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

(no subject)
by Paul Grenyer 25 Nov '01

25 Nov '01

Paul Grenyer pjgrenyer(a)iee.org

1 0

"referenced symbol not found" when loading modules
by dlrubin＠yahoo.com 25 Nov '01

25 Nov '01

I'm getting a "refernced symbol not found" error when loading modules compiled with Boost. I tried using boost using this example code: #include <stdio.h> #include <string> #include <boost/python/class_builder.hpp> namespace { std::string greet() { return "hello, world"; } int square(int number) { return number * number; } } namespace python = boost::python; BOOST_PYTHON_MODULE_INIT(example) { try { // Create an object representing this extension module. python::module_builder this_module("example"); // Add regular functions to the module. this_module.def(greet, "greet"); } catch(...) { python::handle_exception(); // Deal with the exception for Python } } and when I try importing into python, here's the error I get: Python 2.1.1 (#5, Nov 22 2001, 14:04:45) [GCC 2.95.2 19991024 (release)] on sunos5 Type "copyright", "credits" or "license" for more information. >>> import example Traceback (most recent call last): File "<stdin>", line 1, in ? ImportError: ld.so.1: python: fatal: relocation error: file ./example.so: symbol handle_exception__Q25boost6pythonv: referenced symbol not found Any suggestions? thanks, Daniel

1 0

Introducing Me
by Paul Grenyer 24 Nov '01

24 Nov '01

Dear All I don't know if this is the appropriate thing to do here (I'm sure the moderator will filter me out if it isn't), but I failed to introduce myself on other lists and with hindsight I think it would have been better if I had. I've been programming in C++ for about 18 months, since I left university and started working. I was introduced to the C++ standard library and boost libraries a few months ago by a member of the ACCU. I have just started a new job writing distributed software using the C++ standard library and the ATL and I'm making as much use of boost as possible. Previous to that I was writing windows software to control dot markers and laser markers using the MFC. So, I am very new to boost and am not expecting to be able to contribute much for a while and I'll no doubt need the odd nudge in the right direct. Thanks Paul Paul Grenyer pjgrenyer(a)iee.org

1 0

Re: [boost] how to get a *real* copy of a boost-python object that supports virtual function
by Phil Austin 24 Nov '01

24 Nov '01

(I'm posting to both boost and boost-users, but I'm assuming we're supposed to move this kind of discussion to Boost-Users(a)yahoogroups.com?) dkottowk(a)yahoo.com writes: [snip] > my question is, how do i get a copy of PyGenome_Callback, > which also is a copy of the python-side of the object ? I'd also be interested in a pointer to an elegant way to do this, as well as an example of a from_python function that converts PyObject* to PyGenome_Callback. Brute force does work for me -- i.e. if you need the derived class and can't construct the object in Python code, you can call the derived constructor from the C++ side like this: PyObject* theMod=PyImport_ImportModule("mymodule"); PyObject* theDict=PyModule_GetDict(theMod); py::ref theRef(theDict); py::dictionary modDict(theRef); py::ref constructor=bDict.get_item(py::string("theObject").reference()); PyObject* theInstance=PyObject_CallObject(constructor.get(),0); This was ugly enough so that I added a python wrapper around my extension to construct the derived objects I needed in Python and pass them back to C++. Phil

1 0

boost tokenizer
by pjgrenyer＠iee.org 23 Nov '01

23 Nov '01

Dear All I'm using the boost tokenizer to tokenise lines similar to this one: first_initial |23|1| This gives me: first initial 23 1 but what I actually want is: first_initial 23 1 Does anyone know how to get it to ignore the underscore? Thanks Paul

3 2

passing data to a thread
by d33pmind＠yahoo.fr 23 Nov '01

23 Nov '01

How can I pass data to a thread ? Why does the boost::thread's constructor takes a boost::function0<void> as parameter, instead of a boost::function1<void> which would have allowed to pass it a pointer to a thread safe structure ? Julien.

2 1

core dump when invoking jam
by d33pmind＠yahoo.fr 23 Nov '01

23 Nov '01

On a Linux Mandrake 8.1 (i586), I succesfully built jam, but when I invoke it I got no output except a core dump. Has anyone encountered the same problem ? I compiled it with gcc 3.0.1, and modified gcc-tools.jam to use gcc- 3.0.1 and g++-3.0.1 instead of gcc and g++ (which are 2.96). Thanks. Julien.

2 1

What's safe_bool?
by mr_lilirong＠yahoo.com 23 Nov '01

23 Nov '01

Hi, The following code is excerpted from Boost.Function. I just don't unstanding it. Why use "safe_bool"? What's the advantage? Can any one explain it to me? Thanks. // code from <boost/function/function_base.hpp> // ===> private: struct dummy { void nonnull() {}; }; typedef void (dummy::*safe_bool)(); public: operator safe_bool () const { return (this->empty())? 0 : &dummy::nonnull; } safe_bool operator!() const { return (this->empty())? &dummy::nonnull : 0; } // code from <boost/function/function_base.hpp> // <===

2 2

aCC compilator
by lokys1976＠yahoo.com 22 Nov '01

22 Nov '01

Is it possible using aCC A.03.27 to use boost libraries? When I try to compile using boost examples, it dosn't understand "using namespece std" and so on, when I try to use aCC -AA than it can not find some libraries. Any suggestions how must I use aCC with boost libraries is welcomed very much thank you Paulius

2 1

Threads on MSVC 6.0
by brian_todoroff＠yahoo.com 22 Nov '01

22 Nov '01

I've been trying for a few hours now to get the threads lib examples to compile. I built the libraries with jam and all but python built correctly. When I create a new Win32 console application and copy the code ( See end of message from the thread class example and attempt to compile I get: ===================Start Error --------------------Configuration: threadtest - Win32 Debug----------- --------- Compiling... threadtest.cpp c:\projects\boost\threadtest\threadtest.cpp(27) : error C2228: left of '.join' must have class/struct/union type Error executing cl.exe. Creating browse info file... threadtest.exe - 1 error(s), 0 warning(s) ===================End Error Can't figure out what I'm doing wrong. Thanks for any help. Brian Todoroff ========================Start Code (to end) #include <boost/thread/xtime.hpp> #include <boost/thread/thread.hpp> #include <iostream> struct thread_alarm { thread_alarm(int secs) : m_secs(secs) { } void operator()() { boost::xtime xt; boost::xtime_get(&xt, boost::TIME_UTC); xt.sec += m_secs; boost::thread::sleep(xt); std::cout << "alarm sounded..." << std::endl; } int m_secs; }; int main(int argc, char* argv[]) { int secs = 5; std::cout << "setting alarm for 5 seconds..." << std::endl; boost::thread thrd(thread_alarm(secs)); thrd.join(); // Line 27 - Error here return 0; }

2 1

breadth_first_searchable graph, without invalidable vertex descriptors
by cristipp＠ics.uci.edu 21 Nov '01

21 Nov '01

Hello, I am trying to use the BCG as backbone for a ControlFlowGraph in a compiler project. It seems that I cannot use BCG for this task because of the follow contradicting (from BCG point of view) requirements: 1. I want to be able to breadth_first_search through my graph. 2. I want to be able to keep descriptors to vertices 'for ever' so I can refer a specific vertex at any given time (for removal or whatever other reasons). To implement requirement 1 my adjacency_list should have its VertexList of type vecS. To implement requirement 2 my adjacency_list should have its VertexList of type listS. Unfortunately those are mutualy exclusive and it seems that BCG is unsuited for the job. But I can't really think of a decent generic graph library unable to support those requirements. After all, it's all about being able to use basic graph traversals on randomly mutating graphs. STL has the list which has those nice properties, why a graph library should be different?! Perhaps I am missing something and I am asking for help: how should I declare my graph and how can I do a bfs on it? Thanks in advance, Cristi

2 5

RE: [Boost-Users] question about any class ...
by scleary＠jerviswebb.com 10 Nov '01

10 Nov '01

> any is really a powerful class, but sometimes I would also be > able to know directly the type stored into it without 'guessing', for > example to use it in template code. > What I would like to have is something as > > type_info::real_type for the value hidden into any (not legal > C++, maybe > legal C++0x ?), This actually wouldn't help you, still. How would the declaration look? typeof() wouldn't work, either... > or, alternatively, virtual template functions (again not > legal C++) to be > able to write generic > code depending on both hidden value's type and types 'from the outside > world' (sorry cannot explain it better :-( than this) I have run into this same problem. To re-phrase it, Gianni is trying to do an "operation" on a contained value. 'any' supports the "operations" of copying and destruction as-is; manual modification of 'any' could support other "operations" (this is what Gianni means when he talks about "visiting" the any class, below): any { ... void f() { if (content) content->f(); } }; any::placeholder { ... virtual void f() = 0; }; any::holder<...> ... { ... virtual void f() { held.f(); } }; The problem is how to do this generically (i.e., provide a solution for any "operation"). Another way of looking at this is to force 'any' to call a function that is overloaded on the type of the contained value. Strictly speaking, this isn't possible because 'any' makes the compile-time info into run-time info. > I have tried various approaches, but if I 'visit' the any > class adding a > virtual function to reach hidden value I loose generic type > information > (lack of virtual template functions) from the outside while from the > outside of any > I cannot obtain directly the type of the hidden value ... > > is there a way to do the above ? I have not been successful. Again, 'any' changes compile-time information into run-time information, and to go the other way is just not possible in this language. You could take a "nasty-hack" approach, if you *really* wanted to: 1) Add an 'any_base' class with a virtual function 2) Extend 'any' to return an 'any_base *' referring to the held value, using C-cast notation (derived-to-base when possible, reinterpret when not) 3) Derive your held value types from a common base class 'my_base', which was derived from 'any_base'; add the operations you want to perform to 'my_base' 4) Call the 'any' function returning 'any_base *', downcasting it to 'my_base'. This will result in undefined behaviour if the contained value is not actually derived from 'my_base' (because of the reinterpret_cast behaviour in step 2) I *think* that will work (never tried it). Messy, though, and IMO not generic enough to mess with. Alternatively, you could: 1) Visit 'any'/'any::placeholder'/'any::holder' to provide: void any::action(any which) { content->action(which); } virtual void any::holder::action(any which) { any_action(held, which); } 2) Declare a function 'any_action(T, any)', where the any parameter designates which "operation" to perform. Overloading is then possible on T. Also, where T is a user-defined type, Koenig lookup can be used (abused?) to create different (non-overlapping) sets of "operations". Just some thoughts. Again, I've never tried it. -Steve

2 1

question about any class ...
by Gianni Luciani 10 Nov '01

10 Nov '01

Hi, any is really a powerful class, but sometimes I would also be able to know directly the type stored into it without 'guessing', for example to use it in template code. What I would like to have is something as type_info::real_type for the value hidden into any (not legal C++, maybe legal C++0x ?), or, alternatively, virtual template functions (again not legal C++) to be able to write generic code depending on both hidden value's type and types 'from the outside world' (sorry cannot explain it better :-( than this) I have tried various approaches, but if I 'visit' the any class adding a virtual function to reach hidden value I loose generic type information (lack of virtual template functions) from the outside while from the outside of any I cannot obtain directly the type of the hidden value ... is there a way to do the above ? Gianni Luciani

2 2

Re: any question
by Kevlin Henney 09 Nov '01

09 Nov '01

> From: Thomas Matelich <toms-mailing-lists(a)home.com> > >In the docs, it uses a function called any_ptr, but I don't think it >exists. Could someone please explain? Yes, nothing more elaborate than a typo, I'm afraid! Should be any_cast. Kevlin ____________________________________________________________ Kevlin Henney phone: +44 117 942 2990 mailto:kevlin@curbralan.com mobile: +44 7801 073 508 http://www.curbralan.com fax: +44 870 052 2289 Curbralan: Consultancy + Training + Development + Review ____________________________________________________________

2 1

casting shared_ptr while preserving use_count..
by John Liberty 09 Nov '01

09 Nov '01

I'd like to essentially downcast (to derived) from a base class with virtual methods while preserving (and incrementing) the use_count.. for an upcast, I can simply assign a shared_ptr<Child> to shared_ptr<Parent>. But for the downcast, I need to incorporate a dynamic_cast<childType> to get what I want.. With an older compiler that didn't like member templates, the boost conf would leave the internal pointer and use count as public, so I wrote my own casting function.. However, the newer compiler doesn't give this public access, so I may be stuck without modifying the header... any thoughts?

2 1

How to build the latest release?
by Sinan 08 Nov '01

08 Nov '01

No matter what I try, ( I set JAMBASE, specified boost-base.jam etc... I always get the following error kanita% jam -f./tools/build/allyourbase.jam -sTOOLS="suncc" ./tools/build/allyourbase.jam: line 1608: syntax error at keyword ( don't know how to make all ...found 1 target... ...can't find 1 target... kanita% Line 1608 is the one that says: rule remember-binding ( target : bound-path ) { gBINDING($(target)) = $(bound-path) ; } ... How do I make jam understand the extended syntax ? ( Using FTjam 2.3.5 ...) Sinan Karasu ---------- # (C) Copyright David Abrahams 2001. Permission to copy, use, # modify, sell and distribute this software is granted provided this # copyright notice appears in all copies. This software is provided # "as is" without express or implied warranty, and with no claim as # to its suitability for any purpose. # compute directories for invoking WS6U2 WS6U2_BIN_DIRECTORY ?= $(WS6U2_ROOT_DIRECTORY)$(SLASH)bin ; WS6U2_BIN_DIRECTORY ?= "" ; # Don't clobber tool names if WS6U2_ROOT_DIRECTORY not set WS6U2_INCLUDE_DIRECTORY ?= $(WS6U2_ROOT_DIRECTORY)$(SLASH)include ; WS6U2_STDLIB_DIRECTORY ?= $(WS6U2_ROOT_DIRECTORY)$(SLASH)lib ; flags cc LINKFLAGS <runtime-link>static : -Bstatic ; flags cc CFLAGS <debug-symbols>on : -g ; flags cc LINKFLAGS <debug-symbols>on : -g ; flags cc CFLAGS <optimization>off : -xO0 ; flags cc CFLAGS <optimization>speed : -xO4 ; # Other optimizations we might want for WS6U2 # # flags cc CFLAGS <optimization>space : -xO2 ; flags cc CFLAGS <inlining>off : -xinline=no%func ; flags cc CFLAGS <inlining>on : -xinline=%auto ; flags cc CFLAGS <inlining>full : -xinline=%auto ; flags cc CFLAGS <profiling>on : -xpg ; flags cc LINKFLAGS <profiling>on : -xpg ; flags cc DEFINES <define> ; flags cc UNDEFS <undef> ; flags cc HDRS <include> ; flags cc STDHDRS : $(WS6U2_INCLUDE_DIRECTORY) ; flags cc STDLIBPATH : $(WS6U2_STDLIB_DIRECTORY) ; flags cc CFLAGS <shared-linkable>true : -KPIC ; flags cc LINKFLAGS <shared-linkable>true : -KPIC ; flags cc LINKFLAGS <target-type>DLL : -G ; flags cc LIBPATH <library-path> ; flags cc NEEDLIBS <library-file> ; flags cc FINDLIBS <find-library> ; #### Link #### rule Link-action { # This will appear before the import library name when building a DLL, but # will be "multiplied away" otherwise. The --exclude-symbols directive # proved to be neccessary with some versions of Cygwin. ##sik IMPLIB_COMMAND on $(<) = "-Wl,--exclude-symbols,_bss_end__:_bss_start__:_data_end__:_data_start__ -Wl,--out-implib," ; # IMPLIB_COMMAND on $(<) = "-Wl,--out-implib," ; cc-Link-action $(<) : $(>) ; } # for cc, we repeat all libraries so that dependencies are always resolved actions cc-Link-action bind NEEDLIBS { $(WS6U2_BIN_DIRECTORY)CC $(IMPLIB_COMMAND)$(<[2]) $(LINKFLAGS) -o "$(<[1])" -L$(LIBPATH) -L$(STDLIBPATH) "$(>)" "$(NEEDLIBS)" "$(NEEDLIBS)" -l$(FINDLIBS) -znodefs } # older actions # $(WS6U2_BIN_DIRECTORY)dlltool -e $(<[1]:S=:D=:G=) -l "$(<[2])" "$(>)" # $(WS6U2_BIN_DIRECTORY)CC $(LINKFLAGS) -L$(STDLIBPATH) "$(>)" "$(NEEDLIBS)" "$(NEEDLIBS)" -o "$(<[1])" #### Cc ##### rule Cc-action { cc-Cc-action $(<) : $(>) ; } actions cc-Cc-action { $(WS6U2_BIN_DIRECTORY)cc -c -U$(UNDEFS) -D$(DEFINES) $(CFLAGS) -I$(BOOST_ROOT) -I"$(HDRS)" -I"$(STDHDRS)" -o "$(<)" "$(>)" } #### C++ #### rule C++-action { cc-C++-action $(<) : $(>) ; } actions cc-C++-action { $(WS6U2_BIN_DIRECTORY)CC -c -U$(UNDEFS) -D$(DEFINES) $(CFLAGS) $(C++FLAGS) -I$(BOOST_ROOT) -I"$(HDRS)" -I"$(STDHDRS)" -o "$(<)" "$(>)" } #### Archive #### rule Archive-action { cc-Archive-action $(<) : $(>) ; } actions updated together piecemeal cc-Archive-action { ar ru "$(<)" "$(>)" } [Non-text portions of this message have been removed]

2 2

any question
by Thomas Matelich 08 Nov '01

08 Nov '01

In the docs, it uses a function called any_ptr, but I don't think it exists. Could someone please explain? -- Thomas O Matelich Senior Software Designer Zetec, Inc. matelich(a)home.com tmatelich(a)zetec.com

2 1

Boost release 1.25.1 now available
by Beman Dawes 07 Nov '01

07 Nov '01

Boost release 1.25.1 is now available at www.boost.org: * Boost Build System: Continued improvements. * Config Library: Continued refinements. * Graph Library: Final cleanup for upcoming the Boost Graph Library book. * Thread Library: Minor fixes - tests now pass on most Win32 and POSIX systems including Linux and Solaris. Semaphore removed as too error prone. * Function Library: direct support for member function pointers and documentation updates. * Boost-Users mailing list has been created to address topics of interest to users of Boost libraries. * Boost Wiki web added. Provides a place for Boost users to openly discuss and document the use of Boost libraries. It is not officially maintained by Boost developers. This release uses a new ftp program for updating the web site; the program used in the past was prone to get confused about dates, and didn't recover gracefully from network or server errors. Please let me know if you spot any out-of-date web pages on the site. --Beman

1 0

Boost-Users announcement in 1.25.1
by Douglas Gregor 02 Nov '01

02 Nov '01

Hello all, We should announce the existence of the Boost-Users group in the 1.25.1 release of Boost. I've added a bullet to the "Latest News" entry for the release (just check index.htm in the top level of CVS). My two questions are: - Is there a more precise goal statement that should be in the announcement? (i.e., more precise that "address topics of interest to users of Boost libraries", which is intentionally vague) - Should we mention the Wiki? (with the usual not-Boost-moderated disclaimer) Doug

2 2

RE: [Boost-Users] array -v- array_traits
by Jim.Hyslop 25 Oct '01

25 Oct '01

Duane Murphy [SMTP:duanemurphy@mac.com] wrote: > I noticed that array.hpp uses > > #ifndef BOOST_ARRAY_HPP > #define BOOST_ARRAY_HPP > .. > #endif > > and array_traits.hpp uses: > > #if !defined(BOOST_ARRAY_HPP) > #define BOOST_ARRAY_HPP 1 > .. > #endif > > This means that you cant use them both at the same time? The first one wins? > > These are certainly similar classes but they dont seem to collide. > > Is this a subtle bug or intentional? Hi, Duane. I don't know the answer, but I've cc'd this to the Boost developer's list. Someone there should have an answer ;-) -- Jim Hyslop (moderator, Boost-Users group)

3 2

array -v- array_traits
by Duane Murphy 25 Oct '01

25 Oct '01

I noticed that array.hpp uses #ifndef BOOST_ARRAY_HPP #define BOOST_ARRAY_HPP .. #endif and array_traits.hpp uses: #if !defined(BOOST_ARRAY_HPP) #define BOOST_ARRAY_HPP 1 .. #endif This means that you cant use them both at the same time? The first one wins? These are certainly similar classes but they dont seem to collide. Is this a subtle bug or intentional? ..Duane

1 0

FW: [boost] any
by Jim.Hyslop 19 Oct '01

19 Oct '01

rajanikanth jammalamadaka [SMTP:rajanik3@rediffmail.com] wrote: > Hi! Jim, > Could you please tell me why this code is not working: > > #include <iostream> > #include <boost/any.hpp> > using namespace boost; > int main() > { > > any holdsanything ; > holdsanything = 9; > std::cout<<holdsanything<<endl; > holdsanything = std::string("hello"); > std::cout<<holdsanything<<endl; > return 0; > } Hi, Raj You didn't specify what problems you were having. Am I correct in assuming you get a compiler error on the std::cout lines? If so, then the problem is that class 'any' does not have an extraction operator defined. If you change the lines to: std::cout << any_cast<int>(holdsanything)<<std::endl; and std::cout << any_cast<const char *>(holdsanything) << std::endl; it should work properly. -- Jim

2 1

compiling boost on Win2k
by Mike Owens 19 Oct '01

19 Oct '01

I got a this suggestion from Damien Mulligan which worked really well. Use subst to map a drive letter to the path for the MSVC directory then use that drive letter in the VISUALC environment variable. Works like a charm. I should of thought of that. Thanks Damien. Mike > Date: Thu, 18 Oct 2001 08:12:05 -0400 > From: Anthony Kolarik <a.kolarik(a)computer.org> >Subject: RE: Digest Number 11 > >I don't have an answer for your specific JAM/Boost problem but I've seen >similar build problems (using cons)on NT and ended up using temp files to >hold the command line for the linker, rather than try to figure out why the >command lines worked fine on some machines and were too much for others. >You could look into registry settings to extend the buffer size for the >command interpreter but unless you can do it programmatically the temp file >approach would probably be a better. > -- Tony K.

1 0

RE: [Boost-Users] Digest Number 11
by Anthony Kolarik 18 Oct '01

18 Oct '01

I don't have an answer for your specific JAM/Boost problem but I've seen similar build problems (using cons)on NT and ended up using temp files to hold the command line for the linker, rather than try to figure out why the command lines worked fine on some machines and were too much for others. You could look into registry settings to extend the buffer size for the command interpreter but unless you can do it programmatically the temp file approach would probably be a better. -- Tony K. > Date: Wed, 17 Oct 2001 15:47:58 -0700 > From: Mike Owens <mikeo(a)medifor.com> > Subject: Build Problem > > I am trying to build boost on windows 2000 and I get "The > input line is too > long" errors. I have included some sample output from jam. > Please note I > switched to short file names to try and alleviate this > problem. Anyone have > any suggestions on overcoming this problem? > > "C:\Progra~1\micros~2\VC98"\bin\cl /Zm400 -nologo -GX -c > -DNDEBUG > -DBOOST_ > RE_NO_LIB=1 /O2 /GX /GR /MD -I"libs\regex\build" > -I"libs\regex\build\..\..\.. > " -I""C:\Progra~1\micros~2\VC98"\include" > -Fo"libs\regex\build\bin\libboost_re > gex.lib\msvc\release\runtime-link-dynamic\posix_api.obj" > -Tp"libs\regex\build\. > ./src\posix_api.cpp" > > ...failed msvc-C++-action > libs\regex\build\bin\libboost_regex.lib\msvc\release\r > untime-link-dynamic\posix_api.obj ... > msvc-C++-action > libs\regex\build\bin\libboost_regex.lib\msvc\release\runtime-lin > k-dynamic\regex.obj > The input line is too long. >

1 0

Build Problem
by Mike Owens 17 Oct '01

17 Oct '01

I am trying to build boost on windows 2000 and I get "The input line is too long" errors. I have included some sample output from jam. Please note I switched to short file names to try and alleviate this problem. Anyone have any suggestions on overcoming this problem? "C:\Progra~1\micros~2\VC98"\bin\cl /Zm400 -nologo -GX -c -DNDEBUG -DBOOST_ RE_NO_LIB=1 /O2 /GX /GR /MD -I"libs\regex\build" -I"libs\regex\build\..\..\.. " -I""C:\Progra~1\micros~2\VC98"\include" -Fo"libs\regex\build\bin\libboost_re gex.lib\msvc\release\runtime-link-dynamic\posix_api.obj" -Tp"libs\regex\build\. ./src\posix_api.cpp" ...failed msvc-C++-action libs\regex\build\bin\libboost_regex.lib\msvc\release\r untime-link-dynamic\posix_api.obj ... msvc-C++-action libs\regex\build\bin\libboost_regex.lib\msvc\release\runtime-lin k-dynamic\regex.obj The input line is too long.

1 0