May 2009 - Boost-users - lists.preview.boost.org

[BGL] Trying to use subgraph
by Lindley M French 19 May '09

19 May '09

I've been trying to work out how to use the subgraph adapter. I've come to realize that in order to use it, my primary graph must *itself* be an object of the subgraph type (unintuitive). However, I'm getting a static assertion error that edge_index is an unknown property. I had *thought* that an adjacency_list<vecS,vecS> would not have that issue, but apparently I was wrong. The page here: http://www.boost.org/doc/libs/1_39_0/libs/graph/doc/subgraph.html says "we add an edge index property to the adjacency list", but the code which follows demonstrates no such thing. The distinction between bundled properties and internal properties is still a bit mysterious to me. Can anyone shed some light on this?

2 1

Serialization Archive Incompatibility between Boost Versions
by David Graf 19 May '09

19 May '09

Hello How is it possible to load serialized objects with a newer boost version than they have been saved with. I have installed boost version 1.36 and I cannot read archives that have been save with boost 1.35. This is not possible because boost 1.35 puts 'serialization::archive 4' into the files but boost 1.36 needs the archive version 5. I found some information in the documentation that is must be possible to read old archives. But I am not able to do it. I always get exceptions. Thank you in advance for help! David

2 8

Boost up for SourceForge Community Choice Award
by David Abrahams 19 May '09

19 May '09

Cast your vote here: http://tinyurl.com/boost-cca09 Cheers, -- Dave Abrahams BoostPro Computing http://www.boostpro.com

2 1

[signals2] connection blocking
by Bryan Green 19 May '09

19 May '09

Hi, I've starting porting some code from signals to signals2, and I've run into a problem because of the removal of the 'block' and 'unblock' methods from a connection object. I understand the rational for using the 'shared_connection_block' object, but it doesn't really apply in my case. I have a GUI widget which receives events via a signal, unless it is in an "inactive" state, during which time events are blocked: class widget { private: boost::signals::connection conn; ... public: void toggle_active_state() { ... if (inactive) conn.block(); else conn.unblock(); } }; It is obviously awkward to be forced to convert this code to use an RAII object, and I'd rather not do a full connect/disconnect for this state change. What do you think? Should I be thinking about this differently? Or can a good case be made for bringing back the block()/unblock() methods? Thanks, -bryan

4 7

[Regex] Source-only link error in MSVC
by dtexg＠hush.com 19 May '09

19 May '09

Hi, I'm trying to include the Boost Regex source in a Visual C++ project, but I get the following linker error: (indented for legibility) error LNK2019: unresolved external symbol "private: class boost::basic_regex< char, struct boost::regex_traits< char, class boost::w32_regex_traits<char> > > & __thiscall boost::basic_regex< char, struct boost::regex_traits< char, class boost::w32_regex_traits<char> > >:: do_assign( char const *, char const *, unsigned int)" (?do_assign@?$basic_regex@DU?$regex_traits@DV?$w32_regex_traits@D@bo ost@@@boost@@@boost@@AAEAAV12@PBD0I@Z) referenced in function "public: class boost::basic_regex< char, struct boost::regex_traits< char, class boost::w32_regex_traits<char> > > & __thiscall boost::basic_regex< char, struct boost::regex_traits< char, class boost::w32_regex_traits<char> > >:: assign( char const *, char const *, unsigned int)" (?assign@?$basic_regex@DU?$regex_traits@DV?$w32_regex_traits@D@boost @@@boost@@@boost@@QAEAAV12@PBD0I@Z) Here are the steps I'm taking: I downloaded the latest Boost (boost_1_39_0.zip), and extracted it all into its own directory; I've tried extracting with both the built-in Windows Zip Folder support and the latest WinRAR. I did not do any making (nmake) or building on the Boost source -- after extracting, I don't touch this directory directly at all. In both Visual Studio 2005 and Visual C++ 2008 Express Edition (http://www.microsoft.com/express/vc/), I created a new empty console project, and I added a new .cpp file to the project. I went to Project -> Properties, Configuration Properties -> C/C++ - > General: Additional Include Directories, and (for the "Active(Debug)" configuration) added the full path of the directory containing the source, so that "boost" is a subdirectory in the selected directory. I wrote the following code: #define BOOST_REGEX_SOURCE #include <boost/regex.hpp> int main() { boost::regex r("a"); return 0; } Everything compiles, but linking fails with the above error. The same error results if I try this: boost::regex * p = new boost::regex("a"); Or if I use boost::wregex, or boost::basic_regex<char,boost::regex_traits<char,boost::w32_regex_tr aits<char>>>, etc. Also, the outcome is the same whether I #define BOOST_REGEX_SOURCE, BOOST_REGEX_NO_LIB, or BOOST_ALL_NO_LIB, and whether I define them in the .cpp or in Project -> Properties, Configuration Properties -> C/C++ -> Preprocessor: Preprocessor Definitions, and whether I #include <boost/regex.hpp> or <boost/regex/v4/regex.hpp>, and whether I use Debug or Release configuration. If I put a #error by the do_assign implementation in boost/regex/v4/basic_regex.hpp, it gets hit, so the implementation isn't being #if'ed away somehow. It will build successfully if, in the .cpp file in which main() resides, I write an implementation and replace all the template arguments, but besides being vulgar, this kludge also fails (with compiler errors) if I try to copy-and-paste the full implementation from basic_regex.hpp, so it's untenable as a workaround. In a post from 2006, John wrote, "if you are doing something, umm, strange, and you can't track down the cause of the error, you can always just add the regex source directly to your application," but I've not seen this successfully demonstrated in Visual C++. I'm just looking for an absolutely parsimonious way to add Boost Regex to a VC++ project. /Eric -- Free information on Digital Photography. Click Now! http://tagline.hushmail.com/fc/BLSrjkqZZHWE3MuglDmDyK5F8gPPnxz3V9yL7kN5dsQj…

2 1

PropertyTree's XML parsers survey
by Tan, Tom (Shanghai) 19 May '09

19 May '09

>From: Nat Goodspeed <nat(a)lindenlab.com> >> - I often embed some comment nodes inside the xml to facilitate the >> manual edit. E.g, enumeration of accepted value some fields. I wish >> boost.serialize ignores them when parsing, > >That seems plausible to me... > >> and embed them when saving (is this possible). >This request feels more difficult. Boost.Serialize is designed to read >from a file into /your/ data structures, and/or write those data >structures to a file. XML is only one of several possible archive >formats for such a file. > >Where in your own data structures would you recommend the Serialize >library store any XML comments it may encounter? > >Supposing there were an appropriate place to collect such comments, how >would you write them again in something like their original file location? Just for the fun of discussion, how about allowing only one comment node only and fixing the position to be the end of the file, like this: First, add an optional parameter to signature. template<class Archive> void serialize(Archive & ar, const unsigned int version, basic_string<Ch> comment = basic_string<Ch>()) Then, toward the end of the file, append this node <!== {comment string here} ==> This does not break the compatibility of the existing version, and that the xml archive version already includes more info(i.e. names in additional to values) than other variants(e.g. bins), it does not actually violate the current philosophy much. >>Instead of incrementally enhancing a minimal XML parser, wouldn't it >>make more sense to jump right to an existing full-blown parser >>implementation? Ah, while this is too limited, a full-blown is too heavy.

1 0

[iostreams][zip]
by Andy Schweitzer 18 May '09

18 May '09

Can someone provide a command line that builds iostreams with zlib and bz2 support? I tried the following but my program still fails to link to libboost_zlib. I did find a boost_zlib-vc90-mt-1_38.dll and a boost_zlib-vc90-mt-1_38.lib, but no libboost_zlib-vc90-mt-gd-1_38.lib bjam toolset=msvc-9.0 -sHAS_ZLIB=1 -sHAS_BZIP2=1 -sNO_COMPRESSION=0 -sNO_BZIP2=0 -s -sNO_ZLIB=0 -sZLIB_INCLUDE=C:/nan/compression/zlib/zlib123-dll/include -sZLIB_LIBPATH=C:/nan/compression/zlib/zlib123-dll/lib -sZLIB_SOURCE=C:/nan/compression/zlib/zlib-1.2.3 -sBZIP2_INCLUDE=C:/nan/compression/bz2/bzip2-1.0.5 -sBZIP2_LIBPATH=C:/nan/compression/bz2/bzip2-1.0.5 -sBZIP2_SOURCE=C:/nan/compression/bz2/bzip2-1.0.5 Thanks Andy

3 7

extending Jam
by gtsml owevwr 18 May '09

18 May '09

Hi, I'm trying to build boost on the iphone OS 3.0 and Jam is telling me: error: "iphone-3.0" is not a known value of feature <macosx-version> error: legal values: "10.6" "10.5" "10.4" "10.3" "10.2" "10.1" "iphone-2.3" "iphonesim-2.3" "iphone-2.2" "iphonesim-2.2" "iphone-2.1" "iphonesim-2.1" "iphone-2.0" "iphonesim-2.0" "iphone-1.x" I'd like to add this OS to Jam, is there a tutorial (or anything helpful) to accomplish such task? Thanks, Frank

2 5

Boost on Eclipse Ganymede 3.4.2
by Matheus Araújo Aguiar 18 May '09

18 May '09

Hello everyone, I'm trying to use boost lib asio in a project on eclipse (running on ubuntu 8.10). So far i'm getting this errors when trying to build the project. '/home/matheus/boost_1_39_0' is where i've unpacked boost lib downloaded from sourceforge.net . I've also set this directory to the include path in eclipse. So, any help? Description Resource Path Location Type /home/matheus/boost_1_39_0/boost/asio/detail/posix_signal_blocker.hpp undefined reference to `pthread_sigmask' Proxy_Server line 48 C/C++ Problem /home/matheus/boost_1_39_0/boost/asio/detail/posix_signal_blocker.hpp undefined reference to `pthread_sigmask' Proxy_Server line 55 C/C++ Problem /home/matheus/boost_1_39_0/boost/asio/detail/posix_thread.hpp undefined reference to `pthread_detach' Proxy_Server line 69 C/C++ Problem /home/matheus/boost_1_39_0/boost/asio/detail/posix_thread.hpp undefined reference to `pthread_join' Proxy_Server line 77 C/C++ Problem /home/matheus/boost_1_39_0/boost/asio/detail/posix_tss_ptr.hpp undefined reference to `pthread_getspecific' Proxy_Server line 67 C/C++ Problem /home/matheus/boost_1_39_0/boost/asio/detail/posix_tss_ptr.hpp undefined reference to `pthread_key_create' Proxy_Server line 47 C/C++ Problem /home/matheus/boost_1_39_0/boost/asio/detail/posix_tss_ptr.hpp undefined reference to `pthread_key_delete' Proxy_Server line 61 C/C++ Problem /home/matheus/boost_1_39_0/boost/asio/detail/posix_tss_ptr.hpp undefined reference to `pthread_setspecific' Proxy_Server line 73 C/C++ Problem /home/matheus/boost_1_39_0/boost/asio/error.hpp undefined reference to `boost::system::get_system_category()' Proxy_Server line 218 C/C++ Problem /home/matheus/boost_1_39_0/boost/system/error_code.hpp undefined reference to `boost::system::get_generic_category()' Proxy_Server line 206 C/C++ Problem /home/matheus/boost_1_39_0/boost/system/error_code.hpp undefined reference to `boost::system::get_generic_category()' Proxy_Server line 211 C/C++ Problem /home/matheus/boost_1_39_0/boost/system/error_code.hpp undefined reference to `boost::system::get_generic_category()' Proxy_Server line 212 C/C++ Problem /home/matheus/boost_1_39_0/boost/system/error_code.hpp undefined reference to `boost::system::get_system_category()' Proxy_Server line 205 C/C++ Problem /home/matheus/boost_1_39_0/boost/system/error_code.hpp undefined reference to `boost::system::get_system_category()' Proxy_Server line 213 C/C++ Problem /home/matheus/boost_1_39_0/boost/system/error_code.hpp undefined reference to `boost::system::get_system_category()' Proxy_Server line 312 C/C++ Problem Thanx and best regards, -- Matheus Araújo Aguiar Computer Scientist matheus.pit(a)gmail.com

3 4

*ignore* [interprocess] managed_mapped_file under cygwin
by Bob Walters 18 May '09

18 May '09

Please disregard that request for help, I had forgotten about a 'experimental' version of boost/interprocess/detail/os_file_functions.hpp (with support for offsets in the file locks) from the 1_38 version, which was being inadvertently included by the new 1_39 code, which I naturally upgraded to at the same time I did the OS change. - Bob Walters

1 0