 'int' or 'class boost::const_mem_fun1_t<int,struct Foo,int>' Error executing cl.exe. test.obj - 2 error(s), 0 warning(s) I am just not seeing where the ambiguity is coming from. The commented out line will actually compile and run as I expected, but I can't figure out why the compiler can't deduce the template argument. Any ideas? Thanks in Advance! Carl Waldbieser waldbie[at-sign-here]attglobal.net

2 1

libbgl-viz.a instead of libbglviz.a
by dhofstee 16 Jul '03

16 Jul '03

Hi, I'm quite new at using boost (downloaded 1.30.0 a couple of days ago). I noticed that when I wanted to make libbglviz.a, as described in http://www.boost.org/libs/graph/doc/read-graphviz.html, there was already a makefile. "make libbglviz.a" does not work, but "make libbgl-viz.a" will. A second documentation error is in http://www.boost.org/libs/property_map/property_map.html, where the example is not equal to the actual cpp file. generic_foo_algorithm is on this page, but the file just mentions foo. I think it is a glitch in the doc, but if different... Comments?, David NL

2 1

Is vc7 test unresolved external cpp_main still a problem?
by dick.bridges＠tais.com 16 Jul '03

16 Jul '03

The last reference to this failure I found in the archive was from Feb 19th. I got boost_1_30_0.zip from SourceForge yesterday. Have I missed an update to something somewhere? TIA <bjam log> ...found 2392 targets... ...updating 28 targets... FileClone libs\regex\build\bin-stage\boost_regex_debug.lib 1 file(s) copied. FileClone libs\regex\build\bin-stage\boost_regex_debug.lib 1 file(s) copied. FileClone libs\regex\build\bin-stage\boost_regex.lib 1 file(s) copied. FileClone libs\regex\build\bin-stage\boost_regex.lib 1 file(s) copied. vc-Link libs\date_time\build\bin\boost_date_time.dll\vc7\debug\runtime-link-dynamic\boost_date_time.dll libs\date_time\build\bin\boost_date_time.dll\vc7\debug\runtime-link-dynamic\boost_date_time.lib vc-Link libs\date_time\build\bin\boost_date_time.dll\vc7\release\runtime-link-dynamic\boost_date_time.dll libs\date_time\build\bin\boost_date_time.dll\vc7\release\runtime-link-dynamic\boost_date_time.lib LINK : warning LNK4089: all references to 'MSVCP71.dll' discarded by /OPT:REF vc-Link libs\test\build\bin\boost_prg_exec_monitor.dll\vc7\debug\runtime-link-static\threading-single\boost_prg_exec_monitor.dll libs\test\build\bin\boost_prg_exec_monitor.dll\vc7\debug\runtime-link-static\threading-single\boost_prg_exec_monitor.lib cpp_main.obj : error LNK2019: unresolved external symbol "int __cdecl cpp_main(int,char * * const)" (?cpp_main@@YAHHQAPAD@Z) referenced in function "public: virtual int __thiscall `anonymous namespace'::cpp_main_caller::function(void)" (?function@cpp_main_caller@?A0xe9549083@@UAEHXZ) libs\test\build\bin\boost_prg_exec_monitor.dll\vc7\debug\runtime-link-static\threading-single\boost_prg_exec_monitor.dll : fatal error LNK1120: 1 unresolved externals "link" /nologo /INCREMENTAL:NO /DEBUG /DLL /subsystem:console /out:"libs\test\build\bin\boost_prg_exec_monitor.dll\vc7\debug\runtime-link-static\threading-single\boost_prg_exec_monitor.dll" /IMPLIB:"libs\test\build\bin\boost_prg_exec_monitor.dll\vc7\debug\runtime-link-static\threading-single\boost_prg_exec_monitor.lib" /LIBPATH:"C:\Program Files\Microsoft Visual Studio .NET 2003\VC7\lib" @"libs\test\build\bin\boost_prg_exec_monitor.dll\vc7\debug\runtime-link-static\threading-single\boost_prg_exec_monitor.CMD" ...failed vc-Link libs\test\build\bin\boost_prg_exec_monitor.dll\vc7\debug\runtime-link-static\threading-single\boost_prg_exec_monitor.dll libs\test\build\bin\boost_prg_exec_monitor.dll\vc7\debug\runtime-link-static\threading-single\boost_prg_exec_monitor.lib... ...skipped <libs!test!build\boost_prg_exec_monitor.dll\vc7\debug\runtime-link-static\threading-single>boost_prg_exec_monitor.dll for lack of <libs!test!build\boost_prg_exec_monitor.dll\vc7\debug\runtime-link-static\threading-single>boost_prg_exec_monitor.lib... vc-Link libs\test\build\bin\boost_prg_exec_monitor.dll\vc7\debug\runtime-link-static\threading-multi\boost_prg_exec_monitor.dll libs\test\build\bin\boost_prg_exec_monitor.dll\vc7\debug\runtime-link-static\threading-multi\boost_prg_exec_monitor.lib cpp_main.obj : error LNK2019: unresolved external symbol "int __cdecl cpp_main(int,char * * const)" (?cpp_main@@YAHHQAPAD@Z) referenced in function "public: virtual int __thiscall `anonymous namespace'::cpp_main_caller::function(void)" (?function@cpp_main_caller@?A0xe9549083@@UAEHXZ) libs\test\build\bin\boost_prg_exec_monitor.dll\vc7\debug\runtime-link-static\threading-multi\boost_prg_exec_monitor.dll : fatal error LNK1120: 1 unresolved externals <snip> ...failed updating 12 targets... ...skipped 12 targets... ...updated 4 targets... </bjam log>

3 2

Using date_time::microsec_clock in windows ?
by J�r�me Poulin 15 Jul '03

15 Jul '03

Hi, I would like to know why we can't use the microsec_clock in a windows env. I know there is no gettimeofday function, but there is certainly a viable workaround. Why it isn't in boost yet, is it planned ? How do you guys suggest to emulate the microsec_clock on a Windows compilation ? Thanks. Jérôme

2 1

profiling ublas code
by cerenoc 15 Jul '03

15 Jul '03

Hi, I am trying to optimize my ublas-containing code. The profile from gprof showed these two functions as the top 2 time-consuming ones. Could somebody explain to me a little bit about what these mean? % cumulative self self total time seconds seconds calls ms/call ms/call name 23.76 93.11 93.11 137604968 0.00 0.00 vector<int, allocator<int> > * __uninitialized_copy_aux<vector<int, allocator<int> > *, vector<int, allocator<int> > *>(vector<int, allocator<int> > *, vector<int, allocator<int> > *, vector<int, allocator<int> > *, __false_type) 21.49 177.31 84.20 41208000 0.00 0.00 vector<int, allocator<int> >::operator=(vector<int, allocator<int> > const &) Thanks, Julian

2 1

Assertion tripped in boost::uniform_int
by mvogt＠juptech.com 15 Jul '03

15 Jul '03

Hello. I just experienced an assertion failure using the boost::minstd_rand generator: /home/packages/boost_1_29_0/boost/random/uniform_int.hpp:50: boost::uniform_int<UniformRandomNumberGenerator, IntType>::uniform_int (UniformRandomNumberGenerator &, IntType, IntType) [with UniformRandomNumberGenerator = boost::random::linear_congruential<int32_t, 48271, 0, 2147483647, 399268537>, IntType = int]: Assertion `min < max' failed. I use the following code to generate random numbers: typedef boost::minstd_rand RngType; extern RngType pseudo_rng; template <class T> T random(T min, T max) { boost::random_number_generator<RngType, T> rng(pseudo_rng); return rng(max - min + 1) + min; } Am I doing something wrong, or is there a fault in the uniform_int algorithm? Thanks, Matt. ********************************************************************** This email and any files transmitted with it are confidential and intended solely for the use of the individual or entity to whom they are addressed. If you have received this email in error please notify mail.administrator(a)jupiters.com.au ********************************************************************** [Non-text portions of this message have been removed]

4 4

Spirit v1.7.0 and v1.6.1 Released
by Joel de Guzman 13 Jul '03

13 Jul '03

Hello, Spirit v1.7.0 and v1.6.1 have been released. Get it here: Spirit v1.7.0: http://tinyurl.com/gssn Spirit v1.6.1: http://tinyurl.com/gsss Take note that by convention, odd minor version releases (e.g. 1.7.0) are developmental while even minor version releases (e.g. 1.6.1) are stable. What's new: http://spirit.sourceforge.net/distrib/change_log.html Have fun!!! Cheers, -- Joel de Guzman joel at boost-consulting.com http://www.boost-consulting.com http://spirit.sf.net

1 0

Logging library ?
by drkm 13 Jul '03

13 Jul '03

Hello. Is there any logging fonctionality in the Boost libraries ? I found nothing. If there is nothing, is there a plan to add a such library to Boost ? --drkm

2 1

MPL::lambda issue: what am I doing wrong?
by Brian Simpson 11 Jul '03

11 Jul '03

I want to express a "type-mapping" scheme. My implementation idea is this: have the user define an MPL sequence comprised exclusively of std::pair derivations. He can then instantiate my template class with his sequence as the argument. My template then derives two MPL sequences, one containing the "first" types, the other containing the "second" types. Here is the attempt: <code> namespace ls { using namespace boost::mpl; using namespace placeholders; template <class TypePairSequence> struct list_splitter { template <class STDPAIR> struct first_type { typedef typename STDPAIR::first_type type; }; template <class STDPAIR> struct second_type { typedef typename STDPAIR::second_type type; }; typedef typename transform < TypePairSequence , first_type<_1> > ::type first_types; typedef typename transform < TypePairSequence , second_type<_1> >::type second_types; }; }//end namespace ls </code> Here is an attempt to instantiate: 1> typedef boost::mpl::list< std::pair<long, short>, std::pair<char, double> > associated_types; 2> typedef ls::list_splitter<associated_types> split_associated_types; 3> typedef split_associated_types::first_types first_types; 4> typedef split_associated_types::second_types second_types; 5> typedef boost::mpl::at_c<first_types, 0>::type first_first_t; Code lines 1-2 compile fine. When I add line 3, I get the following error: "no type named `first_type' in `struct boost::mpl::arg<1>'". If you need more: <error> temp.h: In instantiation of `ls::list_splitter<associated_types>::first_type<boost::mpl::arg<1> >': boost/mpl/transform.hpp:41: instantiated from `boost::mpl::apply1<ls::list_splitter<associated_types>::first_type<boost::m pl::arg<1> >, std::pair<char, double> >' boost/mpl/transform.hpp:41: instantiated from `boost::mpl::aux::transform_op<ls::list_splitter<associated_types>::first_ty pe<boost::mpl::arg<1> > >::apply<boost::mpl::null_node, std::pair<char, double> >' boost/mpl/aux_/preprocessed/gcc/fold_backward_impl.hpp:74: instantiated from `boost::mpl::aux::fold_backward_impl<2, boost::mpl::list_iterator<boost::mpl::list2<std::pair<long int, short int>, std::pair<char, double> > >, boost::mpl::list_iterator<boost::mpl::null_node>, boost::mpl::null_node, boost::mpl::protect<boost::mpl::aux::transform_op<ls::list_splitter<associat ed_types>::first_type<boost::mpl::arg<1> > > >, boost::mpl::arg<1> >' boost/mpl/fold_backward.hpp:45: instantiated from `boost::mpl::fold_backward<associated_types, boost::mpl::null_node, boost::mpl::protect<boost::mpl::aux::transform_op<ls::list_splitter<associat ed_types>::first_type<boost::mpl::arg<1> > > >, boost::mpl::arg<1> >' boost/mpl/transform.hpp:64: instantiated from `boost::mpl::algo_::transform<associated_types, ls::list_splitter<associated_types>::first_type<boost::mpl::arg<1> > >' temp.h:35: instantiated from `ls::list_splitter<associated_types>' temp.h:53: instantiated from here temp.h:19: no type named `first_type' in `struct boost::mpl::arg<1>' </error> I'm sure I'm overlooking something simple. Could someone point me in the right direction? I'm using gcc3.2.2 on Solaris 9. TIA, Brian

2 4

Defining tuples and variants based on mpl::sequences
by Brian Simpson 11 Jul '03

11 Jul '03

Here is the problem: A tuple or variant needs to be defined, whose types are derived from an mpl::list (or other model of MPL Sequence). How is it to be done? I have considered that I might be able to define the tuple/variant with the types and then generate the sequence from there, but that still leaves me wondering how to define another tuple/variant with a sub- or super-set of the types in the generated sequence. Consider that I am supplied with a sequence of types, and that I want to define either 1)a tuple of pointers to them, or 2)a variant based on those types as well as pointers to them. I have considered that it would be a good thing to provide sequence-based tuples/variants and still provide the current variadic template (<--is that the correct term?) versions--which could be implemented in terms of the sequence-based ones (in my uneducated opinion). But I wonder if there is another solution already available which I have not considered. TIA for your consideration, Brian

1 0

Boost.Python interface to Boost Graph Library (BGL)
by BP 11 Jul '03

11 Jul '03

I would like to use the Boost Graph Library (BGL), but I'm not much of a C++ programmer. I'd feel much more comfortable working in Python. Can I download the Boost Graph Library wrapped in Boost.Python? I'd rather not wrap myself if it already exists. Thanks. - BP __________________________________ Do you Yahoo!? SBC Yahoo! DSL - Now only $29.95 per month! http://sbc.yahoo.com

3 2

templated 'handle' in pimpl code? (Slightly off topic)
by Kevin Wheatley 11 Jul '03

11 Jul '03

Hi, I wondered if there are any examples of a templated handle class within the boost code base I could look at? I'm trying to hide both the implementation of the body class and that of the template type, something like that below but my compiler is complaining about partial specialization in the declaration of the implementation class in handle.cpp I know its not strictly boost related, but I don't get any of the comp.lang.* groups. Also before anybody suggests I should be using a smart pointer for the impl variable I'd like to avoid having to introduce the boost library to the user of the two .h files. My goal was that the users would only need the *.h files and I could provide fred.o fred2.o fred3.o etc. as new options for "fred" come along. I know that using the template doesn't appear to be getting me anywhere towads my original goal (generally needing the implementation to get the templated code to work!) but I'd like to see if I could at least get the code to compile by studying anything similar as it may have some use in other areas. Thanks Kevin // -------------- fred.h class fred { public: void somefunction(); }; // -------------- handle.h template <typename T> class handle { public: handle(); ~handle(); void callsomefunctionviaimpl(); private: template <typename U> class handleimpl; handleimpl<T> *impl; }; // -------------- fred.cpp #include <iostream> using namespace std; #include "fred.h" void fred::somefunction() { cout << "fred::somefunction()" << endl; } // -------------- handle.cpp #include <iostream> using namespace std; #include "handle.h" template <class T, class U> class handle<T>::handleimpl<U> { public: void implfunctiononfred() { myfred.somefunction(); } private: static T myfred; }; template <class T> handle<T>::handle() : impl(new handle<T>::handleimpl<T>) { } template <class T> handle<T>::~handle() { delete impl; } template <class T> void handle<T>::callsomefunctionviaimpl() { cout << "handle calling" << endl; impl->implfunctiononfred(); } // ----------- Main code ---------------- #include "fred.h" #include "handle.h" int main(int argc, char** argv) { handle<fred> foo; foo.callsomefunctionviaimpl(); return 0; } -- | Kevin Wheatley | These are the opinions of | | Senior Do-er of Technical Things | nobody and are not shared | | Cinesite (Europe) Ltd | by my employers |

1 1

multi-array versus valarray
by lformaggia 11 Jul '03

11 Jul '03

Hi, I am a new user of boost and I am interest in the multi-array library. However it is not clear to me which are the advantages versus the std::valarray and std::slice classes. There are quite a few numeric functions provided by C++ that work on valarrays (the pow() for example), is there any equivalent for multi-arrays )it would be very nice to have a norm() function at least for two-dimensional arrays!) or you have to write your own. Sorry if this question may seem trivial, yet I have to choose which basic multiarray classes to implemet for my code (it is numerical code based on the finite element method) and so far my choice is between multi-array or valarray + slice. Maybe somebody of you have some experience on this matter and my help in taking an appropriate decision. Thank you Luca

1 0

Having problems with boost::signal and boost::function
by subsample 11 Jul '03

11 Jul '03

I'm having a problem (which was really hard to decipher) when connecting a function object to a signal. Thing is I default construct (no argument) a function object and connect it to the signal (using boost::ref), then at a later point in time I actually assign something to the function object (using boost::bind). But firing the signal after this causes a bad_function_call to be thrown. On the other hand; if I construct the function object with some dummy object before connecting it to the signal, and later do the assignment, the signal does its thing. If someone could enlighten me as to whats going on here, I'd be grateful .. : )

2 1

Signals library safety
by Eugene Talagrand 11 Jul '03

11 Jul '03

Hello to all, I'm looking into using the Boost.Signals. The library claims not to be thread safe. What does this mean exactly? When using Boost.Threads, can I make concurrent calls to connect as long as I wrap each call in a Mutex? If this is possible, do tackable objects also still work (if two trackable objects get destroyed in two different threads there might be a problem from what I understand) On an different note, I am trying to use Boost.Signals for network event handling. I'd like to set it up so that it uses something similar to he JavaScript event model, where every slot returns a boolean value. If that value is false, the remaining slots are not called. Would anyone have any recommendations on how to do this? Should I wrap each slot in a function that takes the return value and sets a global flag, then every subsequent slot checks the flag? It looks like combiners would not be appropriate, as it appears they are used after each slot is called (unless the InputIterator' opeator* actually calls the slot??) Thanks for any help -Eugene

2 1

stream operators and random number generators
by Witz 10 Jul '03

10 Jul '03

hi, in my code i need to be able to save the state of my program including that of he random number generators but under MSVC 7.1.3088 the following code: int main() { boost::mt19937 rng; std::cout << rng; return 0; } won't compile and gives the error: "error C2679: binary '<<' : no operator found which takes a right-hand operand of type 'boost::mt19937' (or there is no acceptable conversion)" there a workaround for this (beyond my hacky solution of adding load and save functions to the mersenne_twister class)? thanks in advance, ian whittley

1 0

BOOST_MPL_DEPENDENT_TEMPLATE_TYPEDEF
by Joaquín Mª López Muñoz 10 Jul '03

10 Jul '03

I've read on some archived post in the boost mailing list about BOOST_MPL_DEPENDENT_TEMPLATE_TYPEDEF, but seems like it is not in the 1.30 distribution anymore. Anyone knows what happened to it? I'm in need of this sort or workaround. Aleksey, are you there? :) Joaquín M López Muñoz Telefónica, Investigación y Desarrollo

2 1

deadlock in filesystem?
by Ken Alverson 10 Jul '03

10 Jul '03

I'm running into an odd deadlock situation, and I was wondering if it's a known issue, or if there's likely something wrong with my code, or if I've misconfigured something? First off, a stack trace: ntdll.dll!77f5b7f4() kernel32.dll!77e61bf5() scanner.exe!boost::detail::sp_counted_base::release() Line 132 + 0xf C++ scanner.exe!boost::detail::shared_count::~shared_count() Line 380 C++ scanner.exe!boost::shared_ptr<boost::filesystem::directory_iterator::dir_itr_i mp>::~shared_ptr<boost::filesystem::directory_iterator::dir_itr_imp>() + 0x2e C++ scanner.exe!boost::shared_ptr<boost::filesystem::directory_iterator::dir_itr_i mp>::reset() Line 217 C++ scanner.exe!boost::filesystem::directory_iterator::m_inc() Line 256 C++ scanner.exe!boost::filesystem::directory_iterator::operator++() Line 88 + 0x2b C++ scanner.exe!std::for_each(boost::filesystem::directory_iterator _First={...}, boost::filesystem::directory_iterator _Last={...}, scan_file _Func={...}) Line 20 + 0xa C++ I'm using boost 1.30.0 running a staticly linked debug build of the filesystem lib, on vc7. The for_each is actually the most recursed level of a recursive directory search - I'm pretty sure that code is fine; I'm very familiar with STL iteration and I'm not doing anything strange, just: std::for_each( fs::directory_iterator(dir), fs::directory_iterator(), scan_files(fileset)); where scan_files will recurse on directories and process on files. I don't touch the iterators directly other than to pass them to for_each. Now, the strange part is this program works completely fine when I simply run it from the command line. However, I am building it into a windows service, and when it runs as a service under the SYSTEM account (I haven't tried running it as a service under a different account, yet), I get deadlock. I suspected the problem might be related to funny permissions on the directory in question, but SYSTEM has full rights to the directory, so that shouldn't be the problem. Oddly, when I use the release filesystem lib it deadlocks in a different location in the search, but in the same sp_counted_base::release() function - this time in an attempt to unwind after a filesystem exception. I tried to search the mailing list for similar problems, but the search seems to be broken at the moment. I'm hoping this is an easy fix, known issue, because I'm at a loss as to how I could be causing it, especially since I'm not getting anywhere near the internals of the directory iterators. I am a fairly new user to boost, but up until now I've had no problems. Sorry if this is a jumble...I've been banging my head against this for the past few hours and my coffee is wearing off. Ken

3 11

newbie build problems with msvc-stlport toolset
by uncle_buck_598 10 Jul '03

10 Jul '03

I am attempting to build the complete boost library set with the command line: bjam -sTOOLS=msvc-stlport "-BUILD=debug release <stlport-iostream>on" Unfortunately, 5 of the release DLLs fail to link with the following error: unresolved external symbol "public: static void __cdecl _STL::_STLP_mutex_spin<0>::_M_do_lock(long volatible *)" ("_M_do_lock@?$_S_STLP_mutex_spin@$0A@@_STL@@SAXPCJ@Z) What is confusing is that the debug DLLs all build with out a hitch! I am using MSVC 6.0 SP5, STLport-4.5.3, and Windows Platform SDK - Nov2002. Any help would be greatly appreciated!! --Jamie

1 0

configuring MSVC 7.1 to use boost build
by Dave Gomboc 10 Jul '03

10 Jul '03

I'm trying to use boost::signals and boost::regex with MSVC 7.1 I retrieved the current CVS state with WinCVS, downloaded a binary of bjam (3.1.4), used the Visual Studio .NET 2003 Command Prompt, invoked bjam with the argument "-sTOOLS=vc7.1", and have added the boost path to Tools|Options|Projects|VC++ Directories (Include Files). When I compile programs I get link errors, e.g. signals_test.obj : error LNK2019: unresolved external symbol "public: void __thiscall boost::signals::connection::disconnect(void)const " (?disconnect@connection@signals@boost@@QBEXXZ) referenced in function "public: __thiscall boost::signals::connection::~connection(void)" (??1connection@signals@boost@@QAE@XZ) signals_test.obj : error LNK2019: unresolved external symbol "public: __thiscall boost::signals::detail::signal_base::~signal_base(void)" (??1signal_base@detail@signals@boost@@QAE@XZ) referenced in function "public: __thiscall boost::signal0<void,struct boost::last_value<void>,int,struct std::less<int>,class boost::function0<void,class std::allocator<class boost::function_base> > >::~signal0<void,struct boost::last_value<void>,int,struct std::less<int>,class boost::function0<void,class std::allocator<class boost::function_base> > >(void)"(??1?$signal0@XU?$last_value@X@boost@@HU?$less@H@std@@V?$functio n0@XV?$allocator@Vfunction_base@boost@@@std@@@2@@boost@@QAE@XZ) I have tried to add library paths in the IDE in a similar fashion to how I added the include path, but I had no success when building a test project. What do I need to do to successfully link? Thanks, Dave P.S. It might be a good idea to add the "vc7.1" target to http://www.boost.org/tools/build/index.html#Configuring.

2 1

Re: CRC problem with 8 bit truncated polynomial
by Daryle Walker 10 Jul '03

10 Jul '03

On Wednesday, July 9, 2003, at 11:02 AM, Victoria Ames wrote: [SNIP] > However, when I have tried examples using 8 bit truncated > polynomials I fail to get a meaningful checksum. > > Example that doesn't work : > std::cout << "Expected CRC-8 : EA" << std::endl; > boost::crc_optimal<8, 0x9B, 0, 0, false, false> crc_8; > crc_8 = std::for_each( data, data + data_len, crc_8 ); > std::cout << "Optimal CRC-8 : " << std::hex << crc_8() << std::endl; > > The results I get are : [SNIP] > Expected CRC-8 : EA > Optimal CRC-8 : Û > > Does anyone know why the 8 bit example does not seem to work? I can't tell at this point. The problem could be that the results of the "crc_8" object is an "unsigned char". The built-in character types can take the same numeric operations as the bigger built-in integral types (i.e. "int"). However, the character types differ from the other integral types since the character types act as C++ text system types. So when an object of a built-in character type is printed, its text value is printed instead of its numeric value. The solution is to convert the result to a numeric-only integral type. Try std::cout << std::hex << static_cast<unsigned>( crc_8() ) << std::endl; to check the numeric value of the result. Then you can tell if the result is correct. If it's not, send us another message. With my hypothesis, the problem is only in the printing. You would have no problem in using the result for further mathematical analyses. Daryle

1 0

RE: [Boost-Users] RE: Re: CRC problem with 8 bit truncated polyno mial
by Jim.Hyslop 09 Jul '03

09 Jul '03

Vladimir Prus wrote: > Sorry, I don't understand what you're saying... > > std::cout << "Optimal CRC-8 : " << std::hex << (unsigned)crc_8() > << std::endl; > > is the same as > > std::cout << "Optimal CRC-8 : " << std::hex << (unsigned int)crc_8() > << std::endl; Sorry, I wasn't very clear in my last message. The results can be unexpected, depending on whether the result of crc_8() (or more precisely, operator() ) is signed or unsigned. If operator() returns an unsigned value, there will be no problems, and the output will be ea. On the other hand, if operator() returns a signed value, then it will first be promoted to a signed int, then converted to an unsigned int. Using the original example values posted by Victoria, the output would be ffffffea. In this scenario, the return value must be first cast to an unsigned char, then to an unsigned int: ... << static_cast<unsigned>( static_cast<unsigned char>( crc_8() ) ) I just had a quick look at the crc.hpp file, and operator() returns an unsigned value, thus banishing my caveat to the dusty corners where all academic footnotes end up ;-) -- Jim

2 1

RE: [Boost-Users] Re: CRC problem with 8 bit truncated polynomial
by Jim.Hyslop 09 Jul '03

09 Jul '03

Vladimir Prus wrote: > I'm guessing this should be > > std::cout << "Optimal CRC-8 : " << std::hex << (unsigned)crc_8() > << std::endl; > > The way iostreams format chars is most likely not what you wanted. True enough - and that includes signed vs. unsigned chars (both are handled the same, at least with my compiler). The result needs to be put into an unsigned *int* for cout to display the numeric value. -- Jim

3 2

RE: [Boost-Users] CRC problem with 8 bit truncated polynomial
by Jim.Hyslop 09 Jul '03

09 Jul '03

victoriaames wrote: > Example that doesn't work : > std::cout << "Expected CRC-8 : EA" << std::endl; > boost::crc_optimal<8, 0x9B, 0, 0, false, false> crc_8; > crc_8 = std::for_each( data, data + data_len, crc_8 ); > std::cout << "Optimal CRC-8 : " << std::hex << crc_8() << std::endl; > > The results I get are : > > Expected CRC-CCITT : 29B1 > Optimal CRC-CCITT : 29b1 > > Expected CRC-8 : EA > Optimal CRC-8 : Û > > Does anyone know why the 8 bit example does not seem to work? I haven't used this library, but judging from the output it looks to me like it's working just fine. The result is an 8-bit value, i.e. a (possibly unsigned) char, with the value 0xEA. Think about what cout does with 'char'. -- Jim

1 0