January 2006 - Boost-users - lists.preview.boost.org

Re: [Boost-users] Spawning a thread on a member function
by Sliwa, Przemyslaw (London) 30 Jan '06

30 Jan '06

Can I spawn the thread within another member function? I believe it should be not a problem. Thanks Kind regards Pshemek Przemyslaw Sliwa Global Foreign Exchange FX Risk Engine Development Merrill Lynch ( +44 (0)20 7995 4182 ) +44 (0)79 6325 2319 * Email: przemyslaw_sliwa(a)ml.com -----Original Message----- From: boost-users-bounces(a)lists.boost.org [mailto:boost-users-bounces@lists.boost.org] On Behalf Of Patrick Sent: 27 January 2006 23:03 To: boost-users(a)lists.boost.org Subject: Re: [Boost-users] Spawning a thread on a member function You are correct the "boost::ref(*this)" is not needed. Simple(boost::bind(&Simple::run, this)) {} -----Original Message----- From: boost-users-bounces(a)lists.boost.org [mailto:boost-users-bounces@lists.boost.org] On Behalf Of Christian Henning Sent: Friday, January 27, 2006 4:47 PM To: boost-users(a)lists.boost.org Subject: Re: [Boost-users] Spawning a thread on a member function This is even simpler. Is there is reason for "boost::ref(*this)" ? class Simple { Simple() : th( boost::bind( run, this )) {} void run() { for(;;){ //Thread's work goes here. } } private: boost::thread th; }; _______________________________________________ Boost-users mailing list Boost-users(a)lists.boost.org http://lists.boost.org/mailman/listinfo.cgi/boost-users _______________________________________________ Boost-users mailing list Boost-users(a)lists.boost.org http://lists.boost.org/mailman/listinfo.cgi/boost-users -------------------------------------------------------- If you are not an intended recipient of this e-mail, please notify the sender, delete it and do not read, act upon, print, disclose, copy, retain or redistribute it. Click here for important additional terms relating to this e-mail. http://www.ml.com/email_terms/ --------------------------------------------------------

2 1

problem with include files
by Frank Pleyer 30 Jan '06

30 Jan '06

Hi, Im new to boost and after building the libraries to c:\boost, I wanted to go through the [String Algorithm First Example]. Im using visual c++ 2005 with windows xp SP2 and when I tried to build my project I get the following error: fatal error C1083: Datei (Include) kann nicht geöffnet werden: "boost/algorithm/string.hpp" corrosponding to that line of code: #include <boost/algorithm/string.hpp> All # include calls have a unix like path description and not a common windows path like : c:\boost\include\algorith\string.hpp. That seems to be a general problem with all #include calls in all boost *.hpp files. I have included the the path[c:\boost] and [c:\boost\include] to my VC++ INCLUDE Path. Can somebody please help. -Frank

2 1

Re: [Boost-users] Boost shared pointers and MSVC++ 6.0
by Sohail Somani 30 Jan '06

30 Jan '06

> My questions: > > * Is this normal ? What have other people experienced ? Normal. If you are sharing heaps, data structures across DLLs, you should be sure that you link with the same runtimes. > * Will this go away if I upgrade to MSVC++ 7 ? Maybe not if you're using C++.NET?

1 0

[random] comments on speed?
by Sanyi 30 Jan '06

30 Jan '06

Sorry if double posted... Can anyone who have tried Boost's `random' library have any thoughts on its speed? I looked into its binomial variate function and apparently it actualy flips a (biased) coin countless times to generate it. I am not an expert in this, but this seems awfully slow for high n's. (Not to mention that I could have done it myself.) AFAIK there are fast and good algorithms/approximations for generating some non-uniform variates. Are any of these at least planned to be included? <OT> I've found some papers on fast binomial generators, but that takes some serious time and effort to translate into working code. As I don't want to reinvent the wheel, if you know of any publicly available libraries for it, let me know. </OT> Thanks, Sanyi

3 3

boost::iostream - no way to set failbit or badbit in stream from device?
by Oliver.Kowalke＠infineon.com 30 Jan '06

30 Jan '06

Hallo, how can I set failbit or badbit in boost::iostream::stream if I detect some error in boost::iostream::device::read/write? Is it possible without throwing exceptions? Example: in the write method of a device which models a socket connection an EPIPE error (peer has closed socket) was detected. How can I set the boost::iostream::stream (which uses the device) to be not useable? Regards, Oliver

1 0

Re: [Boost-users] boost::iostream - returning EOF in device::writeand set failbit
by Oliver.Kowalke＠infineon.com 30 Jan '06

30 Jan '06

Hello Petro, >> would the eofbit be set if I return EOF in device::write? >> How can I set failbit from device? >Does it make sense for write to return EOF? >EOF means "no more data". >But writing is inserting data, not requesting it. If you write to a socket and the peer has already closed the socket you get an EPIPE error. How would you propagate this to your application? Oliver

1 0

[boost][serialization] How to implement serialization classes in two DLLs
by jf332-3 29 Jan '06

29 Jan '06

Hi, After few hour on the subject, Im would like to have some explanations & recommendations one the boost serialization library. To explain, my problem, you will find hereafter a VC++ 7.1 project that demonstrates it. Problem Description We have involved in a large project with for example two DLLs libraries: LibA and LibB. These libraries define & implement one class each other, lets say libA::A and libB::B. These two classes can be serialized through boost serialization engine. The interfaces for the two classes are the same: ---------------------- struct LIBA_API A { std::vector<double> my_vector ; private: friend class boost::serialization::access; template<class Archive> void save(Archive & ar, const unsigned int version) const ; template<class Archive> void load(Archive & ar, const unsigned int version) ; BOOST_SERIALIZATION_SPLIT_MEMBER() }; LIBA_API void save(const libA::A* pObj, std::ostream& os) ; LIBA_API void load(libA::A* &pObj, std::istream& is) ; }; // Namespace libA ---------------------- The two free functions save() and load() are defined to be able to serialize the class libA::A (respectively the class libB::B). When, we try to use these interfaces in an EXE application: ---------------------- #include <LibA/A.h> #include <LibB/B.h> #include <boost/filesystem/path.hpp> #include <boost/filesystem/fstream.hpp> #pragma comment(lib, "liba_d.lib") #pragma comment(lib, "libb_d.lib") int main(int argc, char* argv[]) { libA::A a; libA::save(&a, boost::filesystem::ofstream("a.xml")); libB::B b; libB::save(&b, boost::filesystem::ofstream("b.xml")); // -> Assert in debug mode return 0; } ---------------------- The figure hereafter describes the relationships between LibA, LibB and EXE. ------------ ------------ | LibA | | LibB | ------------ ------------ \ / \ / Exe Application We have to face to an assert in debug mode generate by the file boost::serialization::extended_type_info.cpp, line#71. This assert happens during the instance serialization of class libB:B. Recently, there are some posts on this subject, but I would like to known how I can deal with this assert? Is it possible to avoid it by a code organisation? To have a concrete answer, I have joined my complete VC++7.1 project. Thanks in advance for your help. Development Platform WindowsXP VC++ 7.1 Boost 1.33.1 Marc Viala mailto:mviala@acticm.com

1 0

Re: [Boost-users] boost::serialisation - portable_binary_oarchivenotpart of library
by Oliver.Kowalke＠infineon.com 28 Jan '06

28 Jan '06

Hello Robert, I think a 'finished' protable_binary_oarchive/ protable_binary_iarchive would be very useful for network applications (passing message structures -> header, body, trailer). I'm not familiar with boost::serialization (my first look into it) - I see some specializations of load for integral types an a templated member function for all others. Serialisation for std::strings and chars are supported in a portable manner? Only floating point types have to be implemented for portability? Oliver >I made it as an example on how to extend the library by deriving from an >existing >archive. To "finish" it would require implementing serialization of >floating point >types in a portable manner and this turns out to be a non-trivial task. >Implementing >this would be too much work and would sort of compromise illustrative value >of the example.

3 3

Re: [Boost-users] boost::serialisation -Non-default Constructors
by Merrill Cornish 28 Jan '06

28 Jan '06

Yes. Each class you want to serialize needs a default constructor in addition to whatever other constructors your application needs. Merrill

2 1

::spirit, token_node_d
by elviin 28 Jan '06

28 Jan '06

Before I'll investigate more I'd like to know if anybody knows why it is not allowed a rule object inside this directive: token_node_d.

2 2