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

Link errors with MSVC 9 application using Boost date_time on DEBUG build
by Tom Aylesworth 17 Mar '09

17 Mar '09

I have a very simple app that I'm trying to link with the Boost date_time library. It links and runs fine with I build against the RELEASE configuration. But I get the following link errors when when I use the DEBUG configuration: 2>TimeUtil.obj : error LNK2019: unresolved external symbol "public: char const * __cdecl boost::gregorian::greg_month::as_long_string(void)const " (?as_long_string@greg_month@gregorian@boost@@QBAPBDXZ) referenced in function "public: static class stlpd_std::basic_ostream<wchar_t,class stlpd_std::char_traits<wchar_t> > & __cdecl boost::date_time::month_formatter<class boost::gregorian::greg_month,class boost::date_time::iso_extended_format<wchar_t>,wchar_t>::format_month(class boost::gregorian::greg_month const &,class stlpd_std::basic_ostream<wchar_t,class stlpd_std::char_traits<wchar_t> > &)" (?format_month@?$month_formatter@Vgreg_month@gregorian@boost@@V?$iso_extended_format@_W@date_time@3@_W@date_time@boost@@SAAAV?$basic_ostream@_WV?$char_traits@_W@stlpd_std@@@stlpd_std@@ABVgreg_month@gregorian@3@AAV45@@Z) 2>TimeUtil.obj : error LNK2019: unresolved external symbol "public: char const * __cdecl boost::gregorian::greg_month::as_short_string(void)const " (?as_short_string@greg_month@gregorian@boost@@QBAPBDXZ) referenced in function "public: static class stlpd_std::basic_ostream<wchar_t,class stlpd_std::char_traits<wchar_t> > & __cdecl boost::date_time::month_formatter<class boost::gregorian::greg_month,class boost::date_time::iso_extended_format<wchar_t>,wchar_t>::format_month(class boost::gregorian::greg_month const &,class stlpd_std::basic_ostream<wchar_t,class stlpd_std::char_traits<wchar_t> > &)" (?format_month@?$month_formatter@Vgreg_month@gregorian@boost@@V?$iso_extended_format@_W@date_time@3@_W@date_time@boost@@SAAAV?$basic_ostream@_WV?$char_traits@_W@stlpd_std@@@stlpd_std@@ABVgreg_month@gregorian@3@AAV45@@Z) It is linking against "libboost_date_time-vc90-mt-sgdp-1_38.lib". Using undname, I can see that the library is using the __thiscall convention instead of __cdecl. Is this a problem in my configuration for building the Boost date_time library, or a problem in my app's build configuration? Thanks in advance for any help! Tom Aylesworth Reality Mobile LLC

2 6

Problems upgrading from 1_34_1 to 1_38_0 in VS 2005 Express
by Roman Stepanenko 17 Mar '09

17 Mar '09

Hi, The only thing i changed in my project was new location for headers and libraries (from 1_34_1 to 1_38_0). Now I am getting these messages for a lot of classes: 1>pipeline.cpp 1>c:\program files\microsoft visual studio 8\vc\include\complex(356) : error C2589: '(' : illegal token on right side of '::' 1>c:\program files\microsoft visual studio 8\vc\include\complex(356) : error C2059: syntax error : '::' 1>c:\program files\microsoft visual studio 8\vc\include\complex(376) : error C2589: '(' : illegal token on right side of '::' 1>c:\program files\microsoft visual studio 8\vc\include\complex(376) : error C2059: syntax error : '::' 1>c:\program files\microsoft visual studio 8\vc\include\complex(381) : error C2589: '(' : illegal token on right side of '::' 1>c:\program files\microsoft visual studio 8\vc\include\complex(381) : error C2059: syntax error : '::' 1>c:\program files\microsoft visual studio 8\vc\include\complex(386) : error C2589: '(' : illegal token on right side of '::' 1>c:\program files\microsoft visual studio 8\vc\include\complex(386) : error C2059: syntax error : '::' 1>c:\program files\microsoft visual studio 8\vc\include\complex(391) : error C2589: '(' : illegal token on right side of '::' 1>c:\program files\microsoft visual studio 8\vc\include\complex(391) : error C2059: syntax error : '::' 1>noise.cpp 1>llxmlrpctransaction.cpp 1>c:\program files\microsoft visual studio 8\vc\include\complex(356) : error C2589: '(' : illegal token on right side of '::' 1>c:\program files\microsoft visual studio 8\vc\include\complex(356) : error C2059: syntax error : '::' 1>c:\program files\microsoft visual studio 8\vc\include\complex(376) : error C2589: '(' : illegal token on right side of '::' 1>c:\program files\microsoft visual studio 8\vc\include\complex(376) : error C2059: syntax error : '::' 1>c:\program files\microsoft visual studio 8\vc\include\complex(381) : error C2589: '(' : illegal token on right side of '::' 1>c:\program files\microsoft visual studio 8\vc\include\complex(381) : error C2059: syntax error : '::' 1>c:\program files\microsoft visual studio 8\vc\include\complex(386) : error C2589: '(' : illegal token on right side of '::' 1>c:\program files\microsoft visual studio 8\vc\include\complex(386) : error C2059: syntax error : '::' 1>c:\program files\microsoft visual studio 8\vc\include\complex(391) : error C2589: '(' : illegal token on right side of '::' 1>c:\program files\microsoft visual studio 8\vc\include\complex(391) : error C2059: syntax error : '::' 1>llworldmapview.cpp 1>c:\program files\microsoft visual studio 8\vc\include\complex(356) : error C2589: '(' : illegal token on right side of '::' 1>c:\program files\microsoft visual studio 8\vc\include\complex(356) : error C2059: syntax error : '::' 1>c:\program files\microsoft visual studio 8\vc\include\complex(376) : error C2589: '(' : illegal token on right side of '::' 1>c:\program files\microsoft visual studio 8\vc\include\complex(376) : error C2059: syntax error : '::' 1>c:\program files\microsoft visual studio 8\vc\include\complex(381) : error C2589: '(' : illegal token on right side of '::' 1>c:\program files\microsoft visual studio 8\vc\include\complex(381) : error C2059: syntax error : '::' 1>c:\program files\microsoft visual studio 8\vc\include\complex(386) : error C2589: '(' : illegal token on right side of '::' 1>c:\program files\microsoft visual studio 8\vc\include\complex(386) : error C2059: syntax error : '::' 1>c:\program files\microsoft visual studio 8\vc\include\complex(391) : error C2589: '(' : illegal token on right side of '::' 1>c:\program files\microsoft visual studio 8\vc\include\complex(391) : error C2059: syntax error : '::' any help is appreciated

3 4

[Review] Polynomial library review begins today
by John Maddock 17 Mar '09

17 Mar '09

The review of Pawel Kieliszczyk's Polynomial library begins today and ends on Thurs 19th March. Download of the zip file from the vault is here: http://www.boostpro.com/vault/index.php?action=downloadfile&filename=polyno… Otherwise the library is present in the sandbox here: https://svn.boost.org/svn/boost/sandbox/SOC/2008/polynomial/ And the docs can be read online here: https://svn.boost.org/svn/boost/sandbox/SOC/2008/polynomial/libs/docs/index… The polynomial library contains a single class - polynomial<FieldType> - used for the manipulation of polynomials, along with a selection of algorithms which operate upon them. The library is an extension/rewrite of the existing "implementation detail" polynomial class in Boost.Math, and was written as part of last years Google Summer of Code under the mentorship of Fernando Cacciola. What to include in Review Comments ~~~~~~~~~~~~~~~~~~~~~~~~~ Your comments may be brief or lengthy, but basically the Review Manager needs your evaluation of the library. If you identify problems along the way, please note if they are minor, serious, or showstoppers. The goal of a Boost library review is to improve the library through constructive criticism, and at the end a decision must be made: is the library good enough at this point to accept into Boost? If not, we hope to have provided enough constructive criticism for it to be improved and accepted at a later time. The Serialization library is a good example of how constructive criticism resulted in revisions resulting in an excellent library that was accepted in its second review. Here are some questions you might want to answer in your review: * What is your evaluation of the design? * What is your evaluation of the implementation? * What is your evaluation of the documentation? * What is your evaluation of the potential usefulness of the library? * Did you try to use the library? With what compiler? Did you have any problems? * How much effort did you put into your evaluation? A glance? A quick reading? In-depth study? * Are you knowledgeable about the problem domain? And finally, every review should answer this question: * Do you think the library should be accepted as a Boost library? Be sure to say this explicitly so that your other comments don't obscure your overall opinion. Many reviews include questions for library authors. Authors are interested in defending their library against your criticisms; otherwise they would not have brought their library up for review. If you don't get a response to your question quickly, be patient; if it takes too long or you don't get an answer you feel is sufficient, ask again or try to rephrase the question. Do remember that English is not the native language for many Boosters, and that can cause misunderstandings. E-mail is a poor communication medium, and even if messages rarely get lost in transmission, they often get drowned in the deluge of other messages. Don't assume that an unanswered message means you're being ignored. Given constructively, criticism will be taken better and have more positive effects, and you'll get the answers you want. John Maddock. Review Manager for Polynomial Library.

1 2

How to use lock_guard
by Daniele Barzotti 17 Mar '09

17 Mar '09

Hi, I'm tring to use lock_guard, but I have some doubts... I have a method called OnDataReceived() which is called by another class through a signal. void ATDispatcher::OnDataReceived() { boost::lock_guard<boost::mutex> lock(_mut); // Data received from RS232 _device.GetData(_buffer); // .... CODE .... // unlock the mutex before notifications lock.~lock_guard(); signal_OnData(); } The first time it's called, it works fine. The second time the flow locks on the creation of the lock object. In the caller class I have solved this declaring the lock in this manner: boost::mutex::scoped_lock lock(_mutex); lock.unlock(); while (true) { //main loop lock.lock(); // saving data to vector _read_buffer.append(buf, rd); //Unlock the mutex lock.unlock(); signal_OnData(); } but here because there is the worker thread loop, while in the called class there is no loop. Thanks for suggestions. Daniele.

3 3

[mpl] is_constructible_from
by Roman Perepelitsa 17 Mar '09

17 Mar '09

I need a metafunction that returns true iif a type has a constructor with 2 arguments of a given type. Does mpl have it? If not, is it possible to implement? template <class C, class T1, class T2> struct is_constructible_from; struct foo { foo(int, int); }; struct bar { bar(int*, int); }; struct baz {}; int main() { assert(is_constructible_from<foo, int, int>::value); assert(is_constructible_from<foo, int, double>::value); assert(!is_constructible_from<bar, int, int>::value); assert(!is_constructible_from<baz, int, int>::value); }; Roman Perepelitsa.

6 6

[graph] Looks like there is a bug in some graph-writing function
by Juan Antonio Farré Basurte 17 Mar '09

17 Mar '09

> > > Hi, > > I was correctly reading graphs in graphml and graphviz. > Now I've added support for graph writing. > And now I get tons of > errors at the last line of my code (I mean symbol '}'). > > After looking carefully at them looks like there are two families of errors, > all of them related to property maps: > > - Looks like property maps are being accessed for reading using operator[]. > But if the > property map is backed by std::map, it's documented that operator[] > in fact does an insert in the map. This is not a problem when reading a > graph, > as the graph is mutable. But of course it is a problem when writing a const graph. > I get tons of errors about modifying read-only > locations and discarding qualifiers (const, of course). > > - The other errors are about not finding a matching operator>>, also > related with properties. > In this case, it looks to be related with reading (and I think concretely graphviz), but I don't understand > why, > as it used to work before. May be this is a cryptic bug of my code. > > I don't find clear information in the errors about what graph writing > function is causing the first family > of errors. I can attach compiler output, but I'm afraid messages are in Spanish :S > Moreover, I'm not > sure if it's correct to send attached files to the list. > If anybody wants me to send it, just ask. > I've been able to locate the exact origin point of the error. It hasn't been easy. It isn't a problem of graph writing functions. The problem is in sentences like this: dp.property("min",get(&V::min,g)); where dp is a reference to a dynamic_properties object, g is a const reference to a graph and V is the type of vertex bundled properties of graph g (min is one of those properties). Commenting out all lines like this, everything compiles perfectly. Looks like, somehow, that sentence tries to modify properties in graph g and, as g is const, it fails to compile. Is there any workaround? Thanks, Juan

1 0

Ayan S has sent you a private message
by Ayan S 17 Mar '09

17 Mar '09

1 0

Boost.Graph set of edges
by James Sutherland 17 Mar '09

17 Mar '09

Can you do something like typedef boost::graph_traits<Graph>::edge_descriptor Edge; Edge e; .... std::set<Edge> edges; edgeSet.insert( e ); I am getting compiler errors that imply that the edge_descriptor does not implement the < operator. Thanks, James

5 12

[System] Process ID
by Seweryn Habdank-Wojewódzki 17 Mar '09

17 Mar '09

Dears, Is there any functionality like Process ID in boost system? I found in Flyweight details such a nice header to have portable Process ID functionality, but maybe there is another somewhere in boost. Dangerous is to use functions from "details". Regards, -- Seweryn Habdank-Wojewódzki ---------------------------------------------------- Szczypta baśni, garść makabry. Wysmakowana wizualnie opowieść dla całej rodziny. KORALINA I TAJEMNICZE DRZWI 3D - w kinach! Zobacz zwiastun --> http://klik.wp.pl/?adr=http%3A%2F%2Fcorto.www.wp.pl%2Fas%2Fkoralina.html&si…

4 4

Static boost libraries on Linux
by Lennyk 17 Mar '09

17 Mar '09

Hi, I'm using boost ver 1.35 on Linux. When I've installed boost on my Linux machine, the build created dynamic libraries (only .so). How can I build boost on Linux so it'll create static libraries as well? Thanks!

2 2

[Boost users] MacPorts and Boost path issues
by John Burgoon 17 Mar '09

17 Mar '09

I am a command-line user of C++ and boost. I installed boost on Mac Leopard using mac ports like this: sudo port install boost I saw as it installed that macports was using bjam, and the version was 1.37. I want to use boost/regex.hpp, so I know it's not just a matter of including a header file. When I compile my own source files, I see a variety of paths, such as /opt/local/var/macports/sources/rsync.macports.org/release/ports/devel/boost /opt/local/var/macports/sources/ rsync.macports.org/release/ports/devel/boost/files etc My first question is, when I compile my work on command line using g++: i.e., g++ myfile.cpp -o myfile which of these paths should I include? I think I am supposed to do something like g++ -I /path/to/boost myfile.cpp -o myfile but I do not understand how a person knows which dir to use. Another way of asking this is: If YOU sat down at my machine and decided to use boost, what would you do to a)find it and b)include it in your pre-processor includes? Is there clear documentation on how a fellow figures out where the install is located, and then shows how to include it? Perhaps I simply don't know what to ask for, but when I google for this I gfind a zillion unrelated things. Do I in fact need to do the "-I /silly/long/path/to/the/bowels/of/the/machine/where/macports/hid/buried/installed/boost" every single time I compile something?? It occured to me to try a soft link somewhere in my development dir to save typing, but in order to do that of course I first need to figure out where boost actually resides in the system. I also have fink installed on the system, and fink says that boost1.33 is installed. So how in the world do I update fink to accept a later version? When I try to update, fink says no update is needed. Thanks in advance for the hand-holding. Boost is worth the effort to figure it out, yes? -- John Burgoon Development Staff, GiveALink.org Informatics 304 812-856-1833 office 812-391-2428 cell -- >>Insight without implementation is worthless. Get to work.

2 1

problems with boost::spirit
by Goran 17 Mar '09

17 Mar '09

Hi all, I wanna get boost::spirit to tell me which position in my string is liable that parsing fails. Here's a fully compiling demonstration: #-------------------------------------------------------------------- #include <iostream> #include <string> #include <boost/spirit.hpp> using namespace std; using namespace boost::spirit; int main() { string s = "#x"; rule<> r = chlit<>('#') >> chlit<>('X') >> *space_p; parse_info<> p = parse(s.c_str(),r); if(p.full) cout << "File is OK:\n"; else cout << "Error at pos " << p.length << ".\n"; return 0; } #-------------------------------------------------------------------- With this I get: Error at pos 4294967295. But if "string s = "#Xx" I get: Error at pos 2. This is what I want. But why do I get 4294967295 which is 2^32-1. Thanks for any comments, Goran.

2 1

ASIO certificate help.
by Mohammed Rasmy 16 Mar '09

16 Mar '09

1 0

mutex types and interprocess documentation question
by Archie14 16 Mar '09

16 Mar '09

Boost documention has following code in the paragraph "Scoped lock" (reference to the paragraph is here.../html/interprocess/synchronization_mechanisms.html#interprocess.synchron ization_mechanisms.mutexes) using namespace boost::interprocess; //Let's create any mutex type: MutexType mutex; ... I did not find a reference to available types declarations. Paragraph above "Scoped lock" talks about mutex types, but I did not find type declarations. My question is: where are available types declared? I will also appreciate the example of such declaration that shows what #include should be used.

2 2

another metaprogramming question
by Noah Roberts 16 Mar '09

16 Mar '09

This seems to me like it should work. Yet the check() call resolves, without error, to the first one...which should cause an error and trigger SFINAE. Can anyone help me figure out what gives? Compiler is MSVC++ 8. Normally the SFINAE stuff works so I'm betting on my mistake this time. Basically I need to check it the metafunction meta<T> returns something. If it does then the object is a record. If it doesn't then it is not. #include <iostream> #include <boost/mpl/transform.hpp> #include <boost/mpl/placeholders.hpp> #include <boost/mpl/vector.hpp> template < typename T1, typename T2 > struct construct_metafield_composed { typedef T1 type; // Not what the real one does. }; template < typename RECORD > struct meta { struct type : boost::mpl::transform < typename RECORD::fields , construct_metafield_composed < RECORD , boost::mpl::placeholders::_1 > >::type { typedef RECORD meta_for; }; }; template < typename T > struct is_record_check { typedef char (&no) [1]; typedef char (&yes) [2]; template < typename S > static yes check(typename meta<S>::type *); template < typename S > static no check(...); enum { value = sizeof(check<T>(0)) == sizeof(yes) }; typedef boost::mpl::bool_<value> type; }; struct test { typedef boost::mpl::vector< int, double > fields; }; int main() { // output is 1\n1\n. std::cout << is_record_check<test>::value << std::endl; std::cout << is_record_check<int>::value << std::endl; // meta<int>::type x; - yet this will cause compilation error. std::cin.get(); }

2 2

gcc optimization and boost::any
by John 16 Mar '09

16 Mar '09

Is anyone aware of a bug/quirk in boost::any that is exposed by GCC optimization? (Or vice versa: a bug in GCC optimization exposed by boost::any)? We've got a particularly insidious runtime bug that appears under `-O2` with gcc version 4.2.2, but not with gcc 3.3.3, and not with gcc 4.3.3. We narrowed it down to a single .cpp file, which contains a single method, which does nothing but call boost::any_cast several times to extract a pointer from a boost::any instance. By compiling this file under -O2, (gcc version 4.2.2) we can cause the bug. By compiling without any `-O` flags, the bug does not appear. (boost version 1_34_1). We're working to identify a minimal test case that we can post here. We still haven't pinpointed the exact cause. To be clear, we're not 100% sure the bug is in boost::any or gcc -- it merely appears to be so since changing compiler flags or compiler versions makes the bug disappear. If you know of such a bug or quirk in either GCC or boost::any, or of a write-up of this somewhere... or even some idle speculation... it could save us a lot of time tracking this down. Thanks, -John

3 3

Interprocess general questions
by Andrej van der Zee 16 Mar '09

16 Mar '09

Hi, I am using Boost.interprocess for the first time and read the documentation. It is difficult for me to decide what construct would best fit my scheme, because of all the options. I hope that somebody could help me find out. I am using Boost 1.38 on Ubuntu 8.10. In my setup, I have one parent process that forks many (could be hundreds) child processes. In the parent process I need to maintain a data-queue. The child processes insert data into the queue and once every X seconds the parent removes all messages from the queue (in order to dump it to a database). The data is one type only and fixed length: A struct with a few doubles, a few integers and a fixed-length string. Because there can be hundreds of child processes that insert into the data-queue many times a second, I guess it would make sense to have multiple data regions in one big shared memory segment, each with it's own synchronization mechanism? I mean, it is important that the child processes are able to insert the data fast and do not suffer lock contention too much (a "little" overhead is fine though). So my question is what components from the Boost.interprocess-library should I use? Any other comments and/or tips before I start coding? Thank you! Andrej van der Zee

1 0

installing boost in linux (help needed urgently) !!!!!!
by ayan sinha 15 Mar '09

15 Mar '09

Hi, I am new to linux and also to boost library.. I am planning to install boost library in linux box (which is a production machine with no net connectivity). i have already got the linux version of boost in that machine and now I am planning to install it. Can please somebody tell me what are the steps needs to be done to install linux.. with commands (I cant use any tool that boost provide to install because ther is no net connectivity in that box). I have gone through the boost website .. but it looks very complicated :) also I have gone through couple of websites but it was not much usefull. I will be really greatfull if somebody can tell me what are the steps (with commands) to install in linux box. Thanks in Advance, Ayan.

5 5

Why "Mixing a dll boost library with a static runtime is a really bad idea..."?
by D M 15 Mar '09

15 Mar '09

Hi, I want (need) to use boost libs as dll with a static run-time library. In this case auto_link.hpp gives me an error "Mixing a dll boost library with a static runtime is a really bad idea...". Why is it a bad idea? I don't think that it's a good idea but requirements in our project demand static-runtime linkage for all modules and they don't want to use regex as static library cause of it's size. What problems could it cause? I know only one problem with a memory allocation in different modules. Any more? Thanks, Vladimir

3 3

Re: [Boost-users] proto: analytical-only math functions
by Hicham Mouline 14 Mar '09

14 Mar '09

> ----- Original Message ----- > From: peter_foelsche(a)agilent.com > To: boost-users(a)lists.boost.org > Subject: Re: [Boost-users] proto: analytical-only math functions > Date: Fri, 6 Mar 2009 12:40:22 -0700 > > > Could the originator of this thread try to explain to me what he wants to > accomplish? > I still cannot figure any use case for proto. > > Peter Step1 is to have a derivative metafunction which computers the nth derivative of a function at compile time. In order to do that, I need a mathematical language to write my functions. so I have constants, variables, basic_functions and user-defined functions that are terminals of my language f(x) = log(x); g(x) = derivative(1, f, x); g(x) will be 1/x; I should manage to run derivative purely in compile time. Later, I will look at stochastic and partial differential equations and write functions to solve them analytically whenever possible... proto will help me define terminals, the grammars for what are acceptable expressions. regards,

2 6

Seekg on a gzip'd file...Is this possible?
by Gerrick Bivins 14 Mar '09

14 Mar '09

Hello all, I¹m switching one of my projects over from java to c++. Basically it¹s a gzip file reader, where I¹d like to parse a bunch of numbers (floats, ints, doubles) and in some instances I¹d like to skip some of the data. I¹m a ³noob² to boost and after getting the file reader working using the boost like this: { .... std::ifstream ormodelFile(filenameForTimeStep.c_str(), std::ios_base::in|std::ios_base::binary); boost::iostreams::filtering_istreambuf compressedFileFilter; compressedFileFilter.push(boost::iostreams::gzip_decompressor()); compressedFileFilter.push(ormodelFile); this->inputFile.rdbuf(&compressedFileFilter); //where inputFile is a an istream. ... //do the file reading } I tried to add the skipping parts to the code by simply calling: { ... double value; this->inputFile->seekg(sizeof(double),ios::cur); ... } the fail bit is set on the inputFile istream. Is it possible to do this using boost? I¹m pretty ³green² with boost so I¹m sure it¹s just my misunderstanding of how to use the lib but I couldn¹t find my way around the docs to well. The best I could find was that I need to somehow make my stream ³seekable² but it wasn¹t too clear to me how to do this. Any help would be greatly appreciated. Gerrick

5 8

Re: [Boost-users] shared_ptr, auto_ptr and enable_shared_from_this confusion
by 张亚霏 14 Mar '09

14 Mar '09

shared_ptr's constructor in line 168 calls 'boost::detail::sp_enable_shared_from_this( pn, p, p );'. 'boost::shared_ptr<IDerived> c(new Derived());' is dispatched to 'sp_enable_shared_from_this' in line 86. pay attension to the second parameter, it can accept all types that inherit 'enable_shared_from_this<T>', so does 'Derived*'. While, 'boost::shared_ptr<IDerived> c((IDerived*)new Derived());' is dispatched to 'sp_enable_shared_from_this' in line 111, bacause the static type IDerived* has nothing to do with 'enable_shared_from_this<T>' and can not be implicitly converted to 'enable_shared_from_this<T>'. Then check out the difference between the two 'sp_enable_shared_from_this'. It is obvious that the second usage shall cause a BOOST_ASSERT failed in 'enable_shared_from_this::shared_from_this', so an exception is raised. However, I think the root reason is not the language nor boost, but the design of class IDerived. Its root and root1 seem a little bit irrational because there is not direct relationship between IDerived and Base(IBase). ------------------ Original ------------------ From: "Ryan McConnehey"<mccorywork(a)gmail.com>; Date: Sat, Mar 14, 2009 04:29 AM To: "Boost-users"<boost-users(a)lists.boost.org>; Subject: [Boost-users] shared_ptr,auto_ptr and enable_shared_from_this confusion I'm having some trouble with using shared_ptr, auto_ptr and enable_shared_from_this with my inheritance layout. The following is how my inheritance is laid out. // Inheritance Layout // // boost::enable_shared_from_this<T> // ^ // | // Base --------> IBase // ^ // | // Derived -----> IDerived In my test code (that I've included with this email) the following two ways to instantiate a shared pointer is equivalent. boost::shared_ptr<IDerived> r (new Derived()); std::auto_ptr<IDerived> k (new Derived); boost::shared_ptr<IDerived> l (k.release()); When creating the shared_ptr, both declaration take me to line 185 of shared_ptr.hpp. Instantiating the shared_ptr directly then takes me to line 105 of shared_ptr.hpp. Though when instantiating by releasing the auto_ptr into the shared_ptr I'm taken to line 129. It seems to me that I should be going to the same line of code. This wouldn't be such a problem except when trying to call shared_from_this() an exception is thrown if my shared_ptr was created from an auto_ptr. Is this behavior correct? Ryan

1 0

(no subject)
by isik bicer 14 Mar '09

14 Mar '09

Hello everybody, I am a newly boost user and have a problem in the following code. Since it is long, i put only the relevant parts. Could anyone help me please? When I try to get the vertices from a vector and put them to add_edge(), i get the following error massage : 'boost::add_edge' : cannot convert parameter 1 from 'char' to 'void *' The problematic part is highlighted. Code:typedefvector<Edge> simpleedgevector; vector<Edge>::iterator s_e_i;for simpleedge = Edge((*(*vec_it1)),(*(*vec_it1+1))); simpleedgevector.push_back(simpleedge);}edge_descriptor e; cout<< tie(e, inserted) = add_edge((*s_e_i).first, (*s_e_i).second, g); }for(s_e_i = simpleedgevector.begin();s_e_i != simpleedgevector.end();s_e_i++){boolinserted;"first part : "<<(*s_e_i).first<<" second part : "<<(*s_e_i).second<<endl;(vec_it1 = output1.begin();vec_it1 !=output1.end();vec_it1++){std::pair<char, char> Edge;

1 0

shared_ptr, auto_ptr and enable_shared_from_this confusion
by Ryan McConnehey 13 Mar '09

13 Mar '09

I'm having some trouble with using shared_ptr, auto_ptr and enable_shared_from_this with my inheritance layout. The following is how my inheritance is laid out. // Inheritance Layout // // boost::enable_shared_from_this<T> // ^ // | // Base --------> IBase // ^ // | // Derived -----> IDerived In my test code (that I've included with this email) the following two ways to instantiate a shared pointer is equivalent. boost::shared_ptr<IDerived> r (new Derived()); std::auto_ptr<IDerived> k (new Derived); boost::shared_ptr<IDerived> l (k.release()); When creating the shared_ptr, both declaration take me to line 185 of shared_ptr.hpp. Instantiating the shared_ptr directly then takes me to line 105 of shared_ptr.hpp. Though when instantiating by releasing the auto_ptr into the shared_ptr I'm taken to line 129. It seems to me that I should be going to the same line of code. This wouldn't be such a problem except when trying to call shared_from_this() an exception is thrown if my shared_ptr was created from an auto_ptr. Is this behavior correct? Ryan #include <iostream> #include <memory> #include <boost/shared_ptr.hpp> #include <boost/enable_shared_from_this.hpp> // Inheritance Layout // // boost::enable_shared_from_this<T> // ^ // | // Base --------> IBase // ^ // | // Derived -----> IDerived class IBase { public: virtual void hello(void) = 0; }; class Base : public IBase, public boost::enable_shared_from_this<Base> { public: Base(void){} ~Base(void){} void hello(void) { std::cout << "Hello" << std::endl; } }; class IDerived { public: virtual void world(void) = 0; static std::auto_ptr<IDerived> createAuto(void); static boost::shared_ptr<IDerived> createShared(void); virtual boost::shared_ptr<Base> root(void) = 0; virtual boost::shared_ptr<IBase> root2(void) = 0; }; class Derived : public IDerived, public Base { public: Derived(void){} ~Derived(void){} void world(void) { std::cout << "World" << std::endl; } virtual boost::shared_ptr<Base> root(void) { return shared_from_this(); } virtual boost::shared_ptr<IBase> root2(void) { return shared_from_this(); } }; std::auto_ptr<IDerived> IDerived::createAuto(void) { return std::auto_ptr<IDerived>(new Derived()); } boost::shared_ptr<IDerived> IDerived::createShared(void) { return boost::shared_ptr<IDerived>(new Derived()); } int main(int argc, char* argv[]) { boost::shared_ptr<IDerived> c = IDerived::createShared(); boost::shared_ptr<IBase> d = c->root2(); std::cout << c.use_count() << std::endl; boost::shared_ptr<IDerived> r (new Derived()); std::auto_ptr<IDerived> k (new Derived); boost::shared_ptr<IDerived> l (k.release()); boost::shared_ptr<IBase> m = l->root2(); std::cout << m.use_count() << std::endl; return 0; }

2 1

using boost :: thread in Linux
by ayansinhas＠gmail.com 13 Mar '09

13 Mar '09

HI, I am using boost::thread for my project. I have extracted only boost::thread through BCP and I have build my multithreading application. It is working properly in my local machine (Windows)now I have to port it over to linux box. I had couple of questions ...?? Is it ok to copy the boost::thread folder (That I have extracted) from my local machine to linux box and run my application?? or is it that I have to install the entire boost library in linux box and then by using bcp I have to extract only boost:thread part which I can use for my application?? Please let me know what can be the possible solution.If there is any other solution please feel free to advice. Thanks In Advance Ayan.

2 1