August 2004 - Boost-users - lists.boost.org

RE: [Boost-users] Re: Guidelines for effective exception usage
by Ben Hutchings 25 Aug '04

25 Aug '04

Peter Dimov <pdimov(a)mmltd.net> wrote: > Victor A. Wagner Jr. wrote: > > > > OK, blunt point. > > the standard is (insert favorite expletive or "in error") if it > > allows use of std::runtime_error to terminate the program due to low > > memory situations (run out of memory (due to copying) during stack > > unwinding). > > It does not allow such a thing. This is an known and unresolved issue: <http://www.open-std.org/jtc1/sc22/wg21/docs/lwg-active.html#254>

2 1

Update: exception class that
by Stephen torri 25 Aug '04

25 Aug '04

I thought I would share what I am using for exceptions given the excellent advice from everyone responding to my initial email. There is no usages of any memory allocation or std::string like variables. If there are any 'oops' or 'gotchas' please let me know. My plan is to use a command flag for taking into a value. The value will determine what debug messages are reported at the cause of the problem. The message will be displayed via std::cerr. Stephen ---------------- Header ---------------- #ifndef IO_EXCEPTION_H_ #define IO_EXCEPTION_H_ #include <exception> #include <inttypes.h> namespace kurt { /*! * \class IOException * \brief Rather than inheriting from ifstream this class delegates * any required functionality to it. * \date 2003 * \author Stephen Torri */ class IO_Exception : public virtual std::exception { public: IO_Exception (uint32_t message_id); /*! * \brief Destructor */ virtual ~IO_Exception (void) throw(){} /*! * \brief Return the extra message accompanying the exception. */ const char* what (void) const throw(); /*! * \brief Return tne name of the exception */ const char* name (void) const throw(); enum { CANNOT_OPEN, INPUT_CORRUPTED, INVALID_IO_OP, READ_ERROR, UNSUPPORTED_OP, WRITE_ERROR, }; private: static const char* m_exception_name; static const char* m_messages[]; uint32_t m_id; }; } /* namespace kurt */ ------------- Source ------------- #include "IOException.hpp" namespace kurt { const char* IO_Exception::m_exception_name = "IO_Exception"; const char* IO_Exception::m_messages [] = { "I/O Error: cannot open", // CANNOT_OPEN "Input data is corrupted", // INPUT_CORRUPTED "Invalid i/o operation attempted", // INVALID_IO_OP "Error reading input", // READ_ERROR "Unsupported I/O operation", // UNSUPPORTED_OP "Error writing output", // WRITE_ERROR }; IO_Exception::IO_Exception (uint32_t message_id) : std::exception(), m_id(message_id) {} const char* IO_Exception::what (void) const throw() { return m_messages[m_id]; } const char* IO_Exception::name (void) const throw() { return m_exception_name; } } /* namespace kurt */ #endif /* IO_EXCEPTION_H_ */ -- Email: storri(a)torri.org

1 0

MPL vector size limit
by Arkadiy Vertleyb 25 Aug '04

25 Aug '04

Hi, Can BOOST_MPL_LIMIT_VECTOR_SIZE be increased beyond 50, and if yes, what is involved in this? Thanks, Arkadiy

2 2

licenses...
by MarC 24 Aug '04

24 Aug '04

Hi I subscribed to this list to find out if I can publish one program containig "Boost Filesystem Library" under the GPL2. I saw that Boost license is even more flexible than the GPL2 but I'm not sure if I can release a program containing Boost with GPL2... anybody knows? where should I ask to answer my question? my case is that I'm developping a stand-alone console program which I want to integrate in other GPL licensed projects of web developpment like quanta.sourceforge.net or screem.sourceforge.net. I don't mind about modifying the licene of my program to adapt it to the special license of "Boost Filesystem Library" but I think it's inaffordable to modify all licenses of big projects like quanta or screem... so I'm a little bit confused. any idea? thanks in advance!

4 4

boost.graph build problem in latest anonomous CVS update.
by Jeff Holle 24 Aug '04

24 Aug '04

I've just refreshed my boost via CVS and ran into a build problem with gcc v3.4.1. Initially, seems like everything failed on my system. However, I discovered that there is now a configure option :-) . After installing user.hpp where it does some good stuff, the bjam of the root of boost succeeded. However, in running bjam in $ROOT_BOOST/libs/graph/build. This Jamfile is configured to build both release and debug version of bgl-viz. It fails with similiar problems that I was getting elsewhere before I ran configure. An example of the problems is: boost/config/stdlib/libstdcpp3.hpp:33: error: expected unqualified-id before '<<' token boost/config/stdlib/libstdcpp3.hpp:33: error: expected `,' or `;' before '<<' token

4 4

can the boost multi-array be easily extended to std::map's?
by Steven L. Scott 24 Aug '04

24 Aug '04

I'm working on an application (a data analysis system built in C++) for which it would be helpful to have nested (multidimensional) containers other than ordinary arrays. I have defined a generic 'data' class which I like to store in std::vector<data *>'s. For a very simple example, a 'data' might be a double representing a student's score on an exam, or in more complex settings it might be some sort of spatial map, or a multidimensional set of records for a medical patient. I am interested in structures where 'data' is nested across multiple levels. For example, test scores are nested within students, students nested within teachers, nested within school districts, etc. For a single layer of nesting a std::map<std::string, std::vector<data *> > seems to be an ideal way to store such data, largely because text strings can be used to identify subjects: dat["johnny"] // returns a std::vector of pointers to johnny's test scores. This is often the way that data are stored in files to be analyzed. I'm looking for some sort of C++ container that is more convenient for multiply nested data than std::map<std::string, std::map<std::string .... , std::vector<data *> > > ... >, but which will allow me to write expressions like: dat["LAUSD"]["Central High"]["Ms. Jones' Class"]["Johnny"] // same vector as above. Can something like the Boost multi-array package be extended to handle this?

2 1

problems compiling example files
by MarC 24 Aug '04

24 Aug '04

I suppose, it's a begginer's linux problem but I didn't found the answer neither in the documentation nor in the internet forums... After installing boost on my system (see how at the end of this email), I wanted to compile simple_ls.cpp, the filesystem example. everything works when creating the .o but when linking a lot of undefined references arised. What have I done wrong? g++ -c simple_ls.cpp [marc@localhost ls]$ g++ -c simple_ls.cpp [marc@localhost ls]$ ls ls/ simple_ls.cpp* simple_ls.o [marc@localhost ls]$ g++ simple simple_ls.cpp simple_ls.o [marc@localhost ls]$ g++ simple_ls.cpp -o simple_ls /tmp/ccjMYBxo.o(.text+0x19): In function `main': : undefined reference to `boost::filesystem::initial_path()' /tmp/ccjMYBxo.o(.text+0x44): In function `main': : undefined reference to `boost::filesystem::native(std::basic_string<char, std::char_traits<char>, std::allocator<char> > const&)' /tmp/ccjMYBxo.o(.text+0x55): In function `main': : undefined reference to `boost::filesystem::path::path[in-charge](char const*, bool (*)(std::basic_string<char, std::char_traits<char>, std::allocator<char> > const&))' /tmp/ccjMYBxo.o(.text+0x64): In function `main': : undefined reference to `boost::filesystem::system_complete(boost::filesystem::path const&)' /tmp/ccjMYBxo.o(.text+0x119): In function `main': : undefined reference to `boost::filesystem::exists(boost::filesystem::path const&)' /tmp/ccjMYBxo.o(.text+0x13c): In function `main': : undefined reference to `boost::filesystem::path::native_file_string() const' /tmp/ccjMYBxo.o(.text+0x1bf): In function `main': : undefined reference to `boost::filesystem::is_directory(boost::filesystem::path const&)' /tmp/ccjMYBxo.o(.text+0x1e2): In function `main': : undefined reference to `boost::filesystem::path::native_directory_string() const' /tmp/ccjMYBxo.o(.text+0x291): In function `main': : undefined reference to `boost::filesystem::is_directory(boost::filesystem::path const&)' /tmp/ccjMYBxo.o(.text+0x2ca): In function `main': : undefined reference to `boost::filesystem::path::leaf() const' /tmp/ccjMYBxo.o(.text+0x351): In function `main': : undefined reference to `boost::filesystem::path::leaf() const' /tmp/ccjMYBxo.o(.text+0x418): In function `main': : undefined reference to `boost::filesystem::path::leaf() const' /tmp/ccjMYBxo.o(.text+0x586): In function `main': : undefined reference to `boost::filesystem::path::native_file_string() const' /tmp/ccjMYBxo.o(.gnu.linkonce.t._ZN5boost10filesystem18directory_iteratorC1ERKNS0_4pathE+0x2d): In function `boost::filesystem::directory_iterator::directory_iterator[in-charge](boost::filesystem::path const&)': : undefined reference to `boost::filesystem::detail::dir_itr_init(boost::shared_ptr<boost::filesystem::detail::dir_itr_imp>&, boost::filesystem::path const&)' /tmp/ccjMYBxo.o(.gnu.linkonce.t._ZNK5boost10filesystem18directory_iterator11dereferenceEv+0xd): In function `boost::filesystem::directory_iterator::dereference() const': : undefined reference to `boost::filesystem::detail::dir_itr_dereference(boost::shared_ptr<boost::filesystem::detail::dir_itr_imp> const&)' /tmp/ccjMYBxo.o(.gnu.linkonce.t._ZN5boost10filesystem18directory_iterator9incrementEv+0xd): In function `boost::filesystem::directory_iterator::increment()': : undefined reference to `boost::filesystem::detail::dir_itr_increment(boost::shared_ptr<boost::filesystem::detail::dir_itr_imp>&)' collect2: ld returned 1 exit status thank you in advance again. PS: my system is: [marc@localhost html]$ g++ -v Reading specs from /usr/lib/gcc-lib/i586-mandrake-linux-gnu/3.3.2/specs Configured with: ../configure --prefix=/usr --libdir=/usr/lib --with-slibdir=/lib --mandir=/usr/share/man --infodir=/usr/share/info --enable-shared --enable-threads=posix --disable-checking --enable-long-long --enable-__cxa_atexit --enable-clocale=gnu --enable-languages=c,c++,ada,f77,objc,java,pascal --host=i586-mandrake-linux-gnu --with-system-zlib Thread model: posix gcc version 3.3.2 (Mandrake Linux 10.0 3.3.2-6mdk) And i've installed boost using these packages: Nombre: libboost1 Versión: 1.31.0-5mdk ‎/usr/lib/libboost_date_time.so.1 ‎/usr/lib/libboost_date_time.so.1.31.0 ‎/usr/lib/libboost_filesystem.so.1 ‎/usr/lib/libboost_filesystem.so.1.31.0 ‎/usr/lib/libboost_prg_exec_monitor.so.1 ‎/usr/lib/libboost_prg_exec_monitor.so.1.31.0 ‎/usr/lib/libboost_python.so.1 ‎/usr/lib/libboost_python.so.1.31.0 ‎/usr/lib/libboost_regex.so.1 ‎/usr/lib/libboost_regex.so.1.31.0 ‎/usr/lib/libboost_signals.so.1 ‎/usr/lib/libboost_signals.so.1.31.0 ‎/usr/lib/libboost_test_exec_monitor.so.1 ‎/usr/lib/libboost_test_exec_monitor.so.1.31.0 ‎/usr/lib/libboost_thread.so.1 ‎/usr/lib/libboost_thread.so.1.31.0 ‎/usr/lib/libboost_unit_test_framework.so.1 ‎/usr/lib/libboost_unit_test_framework.so.1.31.0 ‎/usr/share/doc/libboost1-1.31.0 ‎/usr/share/doc/libboost1-1.31.0/LICENSE_1_0.txt Nombre: libboost1-devel Versión: 1.31.0-5mdk Archivos: ‎/usr/include/boost ‎/usr/include/boost/aligned_storage.hpp ‎/usr/include/boost/any.hpp ‎/usr/include/boost/array.hpp ‎/usr/include/boost/assert.hpp ‎/usr/include/boost/bind ‎/usr/include/boost/bind.hpp ‎/usr/include/boost/bind/apply.hpp ‎/usr/include/boost/bind/arg.hpp ‎/usr/include/boost/bind/bind_cc.hpp ‎/usr/include/boost/bind/bind_mf_cc.hpp ‎/usr/include/boost/bind/bind_template.hpp ‎/usr/include/boost/bind/make_adaptable.hpp ‎/usr/include/boost/bind/mem_fn_cc.hpp ‎/usr/include/boost/bind/mem_fn_template.hpp ‎/usr/include/boost/bind/mem_fn_vw.hpp ‎/usr/include/boost/bind/placeholders.hpp ‎/usr/include/boost/bind/protect.hpp ‎/usr/include/boost/blank.hpp ‎/usr/include/boost/blank_fwd.hpp ‎/usr/include/boost/call_traits.hpp ... /usr/lib/libboost_date_time.so ‎/usr/lib/libboost_filesystem.so ‎/usr/lib/libboost_prg_exec_monitor.so ‎/usr/lib/libboost_python.so ‎/usr/lib/libboost_regex.so ‎/usr/lib/libboost_signals.so ‎/usr/lib/libboost_test_exec_monitor.so ‎/usr/lib/libboost_thread.so ‎/usr/lib/libboost_unit_test_framework.so ‎/usr/share/doc/libboost1-devel-1.31.0 ‎/usr/share/doc/libboost1-devel-1.31.0/README ‎/usr/share/doc/libboost1-devel-1.31.0/boost.css ‎/usr/share/doc/libboost1-devel-1.31.0/c++boost.gif Nombre: libboost1-static-devel Versión: 1.31.0-5mdk ‎/usr/lib/libboost_date_time.a ‎/usr/lib/libboost_filesystem.a ‎/usr/lib/libboost_prg_exec_monitor.a ‎/usr/lib/libboost_python.a ‎/usr/lib/libboost_regex.a ‎/usr/lib/libboost_signals.a ‎/usr/lib/libboost_test_exec_monitor.a ‎/usr/lib/libboost_unit_test_framework.a

3 2

boost.graph question concerning edge descriptor details
by Jeff Holle 23 Aug '04

23 Aug '04

I'm using gcc v3.4.1 and boost_CVS captured on Tue Jun 29 05:25:17 EDT 2004. I'm developing a boost.graph application that must deal with multiple graphs. The following typedefs describe how its using boost.graph: typedef boost::adjacency_list<boost::vecS,boost::vecS,boost::bidirectionalS> GraphT; typedef boost::shared_ptr<GraphT> GraphPtr; typedef boost::graph_traits <GraphT>::vertex_descriptor Vertex; typedef boost::graph_traits <GraphT>::edge_descriptor Edge; In addition, I have various maps that must act as a "bridge" between these graphs. I'm using the following typedefs as keys in such maps: typedef std::pair<GraphPtr,Vertex> ExternalVertex; typedef std::pair<GraphPtr,Edge> ExternalEdge; I've used ExternalVertex successfully in two different maps. Now I'm having trouble populating the first map that uses ExternalEdge. The compiler warnings that I'm getting sugguest that Edge does have an "operator<" method. Specifically its: error: no match for 'operator<' in '__x->std::pair<GraphPtr, Edge>::second < __y->std::pair<GraphPtr, Edge>::second' I'm wondering why Edge is different from Vertex. As you can see, as I've defined GraphT, both edge and vertex properties should be no_property. Any guidance that assists me to, at worst case, write a Compare operator for my "map<ExternalEdge,...>" containers would be appreciated.

2 3

Removing edges/vertices from a boost.graph object
by Jeff Holle 23 Aug '04

23 Aug '04

I'm using boost.graph captured from CVS on Tue Jun 29 05:25:17 EDT 2004. The type of graph that I'm using is: typedef boost::adjacency_list<boost::vecS,boost::vecS,boost::bidirectionalS> GraphT; I need to trim vertex/edges that was moved to other graphs. I can't seem to do this. Calling remove_vertex appears to be a NOOP. The last thing I tried was: clear_vertex(*vi); remove_vertex(*vi); The gerdemann.cpp example gave me this idea. The clear_vertex did the job that I need in that the necessary edges disappeared. However, the remove_vertex still does not remove the vertexs. What do I have to do to get the ability to clear vertices?

2 2

boost::thread
by Raja 23 Aug '04

23 Aug '04

Does using thread classes from boost need additional library files other than header files ? I am getting the following error when I use the thread class: /usr/include/c++/3.2.3/bits/stl_heap.h:188: undefined reference to `boost::thread::thread[in-charge](boost::function0<void, std::allocator<boost::function_base> > const&)' I also see 'libboost_thread-gcc-mt-d-1_31.so.1.3.0' library under boost/bin/... Does this have to be used ? Thanks in advance Raja

2 1

boost::filesystem problems
by Carlo Wood 23 Aug '04

23 Aug '04

A few months ago I started to write my first multi-platform application. In the hope to be shielded from all kinds of filesystem specific issues, I started to use boost - also for the first time - mainly because of of boost::filesystem. However - I have found that it doesn't shield me from a LOT of filesystem specific things. In the meantime I was in the need of writing six additional functions that are filesystem specific: std::string native_file_string(fs::path const&); void copy_file(fs::path const& source_file, fs::path const& target_file); void chmod(fs::path const& filename, bool readonly); bool is_readonly(fs::path const& filename); void rename_file(fs::path const& source_file, fs::path const& target_file); size_t get_file_size(fs::path const& filename); Today I ran into a very serious problem. Hopefully it is just me who is making a stupid mistake (however unlikely that may sound ;) but it sure seems like a huge bug in boost::filesystem of a magnitude that not only amazes me but will also force me to stop using boost::filesystem if this is true. The reason that I didn't find this problem earlier in the pasts months is because so far I've done all development on GNU/Linux - and only recently have started to actually try and get things to work on cygwin. The reason I subscribed to this list is in the hope to find Beman here, is he reading this list? My hope is to start a public discussion about boost::filesystem - problems, improvements, missing features etc. I will sent a separate mail for the 'serious bug' that I just mentioned, because it isn't 100% sure yet that this is actually a bug ;). -- Carlo Wood <carlo(a)alinoe.com>

13 22

Build with BCB6 failed
by Steffen Rauh 21 Aug '04

21 Aug '04

Hello, creating a custom build, I get lots of compiler errors during Boost.Python compilation. Some of these compiler errors are: c:\...\boost_1_31_0\libs\python\build\../src/numeric.cpp: Error E2231 C:\...\boost_1_31_0\boost/python/detail/value_is_shared_ptr.hpp 14: Member value_is_shared_ptr<X_>::value cannot be used without an object. Error E2299 C:\...\boost_1_31_0\boost/python/detail/value_is_shared_ptr.hpp 14: Cannot generate template specialization from 'mpl::bool_<C_>' Error E2040 C:\...\boost_1_31_0\boost/python/detail/value_is_shared_ptr.hpp 14: Declaration syntax error These error messages repeat in a lot of other files. Sometimes the compiler aborts compilation because of too many errors or warnings. The rest of boost seems to compile fine. Does anybody have an idea on how to solve this problem? Is this difficulty already known? I tried to compile with Python 2.3 at first but switched back to Python 2.2 because I read somewhere that there may be some problems with this version. But it did not solve the problem. I already tried to compile the current CVS Snapshot but it did not change a thing. My configuration is: - Python 2.2 - Boost 1.31.0 - bjam 3.1.10 - Windows XP - Borland C++ Builder 6 Personal Building Boost with: c:\...\boost_1_31_0>bjam -sTOOLS=borland -sPYTHON_ROOT=c:\...\python22 install I would appreciate any suggestions. Regards, Steffen Rauh

3 4

boost erreur
by adel.essafi 20 Aug '04

20 Aug '04

hi list I am using boost to build a small tool that generate a (GRAPHICAL) graph showing dependecies between files ;) I donc guess why g++ returns to me this error. any input will be helpful thanks Ps : excuse me. I will copy all the source file and the error. you can find it a long. #include <qapplication.h> #include <qdir.h> #include <qcanvas.h> #include <iostream.h> #include <vector.h> #include <fstream.h> #include <string.h> #include <fstream> #include <string> #include <iostream> #include <algorithm> #include <string> #include <boost/config.hpp> #include <boost/graph/adjacency_list.hpp> #include <boost/property_map.hpp> #include <qstring.h> using namespace std; using namespace boost; enum vertex_first_name_t { vertex_first_name }; namespace boost { BOOST_INSTALL_PROPERTY(vertex, first_name); } template <class EdgeIter, class Graph> void who_owes_who(EdgeIter first, EdgeIter last, const Graph& G) { // Access the propety acessor type for this graph typedef typename property_map<Graph, vertex_first_name_t> ::const_type NamePA; NamePA name = get(vertex_first_name, G); typedef typename boost::property_traits<NamePA>::value_type NameType; NameType src_name, targ_name; while (first != last) { src_name = boost::get(name, source(*first,G)); targ_name = boost::get(name, target(*first,G)); cout << src_name << " owes " << targ_name << " some money" << endl; ++first; } } int main(int argc,char **argv) { // Create the graph, and specify that we will use std::string to // store the first name's. typedef adjacency_list<vecS, vecS, directedS,property<vertex_first_name_t, QString> > MyGraphType; MyGraphType G; property_map<MyGraphType, vertex_first_name_t>::type name= get(vertex_first_name, G); QApplication a( argc, argv ); QCanvas c; QDir dir (".") ; int filesnum=0; QStringList lst = dir.entryList( "*.cpp; *.h" ); for ( QStringList::Iterator it = lst.begin(); it != lst.end(); ++it ) { boost::graph_traits<MyGraphType>::vertex_descriptor v; name[v]=*it; add_vertex(v,G); cerr << *it << endl; filesnum++; } return 0; } g++ -o filetree main.cpp -I /usr/lib/qt-3.1/include -L /usr/lib/qt-3.1/lib -lqt Dans le fichier inclus à partir de /usr/include/c++/3.3.2/backward/iostream.h:31, à partir de main.cpp:4: /usr/include/c++/3.3.2/backward/backward_warning.h:32:2: attention : #warning This file includes at least one deprecated or antiquated header. Please consider using one of the 32 headers found in section 17.4.1.2 of the C++ standard. Examples include substituting the <X> header for the <X.h> header for C++ includes, or <sstream> instead of the deprecated header <strstream.h>. To disable this warning use -Wno-deprecated. main.cpp: Dans function « int main(int, char**) »: main.cpp:67: error: invalid initialization of reference of type 'const boost::property<vertex_first_name_t, QString, boost::no_property>&' from expression of type 'size_t' /usr/include/boost/graph/detail/adjacency_list.hpp:1916: error: in passing argument 1 of `typename Config::vertex_descriptor boost::add_vertex(typename Config::vertex_property_type&, boost::vec_adj_list_impl<G, C, B>&) [with Graph = boost::adjacency_list<boost::vecS, boost::vecS, boost::directedS, boost::property<vertex_first_name_t, QString, boost::no_property>, boost::no_property, boost::no_property, boost::listS>, Config = boost::detail::adj_list_gen<boost::adjacency_list<boost::vecS, boost::vecS, boost::directedS, boost::property<vertex_first_name_t, QString, boost::no_property>, boost::no_property, boost::no_property, boost::listS>, boost::vecS, boost::vecS, boost::directedS, boost::property<vertex_first_name_t, QString, boost::no_property>, boost::no_property, boost::no_property, boost::listS>::config, Base = boost::directed_graph_helper<boost::detail::adj_list_gen<boost::adjacency_list<boost::vecS, boost::vecS, boost::directedS, boost::property<vertex_first_name_t, QString, boost::no_property>, boost::no_property, boost::no_property, boost::listS>, boost::vecS, boost::vecS, boost::directedS, boost::property<vertex_first_name_t, QString, boost::no_property>, boost::no_property, boost::no_property, boost::listS>::config>]' gmake: *** [filetree] Erreur 1 *** échec *** Accédez au courrier électronique de La Poste : www.laposte.net ; 3615 LAPOSTENET (0,34/mn) ; tél : 08 92 68 13 50 (0,34/mn)

3 2

RE: [Boost-users] New design proposal for boost::filesystem
by Rozental, Gennadiy 20 Aug '04

20 Aug '04

> I propose the following design. The aim of boost::filesystem > should be to support the following coding idiom: I don't think this is a proper forum for this. Try development mailing list Gennadiy.

2 1

RE: [Boost-users] New design proposal for boost::filesystem
by Ben Hutchings 20 Aug '04

20 Aug '04

Carlo Wood <carlo(a)alinoe.com> wrote: > I propose the following design. The aim of boost::filesystem should > be to support the following coding idiom: > > * The programmer should take care to only handle two types of paths > in his application: > > 1) Complete paths > 2) Relative paths Are there any other types? > * The programmer will have to specifically tell the libary when a > constructed path is 'native' and when not. <snip> Sounds reasonable. > I propose two design changes: > > 1) 'native' is now not only a representation, but an *internal > state* of fs::path. (this has no effect on the representation > as returned by fs::path::string()). > 2) All 'complete' paths are automatically marked as 'native'. > > Examples, the following code is legal: > > fs::path p1("C:\foo\a.exe", native); // Fails on linux, > // success on windows. You have that backwards. "\f" and "\a" are not valid in a filename in Win32. Now if those backslashes are changed to "\\" this will work on Win32. Either version of the name will work on Linux because all characters except "\0" are valid in a Unix filename. > fs::path p2("/usr/src/a.out, native); // Fails on windows, > // success on linux. That should work on Windows too because "/" is supported as an alternative to "\\". <snip> > But in case more than one 'working' directory seems needed > then we can add support for that too by allowing to > construct paths with a reference to the (complete/native) > working directory. Ie, > > fs::path home_base(...); > // ... > > fs::path runtime_rcfile(home_base); > // ... > > runtime_rcfile = "myapplication/config/runtime/rc"; > > which is then relative to `home_base' instead of > a single, global 'working directory' (as now returned > by fs::current_path()). <snip> That doesn't seem right to me. Assignment should overwrite the old value rather than combining with it.

2 1

Problems using custom iterator with std::partial_sort_copy
by Martin Magnusson 20 Aug '04

20 Aug '04

I have written a custom iterator based on boost::iterator_facade, and I'm trying to use this with the std::partial_sort_copy algorithm. However, when compiling with g++ I get an error message saying that there is no matching function for a call to counting_iterator(Vertex<3>&). The way I interpret this is that my code is trying to construct a Vertex_Iterator from a Vertex object, but that was not my intenton. Is there something obvious that I have missed in my definition of Vertex_Iterator, which should be a bidirectional iterator? A Vertex_Iterator iterates over the vertices of a mesh, and therefore it needs a pointer to the mesh in question. Does an iterator which returns a Vertex<3>& when dereferenced need a constructor which takes only a Vertex<3>& argument? If so, how can I supply it with the pointer to its parent mesh? I hope that some of this makes sense. ////////////////////////////////////////////// // Some typedefs from the parent class 'Mesh': ////////////////////////////////////////////// class Mesh { public: class Vertex_Iterator; //... private: friend class Vertex_Iterator; typedef boost::property< boost::vertex_root_t, Vertex<3> > Vertex_Property; typedef boost::adjacency_list < boost::vecS, boost::vecS, boost::undirectedS, Vertex_Property > Vertex_Graph; typedef boost::property_map<Vertex_Graph, boost::vertex_root_t>::type Vertex_Data_Map; Vertex_Data_Map its_vertex_data; //... }; ////////////////////////////////////////////// // Definition of my custom iterator: ////////////////////////////////////////////// class Mesh::Vertex_Iterator: public boost::iterator_facade <Vertex_Iterator, Vertex_Graph::vertex_iterator, boost::bidirectional_traversal_tag, Vertex<3>& > { public: Vertex_Iterator() {} explicit Vertex_Iterator( const Mesh* const mesh, Vertex_Graph::vertex_iterator p ) : its_parent( mesh ), its_vertex( p ) {} private: friend class boost::iterator_core_access; Vertex<3>& dereference() const { return its_parent->its_vertex_data[*its_vertex]; } void increment() { ++its_vertex; } void decrement() { --its_vertex; } bool equal( Vertex_Iterator const& other) const { return this->its_vertex == other.its_vertex; } const Mesh* const its_parent; Vertex_Graph::vertex_iterator its_vertex; }; ////////////////////////////////////////////// // The call with an error looks like this: ////////////////////////////////////////////// void Mesh::subset() { //... Compare_Normals compare( its_mean_normal ); Vertex<3> temp_holder[size]; Vertex_Iterator first_vertex( this, boost::vertices( its_vertices ).first ); Vertex_Iterator last_vertex( this, boost::vertices( its_vertices ).second ); std::partial_sort_copy( first_vertex, last_vertex, temp_holder, temp_holder + size, compare ); //... } ////////////////////////////////////////////// // This is gcc's error message: ////////////////////////////////////////////// Building object file Mesh.o... /usr/include/c++/3.3.1/bits/stl_algo.h: In function `_RandomAccessIter std::partial_sort_copy(_InputIter, _InputIter, _RandomAccessIter, _RandomAccessIter, _Compare) [with _InputIter = Mesh::Vertex_Iterator, _RandomAccessIter = Vertex<3>*, _Compare = Mesh::Compare_Normals]': /home/martin/source/scan-match/Mesh.cpp:532: instantiated from here /usr/include/c++/3.3.1/bits/stl_algo.h:2661: error: no matching function for call to `boost::counting_iterator<size_t, boost::use_default, boost::use_default>::counting_iterator(Vertex<3>&)' /usr/local/include/boost/iterator/counting_iterator.hpp:169: error: candidates are: boost::counting_iterator<Incrementable, CategoryOrTraversal, Difference>::counting_iterator(Incrementable) [with Incrementable = size_t, CategoryOrTraversal = boost::use_default, Difference = boost::use_default] /usr/local/include/boost/iterator/counting_iterator.hpp:166: error: boost::counting_iterator<Incrementable, CategoryOrTraversal, Difference>::counting_iterator(const boost::counting_iterator<Incrementable, CategoryOrTraversal, Difference>&) [with Incrementable = size_t, CategoryOrTraversal = boost::use_default, Difference = boost::use_default] /usr/local/include/boost/iterator/counting_iterator.hpp:164: error: boost::counting_iterator<Incrementable, CategoryOrTraversal, Difference>::counting_iterator() [with Incrementable = size_t, CategoryOrTraversal = boost::use_default, Difference = boost::use_default] make[1]: *** [Mesh.o] Error 1 make: *** [default_target] Error 2 Thanks / martin

2 1

[bgl] kamada_kawai_spring_layout
by Gordon Smith 20 Aug '04

20 Aug '04

Just playing with the kamada_kawai_spring_layout , so far for any non trivial graph (more then 4 nodes) - it seems to just go into an infinite loop. The graph is undirected but would contain "circular" paths - thoughts? Gordon.

2 12

New design proposal for boost::filesystem
by Carlo Wood 20 Aug '04

20 Aug '04

I propose the following design. The aim of boost::filesystem should be to support the following coding idiom: * The programmer should take care to only handle two types of paths in his application: 1) Complete paths 2) Relative paths * The programmer will have to specifically tell the libary when a constructed path is 'native' and when not. A native path is accepted according to native rules and never gives any problem (exceptions) further on. A non-native path is checked according to the existing rules, which basically means that the programmer can set a default check routine that will in effect determine how portable the application will be. I propose two design changes: 1) 'native' is now not only a representation, but an *internal state* of fs::path. (this has no effect on the representation as returned by fs::path::string()). 2) All 'complete' paths are automatically marked as 'native'. Examples, the following code is legal: fs::path p1("C:\foo\a.exe", native); // Fails on linux, success on windows. fs::path p2("/usr/src/a.out, native); // Fails on windows, success on linux. std::cout << p1.native_file_string(); // ok, p1 is native. fs::path p3("foo/bar"); // Relative path, always succeeds. std::cout << p.string(); // ok std::cout << p.native_directory_string(); // Useless, but ok. fs::path p4(complete(p3)); // p4 is now "native", because now it is complete. std::cout << p4.native_directory_string(); // ok And the following will throw: std::cout << p4.string() // Not allowed because p4 is native (complete). Since there would be a default way defined of how a relative path is completed, all operation functions will accept both, relative and complete paths. For example: fs::path p1("C:\cygwin\usr/bin/ls", native); // Legal path on Cygwin. if (fs::exists(p1)) // Ok, access complete path. // For clarification fs::path p2("C:\cygwin\usr"); // Just an example fs::default_working_directory(p2); // fictuous function. fs::path p3("bin/ls"); // Portable representation (refers in fact to "C:\cygwin\usr\bin\ls.exe"). if (fs::exists(p3)) // Ok : exists() will make the path complete before testing. And this throws: fs::path p4("/bin/ls"); // Not allowed: this path has a root but is not marked 'native'. Setting the default_working_directory shall allways need to be done for each supported OS seperately. Of course you can set it to "/" on single root machines, and set it to "E:/" after extracting the 'E:' from the current path at application start up, in effect simulating a 'single root': int main() { fs::default_working_directory(fs::current_path().root_path()); // ... simulate single root machine below. fs::path p("bin/ls"); } The average application will work with relative paths, relative to some (native) base directory and next to that have some arbitrary, complete and thus native directories (ie, read from environment variables). But in case more than one 'working' directory seems needed then we can add support for that too by allowing to construct paths with a reference to the (complete/native) working directory. Ie, fs::path home_base(...); // ... fs::path runtime_rcfile(home_base); // ... runtime_rcfile = "myapplication/config/runtime/rc"; which is then relative to `home_base' instead of a single, global 'working directory' (as now returned by fs::current_path()). Most of the current boost::filesystem API can be preserved with this design. -- Carlo Wood <carlo(a)alinoe.com>

1 0

Question about static linking of boost
by Daniel Huber 20 Aug '04

20 Aug '04

Thanks to Stephen for helping me get my boost to link. I feel bad for not thinking of this solution myself. I have one final question/suggestion about the getting started document (which I found to be very well written). I compile and install boost using the following command: tools/build/jam_src/bin.linuxx86/bjam "-sTOOLS=gcc" "-sBUILD=release <runtime-link>static" --prefix=$HOME/linux/lib/boost install My goal was to create a minimal set of boost libraries (basically, I just need the filesystem non-debug static version. I was surprised to find that the above command installs shared libraries for some of the libraries, and it also installs both the -s- version of the static libraries and the -sd- version. It seems to me that this might be a bug, since I specifically requested only the release version. Am I confused about the difference between release version and debug version? Also, I am confused about the meaning of the -s- flag in the runtime section of the filename. Why do we get libraries that are shared but have the -s- flag? That seems like a contradiction. Similarly, if I don't specify static linking using the -sBUILD option (I assume the absence of the -s- flag means dynamic implicitly), I still get static libraries (i.e. .a extension files). These files seem to be unusable more or less (when I try to link against them, the compiler uses the shared libraries instead even if they don't exist). Anyway, all this is not much use to me, since I have it working now, but maybe someone could add a bit of text to the getting started document explaining this. Also, it might be helpful to give an example compile command that shows linking in the boost libraries. Thanks again for all your help. Sincerely, Daniel

2 1

Using boost::date_time for timezones
by Duncan Woods 19 Aug '04

19 Aug '04

Hello All, I'm new to Boost and interested in using it for timezone conversions but I'm not finding it terribly clear. An example, which works fine, uses something of the form: typedef boost::date_time::local_adjustor<ptime, -5, us_dst> us_eastern; 1. My problem is that I wish to use DST settings other than US and found definitions such as eu_dst_trait in local_timezone_defs.hpp but these seem unused and unusable. Would it require implementation of dst_rules.hpp type classes? If so, Boost would not give much advantage. 2. I would also have liked to specify dst & timezone combinations at run-time which dynamic_local_time_adjustor sounds like it might do - but it appears to have no members. 3. Will date_time use the 'tz database' in the future? Thank you in advance, Duncan

2 4

Inter Process Communication lib?
by Fernando Cacciola 19 Aug '04

19 Aug '04

Just a silly question... none of the boost libs provides cross platform IPC right? Txs Fernando Cacciola SciSoft

2 2

File size and attributes
by Surya Kiran Gullapalli 19 Aug '04

19 Aug '04

Hi all, I'm new to Boost library, working with boost-1.31.0. I'm trying to find out the size of the file and its attributes (hidden, readonly etc) using boost filesystem library. I had a go at the documentation, but did not get much help. Thanks in advance, Surya

1 0

mpl::switch question
by Istvan Buki 18 Aug '04

18 Aug '04

Hello, I've just tried to compile a small piece of code using mpl::switch_ starting from an example Aleksey Gurtovoy sent to this list a while ago. I slightly modified the code to use the boost::is_same<T,U> template but I can't make it compile. On the other hand, if I comment out the line in the switch_body containing the is_same template, it compile without any problem. I spent the whole day on this and I'm lost. Any help would be welcome. Thanks, Istvan PS: I'm using boost 1.30.2 and g++ 3.3.3. #include "boost/mpl/vector.hpp" #include "boost/mpl/switch.hpp" #include "boost/mpl/project1st.hpp" #include "boost/mpl/pair.hpp" #include "boost/mpl/always.hpp" #include "boost/mpl/bool.hpp" #include "boost/mpl/assert_is_same.hpp" #include "boost/type_traits/remove_reference.hpp" #include "boost/type_traits/add_const.hpp" #include "boost/type_traits/is_pointer.hpp" #include "boost/type_traits/is_reference.hpp" #include "boost/type_traits/is_same.hpp" using namespace boost::mpl ; using boost::mpl::_ ; struct MyType { typedef int type ; } ; int main() { typedef vector< pair< boost::is_pointer<_1>, _1 > // If the following line is commented out it compile. , pair< boost::is_same<_1, _1 >, boost::add_const<_1> > , pair< boost::is_reference<_1>, boost::remove_reference<_1> > , pair< always<bool_<true > >, boost::add_const<_1> > > switch_body; typedef switch_< switch_body, char& >::type t1; typedef switch_< switch_body, int* >::type t2; typedef switch_< switch_body, int >::type t3; BOOST_MPL_ASSERT_IS_SAME(t1, char); BOOST_MPL_ASSERT_IS_SAME(t2, int*); BOOST_MPL_ASSERT_IS_SAME(t3, int const); }

1 0

RE: [Boost-users] regex: match_results::const_reference is someti messub_match<Bidi Iterator>, sometimes const char?
by Richard Reitmeyer 17 Aug '04

17 Aug '04

Thanks for your help. (And all the effort that went into the regex library.) I'm using 1.31.0 (unpatched). After a more careful examination, my problem was that I was passing the wrong kind of allocator as a template parameter. Richard > -----Original Message----- > From: boost-users-bounces(a)lists.boost.org > [mailto:boost-users-bounces@lists.boost.org] On Behalf Of John Maddock > Sent: Tuesday, August 17, 2004 5:33 AM > To: boost-users(a)lists.boost.org > Subject: Re: [Boost-users] regex: > match_results::const_reference is sometimessub_match<Bidi > Iterator>, sometimes const char? > > > > The regex library is extremely helpful: thanks for all your efforts. > > > > I've got a problem with gcc and stlport and boost: when I try > > to get a submatch on other platforms, I can take my match object, m, > > and do things like > > > > m[n].str(); // works great elsewhere > > > > But with gcc / stlport, with BOOST_NO_STD_ALLOCATOR not defined, > > the const_reference type is defined as Allocator::const_reference > > (which is `const char'), rather than sub_match<BidiIterator>. > > So operator[] gives me a char, and life isn't so good. > > > > Any hints as to where I went astray? > > Works for me (with Boost-1.31.0 + latest patches), what > version are you > using? > > John. > > _______________________________________________ > Boost-users mailing list > Boost-users(a)lists.boost.org > http://lists.boost.org/mailman/listinfo.cgi/boost-users >

1 0

BGL-kamada_kawai_spring_layout
by Gordon Smith 17 Aug '04

17 Aug '04

Just playing with the kamada_kawai_spring_layout, so far for any non trivial graph (more then 4 nodes) - it seems to just go into an infinite loop. The graph is undirected but would contain "circular" paths - thoughts? Gordon.

1 0