April 2008 - Boost-users - lists.preview.boost.org

[Filesystem] operator/ vs operator/=
by Greg Schmid 01 Apr '08

01 Apr '08

I have a string class (tstring) which takes care of narrow <=> wide string conversions, and which provides operators for conversion to std::string and std::wstring: operator std::wstring &(); operator std::string &(); operator const std::wstring &() const; operator const std::string &() const; In Boost 1.33, I could do the following: boost::filesystem::path dir ("/foo"); tstring file ("bar"); boost::filesystem::path p = dir / file; In Boost 1.35, this no longer works; VS8 gives error C2679: binary '/' : no operator found which takes a right-hand operand of type 'const tstring' (or there is no acceptable conversion). However, if I change it to boost::filesystem::path p = dir; p /= file; then all is well. Also, boost::filesystem::path p = dir / std::string(file); boost::filesystem::path p = dir / boost::filesystem::path(file); work just fine. I don't understand the C++ type conversion rules well enough to understand why the conversion to std::string can happen automatically with operator/= but not operator/. I'd rather not have to specify the cast, as the original form of the code is much cleaner. Suggestions? Greg Schmidt (gregs(a)trawna.com) Trawna Publications

2 1

Refactoring
by Jean Vittor 01 Apr '08

01 Apr '08

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi all, I'm trying to write a parser which can write on the input string (mainly insert string at the current position). Does spirit contain any stuff that can help me in that way ? Thanks, Jean -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.7 (GNU/Linux) Comment: Using GnuPG with Mandriva - http://enigmail.mozdev.org iD8DBQFH8fODQ0w7PvnJLt4RAnzxAJ4zbimcHG9G6IvCusqcZ6OXBS+cKQCcDsIL 6nhnvOMeJisGXPNjT/0c5Hs= =lH7Z -----END PGP SIGNATURE-----

2 2

serialization, change from 1.34 to 1.35? (BOOST_CLASS_EXPORT problem)
by Filip Peters 01 Apr '08

01 Apr '08

Hi All, I'm upgrading some code from 1.34 to 1.35. I seem to have problem with BOOST_CLASS_EXPORT macro... according to the help files: http://www.boost.org/doc/libs/1_35_0/libs/serialization/doc/special.html >>>>If no archive headers are included - no code should be instantiated. This will permit BOOST_CLASS_EXPORT to be a permanent part of the my_class.hpp .<<<< my code is structured like that, header file for each class, this also has the export macro, one dll contains the archive includes (#include <boost/archive/binary_iarchive.hpp>), other dll's don't contain the archive code, but do need to use my classes... Now, the other dll's can't compile anymore, I get compile errors like: boost\boost\serialization\type_info_implementation.hpp(48) : error C2027: use of undefined type 'boost::serialization::extended_type_info_impl<T>' some info: boost 1.35, visual studio 2005, sp1 Are there any obvious changes to the serialization code that necessitate changes? thanks, Filip Peters ____________________________________________________________________________________ You rock. That's why Blockbuster's offering you one month of Blockbuster Total Access, No Cost. http://tc.deals.yahoo.com/tc/blockbuster/text5.com

1 0

[GIL] Level 4 warnings on VC8
by John Femiani 01 Apr '08

01 Apr '08

I am trying to compile my code on VC8 at warning level (Level 4). There are several places in boost that produce warnings at this level, but the warnings have obvious simple fixes. (I can send a patch to this list when I have removed enough of them from my working copy in case the library authors care to address them). Along these lines I have a question/comment about GIL. The library uses a GIL_FORCEINLINE macro, presumably to increase performance of some functions. When I compile I get a slew of warning messages with warning "C4714 : function 'function-name' marked as __forceinline not inlined.". I am compiling whith /O2 option (Optimize for speed) and the /Ob2 flag (inline any suitable funciton). My question is: is this a problem for GIL? Obviously the authors have been trying hard to get that function call overhead removed and apparantly vc8 does not play nice with them. I don't know how to address this warning other than to temporarily disable warning 4714, but this is a case where the warning looks like it may be important. The warning indicates that the compiler is unable to inline the functions at these locations: c:\john\hvm\src\boost\boost\gil\algorithm.hpp(251) : copy_with_2d_iterators c:\john\hvm\src\boost\boost\gil\algorithm.hpp(587) : default_construct_range In my case the functions point me to locations in the STL: C:\Program Files\Microsoft Visual Studio 8\VC\include\functional(54) Unfortunately I can not list my code that exhibits the problem here (these warnings come from a large unreleasable project compiled with W3 in the past, and the messages don't help me find which specific parts of my code are instantiating the gil code). I am hoping that the authors will immediately understand the issue (since they used GIL_FORCEINLINE to begin with). Cheers, --John C. Femiani

1 0

[python] can I store reference to c++ shared_ptr instance in python list ?
by Kamil Zubair 01 Apr '08

01 Apr '08

Hi, I'm trying to do this : Python: ptr_list=[] def func(ptr, msg): if "insert"==msg: ptr_list.append(ptr) elif "remove"==msg: ptr_list.remove(ptr) C++: class MyClass{}; int main() { //initialize python exec_file("func.py", global, global); //func.py is the code above object func = global["func"]; shared_ptr<MyClass> mc(new MyClass()); func(mc, "insert"); func(mc, "remove"); } My question is how do I expose my class to python ? I tried class_<MyClass, shared_ptr<MyClass> >("MyClass") but it always fails when I tried to remove an already inserted mc. --------------------------------- You rock. That's why Blockbuster's offering you one month of Blockbuster Total Access, No Cost.

1 0

boost::variant build crashes Visual C++
by a a 01 Apr '08

01 Apr '08

Hello, I have a simple program to test boost::variant. Under Visual C++ 2005, configuration Release (/MD), the build fails with C1060 (out of heap space). The program compiles and runs as expected under Visual C++ 2005 Debug (/MTd), and under Visual C++ 2008 Express, Release and Debug. Below is the code. Any idea what's wrong? Many Thanks, Eric #include <set> #include <string> #include <vector> #include <iostream> #include <fstream> #include <map> #include <boost/variant.hpp> #include <boost/archive/xml_iarchive.hpp> #include <boost/archive/xml_oarchive.hpp> #include <boost/serialization/variant.hpp> #include <boost/serialization/vector.hpp> #include <boost/serialization/map.hpp> struct empty_property_tag { template<class Archive> void serialize(Archive &ar, const unsigned int) {} }; typedef boost::make_recursive_variant<empty_property_tag, std::string, long, double, std::vector<bool>, std::vector<long>, std::vector<double>, std::vector<std::string>, std::vector<std::vector<bool> >, std::vector<std::vector<long> >, std::vector<std::vector<double> >, std::vector<std::vector<std::string> >, std::vector<boost::recursive_variant_>, std::vector<std::vector<boost::recursive_variant_> > >::type property_t; class print_visitor : public boost::static_visitor<> { std::string m_indent; public: print_visitor(const std::string& indent) : m_indent(indent) {} print_visitor(const print_visitor& p) : m_indent(p.m_indent) {} void operator()(const bool& i) const { std::cout << m_indent << "It's an bool: " << i << '\n'; } void operator()(const std::string& s) const { std::cout << m_indent << "It's a std::string: " << s << '\n'; } void operator()(const double& d) const { std::cout << m_indent << "It's a double: " << d << '\n'; } void operator()(const long& d) const { std::cout << m_indent << "It's a long: " << d << '\n'; } void operator()(const empty_property_tag& e) const { std::cout << m_indent << "It's an empty_property_tag.\n"; } void operator()(std::vector<property_t>& v) const { print_visitor p(m_indent + "\t"); std::cout << m_indent << "It is a vector of size: " << v.size() << "\n"; for(std::vector<property_t>::iterator i = v.begin(); i != v.end(); ++i) boost::apply_visitor(p, *i); } void operator()(std::vector<std::vector<property_t> >& v) const { print_visitor p(m_indent + "\t"); std::cout << m_indent << "It is a vector of of vectors size: " << v.size() << "\n"; for(std::vector<std::vector<property_t> >::iterator i = v.begin(); i != v.end(); ++i) operator()(*i); } template <typename T> void operator()(std::vector<T>& v) const { std::cout << m_indent << "It is a vector of size: " << v.size() << "\n"; for(typename std::vector<T>::iterator i = v.begin(); i != v.end(); ++i) std::cout << m_indent << '\t' << *i << '\n'; } template<typename T> void operator()(std::vector<std::vector<T> >& v) const { print_visitor p(m_indent + "\t"); std::cout << m_indent << "It is a vector of size: " << v.size() << "\n"; for(typename std::vector<std::vector<T> >::iterator i = v.begin(); i != v.end(); ++i) operator()<T>(*i); } }; void main() { property_t q = "aaa"; print_visitor p(""); boost::apply_visitor(p, q); }

3 6

Using boost::gregorian (date_time) date with wide strings
by Robin Tucker 01 Apr '08

01 Apr '08

Is it possible to use the boost date_time date class with wide strings? ie. std::wstring toConvert(L"12/12/2008"); date myDate; myDate = from_simple_string(toConvert); What do I have to do to enable this? I see the library is using std::string only. Thanks. (sorry, newbie post). _________________________________________________________________ Win 100’s of Virgin Experience days with BigSnapSearch.com http://www.bigsnapsearch.com

1 0

Re: [Boost-users] Defining Macro's using bjam or
by Stephen Hewitt 01 Apr '08

01 Apr '08

John Maddox wrote: >> Try something like: >> bjam msvc define=_CRT_SECURE_NO_WARNINGS=0 define=_SCL_SECURE_NO_WARNINGS=0 >> define=_SECURE_SCL=0 >> HTH, John. Thanks john. I ended up using the following: * bjam --toolset=msvc-9.0 --build-type=complete define=_CRT_SECURE_NO_WARNINGS define=_SCL_SECURE_NO_WARNINGS define=_SECURE_SCL=0 stage | tee BuildLog.txt* Without "--build-type=complete" only the release multi-threaded libraries were build. This wasn't the case with 1.3.3.1, and there seems to reference to this in the docs. I would like the following in debug builds: *define=_SECURE_SCL=1* And this in release builds: *define=_SECURE_SCL=0* I'm not having much luck. Again, any insights would be greatly appreciated. Steve

2 1

Problems on using ostringstream with boost.format and boost.xpressive
by Olaf Peter 01 Apr '08

01 Apr '08

Hi, the following code did not work as expected: --8<--- #include <iostream> #include <sstream> #include <boost/format.hpp> #include <boost/xpressive/xpressive.hpp> using namespace boost::xpressive; int main() { std::string input( "the answer is\r42" ); const std::string f( "[Input]\n%1%\n|\n" ); std::ostringstream oss; #if 0 oss << boost::format( f ) % input ; #else oss << input; #endif sregex re = as_xpr("\r"); std::string format("\n"); std::ostream_iterator< char > out_iter( std::cout ); regex_replace( out_iter, oss.str().begin(), oss.str().end(), re, format ); } --->8--- The first version displays nothings, the second adds garbage around the text and prints the input string twice. Does exist some limitations or did I ignore some basics? Thanks, Olaf

1 0

Eclipse integration of bjam
by Ruediger Berlich 01 Apr '08

01 Apr '08

Dear all, I am using Eclipse/CDT under Linux as my standard IDE. So far I'm building my projects using Eclipse's own build mechanisms. Is anybody using bjam as a build mechanism in Eclipse/CDT and should I expect problems when making a switch to bjam ? Thanks and Best Regards, Ruediger

12 15