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

[proto-docs] Customing Expression Trees - why three args in operator()?
by Markus Werle 25 Sep '08

25 Sep '08

Hi! [@Eric: The docs look much greater now. Thank you for the hard work. As I can afford time I will go through the docs step by step, give comments again and ask silly questions like before. Is it OK for You that this happens on the public ML? IMHO this may help others, too. Some remarks are rather nitpicking stuff, but still considered to be helpful for first time users of proto] In this post I refer to <http://boost-sandbox.sourceforge.net/libs/proto/doc/html/ boost_proto/users_guide/getting_started/hello_calculator.html> Q1: Probably a typo: Is it "Customing" or "Customizing"? Q2: It is not clear to me why operator() has 3 arguments: double operator()(double a1 = 0, double a2 = 0, double a3 = 0) const { calculator_context ctx; ctx.args.push_back(a1); ctx.args.push_back(a2); ctx.args.push_back(a3); return proto::eval(*this, ctx); } IMHO the whole thing should work even without the 3rd argument. Can you explain? R1: For didactic reasons I would add examples that do not rely on the assumption that the user knows STL by heart, e.g. immediately before <cite> // Use std::transform() and a calculator expression // to calculate percentages given two input sequences: std::transform(a1, a1+4, a2, a3, (_2 - _1) / _2 * 100); </cite> please add the following code: double const result = ((_2 - _1) / _2 * 100)(45.0, 50.0); BOOST_ASSERT(result == (50.0 - 45.0) / 50.0 * 100)); // is this portable? This proposal is in analogy to the code presented in the boost::bind docs at http://www.boost.org/doc/libs/1_36_0/libs/bind/bind.html, which I always found _very_ helpful, since they separate one thing from another. R2: In "Controlling Operator Overloads" it would help a lot if the difference between the previous approach and the new approach was written in *bold*, in order to hinder a scroll proto::domain< proto::generator<calculator>, <BOLD>calculator_grammar</BOLD> > or like this: struct calculator_domain // OLD version had default grammar //: proto::domain< proto::generator<calculator> > : proto::domain< proto::generator<calculator>, calculator_grammar > {}; best regards, Markus

3 4

comparing function objects
by Noah Roberts 25 Sep '08

25 Sep '08

"Function object wrappers can be compared via == or != against any function object that can be stored within the wrapper. If the function object wrapper contains a function object of that type, it will be compared against the given function object (which must be either be EqualityComparable or have an overloaded boost::function_equal)." But how do I compare two boost::functions? This code doesn't work: #include <boost/function.hpp> #include <boost/bind.hpp> double f(double) { return 42.0; } int main() { boost::function<double(double)> f1 = boost::bind(&f, _1); boost::function<double(double)> f2 = boost::bind(&f, _1); //if (f1 == f2) if (boost::function_equal<boost::function<double(double)>, boost::function<double(double)> >(f1,f2)) ; } NOTE: tried to specifically instantiate the function_equal template as attempt3. Previous attempt didn't do that...it also did not work.

2 3

Re: [Boost-users] PDF Documentation
by Dustin Spicuzza 25 Sep '08

25 Sep '08

Dustin Spicuzza wrote: >> One way to test your setup is to cd into libs/math/doc/complex and do a: >> >> bjam pdf >> >> as I know that works (and doesn't take long to build!) >> >> BTW you may find that the FO parser bombs out with out of memory exceptions, >> you can fix this by either: >> >> 1) Changing your JRM settings globally to allow Java to use more memory or >> 2) Edit the batch file that starts the FO processor (Apache FOP or RenderX's >> XEP) to invoke Java with a specific memory allowance (I use -Xmx256m in my >> xep.bat to give XEP 256Mb). >> >> HTH, John. >> I went back and reinstalled everything again, I must have installed it incorrectly the first time because I hadn't been getting OutOfMemory errors from FOP before, but now I am. Looks like increasing the memory limit to 2GB allowed FOP to work correctly for the one that I wanted to build for (instrusive), 512MB failed. It appears that FOP 0.94 has a problem with code sections that are longer than a page, as it overflows the bottom and cuts it off. Go figure. See http://www.virtualroadside.com/stuff/intrusive.pdf , page 33 for an example of what I'm talking about. :) Dustin

1 0

[serialization] conflicting declaration (BOOST_CLASS_EXPORT)
by jens weller 25 Sep '08

25 Sep '08

Hi, I just had this similar error today, but it took me quite a few minutes to build a working "example". I'll start off with the error(s): L:\project\serializationtest\Mixed.hpp:27: error: conflicting declaration 'const boost::archive::detail::guid_initializer<Mixed>&<unnamed>::boost_serialization_guid_initializer_27' L:\project\serializationtest\Child.hpp:27: error: '<unnamed>::boost_serialization_guid_initializer_27' has a previous declaration as `const boost::archive::detail::guid_initializer<Child>&<unnamed>::boost_serialization_guid_initializer_27' L:\project\serializationtest\Mixed.hpp:27: error: type mismatch with previous external decl of `const boost::archive::detail::guid_initializer<Mixed>&<unnamed>::boost_serialization_guid_initializer_27' L:\project\serializationtest\Child.hpp:27: error: previous external decl of `const boost::archive::detail::guid_initializer<Child>&<unnamed>::boost_serialization_guid_initializer_27' L:\project\serializationtest\Mixed.hpp:27: error: declaration of `const boost::archive::detail::guid_initializer<Mixed>&<unnamed>::boost_serialization_guid_initializer_27' L:\project\serializationtest\Child.hpp:27: error: conflicts with previous declaration `const boost::archive::detail::guid_initializer<Child>&<unnamed>::boost_serialization_guid_initializer_27' :: === Build finished: 6 errors, 0 warnings === This is my class structure: Base Child : Base Mixed : Base or #include "Base.hpp" is actually enough. Easy, but maybe its already a hint to the problem. The "original" Class structure in my app is far more complex, so I just will refer to my little example, as I guess its the same source some how. I tend to believe that the problem resides somewhere in the Macro BOOST_CLASS_EXPORT, as the compiler points to this. I'm working on Windows XP, with MinGW and the latest (Monday) SVN Build of boost. (as serialization is in 1.36 partly broken afaik/imho). I'll attach my code. So, how to resolve this? regards Jens Weller

7 16

Problems using thread and asio libraries together with Boost 1.36.0 on Cygwin
by Clarke, Carus V 25 Sep '08

25 Sep '08

> I have been using Boost 1.34.1 for quite awhile for a Windows > application that uses both the Boost thread and asio libraries. The > application builds successfully using gcc 3.4.4 on Cygwin and both > MSVC 8.0 and 9.0. Now I am trying to port the application to Boost > 1.36.0 and have had some problems with the gcc compiler on Cygwin. > The following (or similar) errors have occured when compiling any file > that includes a thread library header: > > ... > g++ -o build/ftds/out/OutputManager.o -c -g -mwin32 -mthreads -Wall > -ansi -D_REENTRANT -D__STRICT_ANSI__ -D_WIN32_WINNT=0x0501 > -D__USE_W32_SOCKETS -DWIN32_LEAN_AND_MEAN -DRWDEBUG=1 > -DRW_MULTI_THREAD -DZAF_BOOL_DEFINED -DZAF_POSIX -DNOMINMAX=1 > -Iinclude -Ic:/cygwin/home/cvc7986/work/boost_1_36_0 > -Ic:/rwave/workspaces/winxp/gcc/15s "-Ic:/Program Files/IVI > Foundation/VISA/winnt/include" "-Ic:/Program Files/Agilent/IO > Libraries Suite/include" -Ic:/zaf530/include > source/ftds/out/OutputManager.cpp > In file included from /usr/include/cygwin/sys_time.h:13, > from /usr/include/sys/time.h:27, > from > c:/cygwin/home/cvc7986/work/boost_1_36_0/boost/date_time/c_time.hpp:28 > , > from > c:/cygwin/home/cvc7986/work/boost_1_36_0/boost/date_time/microsec_time > _clock.hpp:18, > from > c:/cygwin/home/cvc7986/work/boost_1_36_0/boost/thread/thread_time.hpp: > 9, > from > c:/cygwin/home/cvc7986/work/boost_1_36_0/boost/thread/pthread/timespec > .hpp:9, > from > c:/cygwin/home/cvc7986/work/boost_1_36_0/boost/thread/pthread/conditio > n_variable.hpp:8, > from > c:/cygwin/home/cvc7986/work/boost_1_36_0/boost/thread/condition_variab > le.hpp:16, > from > c:/cygwin/home/cvc7986/work/boost_1_36_0/boost/thread/condition.hpp:9, > from include/cds/Server.h:29, > from source/ftds/out/OutputManager.cpp:27: > /usr/include/sys/select.h:31: error: `fd_set' has not been declared > /usr/include/sys/select.h:31: error: `fd_set' has not been declared > /usr/include/sys/select.h:31: error: `fd_set' has not been declared > /usr/include/sys/select.h:33: error: `fd_set' has not been declared > /usr/include/sys/select.h:33: error: `fd_set' has not been declared > /usr/include/sys/select.h:33: error: `fd_set' has not been declared > In file included from > c:/cygwin/home/cvc7986/work/boost_1_36_0/boost/asio/detail/socket_type > s.hpp:75, > from > c:/cygwin/home/cvc7986/work/boost_1_36_0/boost/asio/detail/win_iocp_io > _service_fwd.hpp:24, > from > c:/cygwin/home/cvc7986/work/boost_1_36_0/boost/asio/io_service.hpp:37, > from > c:/cygwin/home/cvc7986/work/boost_1_36_0/boost/asio/basic_io_object.hp > p:20, > from > c:/cygwin/home/cvc7986/work/boost_1_36_0/boost/asio/basic_datagram_soc > ket.hpp:25, > from > c:/cygwin/home/cvc7986/work/boost_1_36_0/boost/asio.hpp:20, > from include/cds/srvr/ConfigurationServer.h:26, > from include/cds/Server.h:41, > from source/ftds/out/OutputManager.cpp:27: > /usr/lib/gcc/i686-pc-cygwin/3.4.4/../../../../include/w32api/winsock2. > h:109: error: redefinition of `struct timeval' > /usr/include/sys/time.h:16: error: previous definition of `struct > timeval' > /usr/lib/gcc/i686-pc-cygwin/3.4.4/../../../../include/w32api/winsock2. > h:632: error: declaration of C function `int select(int, fd_set*, > fd_set*, fd_set*, const timeval*)' conflicts with > /usr/include/sys/select.h:31: error: previous declaration `int > select(int, int*, int*, int*, timeval*)' here > scons: *** [build/ftds/out/OutputManager.o] Error 1 > ... > > I was able to determine that the problem was caused by the inclusion > of sys/time.h, which includes cygwin/sys_time.h, which then includes > sys/select.h. The definition of struct fd_set was prevented in > sys/types.h because macro __USE_W32_SOCKETS had been defined, as is > required for using the asio library on Cygwin. > > My temporary solution to the problem was to disable the definition of > the BOOST_HAS_GETTIMEOFDAY macro in the Boost header file > boost/config/platform/cygwin.hpp and to define BOOST_HAS_FTIME macro > instead: > > $ diff ~/work/boost_1_36_0/boost/config/platform/cygwin.hpp > boost/config/platform/cygwin.hpp > 27c27,32 > < # define BOOST_HAS_GETTIMEOFDAY > --- > > // temp: # define BOOST_HAS_GETTIMEOFDAY > > # if !defined(__USE_W32_SOCKETS) // temp > > # define BOOST_HAS_GETTIMEOFDAY // temp > > # else // temp > > # define BOOST_HAS_FTIME // temp > > # endif // temp > > This has solved the problem for me as of now, but it would be good to > know if there is a better way to solve this. I was not able to find > any references to this problem in any of the Boost lists. > Thanks in advance, > Carus V. (Bud) Clarke > carus.v.clarke(a)boeing.com >

2 1

threads in a DLL with callbacks
by Nicholas Cain 25 Sep '08

25 Sep '08

Hi, I've created a component that's distributed as a DLL. The user loads the DLL, calls various methods on it and gets some data back asyncronously via callbacks. Within the component there are several threads (using boost threads of course) any of which might call the callbacks. In all our sample code demonstrating how to create callbacks and call the methods we use boost mutexes to protect the callback functions and this works just fine. Some of our clients however don't want to dip their toes into boost, and want to use windows methods to protect the callbacks. For instance so far they've tried various methods such as CComAutoCriticalSection and EnterCriticalSection. They still have a crash which they're attributing to my DLL. I keep pushing them to at least try the boost mutex as in our sample code, but no luck so far. My question is; should the windows critical section protect the callbacks just as well as the boost mutexes, or does it require a boost mutex protection? (or do you require a code example to know the answer to this?) Nik

1 0

Is there a boost equivalent of eastl::fixed_vector?
by dariomt＠gmail.com 25 Sep '08

25 Sep '08

In EASTL http://www.open-std.org/jtc1/sc22/wg21/docs/papers/2007/n2271.htmlthere is something like SSO (Small String Optimization) applied to std::vector (and other STL containers) [begin quote] "Fixed containers are fixed-size containers with their memory stored right within the container itself. Fixed containers allocate no dynamic memory and their memory tends to be cache-friendly due to its contiguity and proximity to the container's housekeeping data. The user declares the max container size as a template parameter, and can also specify that if the container overflows that an auxiliary dynamic allocator is used. template <typename T, size_t nodeCount, bool enableOverflow = true, typename Allocator overflowAllocator = EASTLAllocator> class fixed_vector { ... }; [end quote] I have tried a similar approach using some sort of fixed_buffer_with_overflow allocator, with no success. I'm not sure SSO is even possible just using an allocator with the standard containers, because I always end up with stateful allocators. So my question is if there is something similar already in boost. TIA

5 5

bjam JAMSHELL
by Gerhard Lipp 25 Sep '08

25 Sep '08

hello! i would like to specify a different shell for bjam on windows xp (namely sh.exe). i tried to place JAMSHELL = C:\\Programme\\A-Toolchain\\bin\\sh.exe -c % ; in my Jamroot file, though without success, the windows cmd "shell" gets called. background: i have to invoke a shell script (wx-config) to get my cxxflags like that local wx_cxxflags = [ SHELL "wx-config --cxxflags | tr -d '\\n'" ] ; regards, gerhard

2 1

make check results version 1.36.0
by mess-mate 25 Sep '08

25 Sep '08

Hi, this is what i have after a 'make check' when compiling boost: .../... ...failed updating 13 targets... ...skipped 18 targets... ...updated 19995 targets... Some Boost regression tests failed. This is normal for many compilers. ..../.... Is there any reason to not continue the installation ? best regards

1 0

[logging]Redirecting process output to log
by Vladimir Kiryazev 25 Sep '08

25 Sep '08

Hello, I am using the logging library by John Torjo. I create a process in my program and need to redirect its output to log. How do I do this? Thanks.

4 4