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

RE: shared_ptr - implicit conversion to bool
by scleary＠jerviswebb.com 30 Jan '03

30 Jan '03

> From: wasekvesely <vaclav.vesely(a)volny.cz> > > I've found this in "shared_ptr.hpp": > > // implicit conversion to "bool" > typedef T * (this_type::*unspecified_bool_type)() const; > > operator unspecified_bool_type() const // never throws > { return px == 0? 0: &this_type::get; } > > Why this is better than: > > operator bool() const > { return px != 0; } > > What more can I do with somethin like this? Actually, that's used to prevent mistakes; bool is an integral type, so if you add an implicit conversion to bool, then your smart pointer class can participate in integral promotions. Which allows lots of bad code to compile. :) For an example, implicit conversion to an integral type defines all the comparision operators with any other integral-convertible class. e.g., shared_ptr<int> x; scoped_ptr<int> y; (x == y); // does not do what you might expect... For more information on the problem, see the following reference: More Effective C++, Scott Myers :: Item 28 :: Testing Smart Pointers for Nullness or search the Boost developer archives. -Steve

1 0

shared_ptr - implicit conversion to bool
by wasekvesely <vaclav.vesely＠volny.cz> 29 Jan '03

29 Jan '03

Hi, I've found this in "shared_ptr.hpp": // implicit conversion to "bool" typedef T * (this_type::*unspecified_bool_type)() const; operator unspecified_bool_type() const // never throws { return px == 0? 0: &this_type::get; } Why this is better than: operator bool() const { return px != 0; } What more can I do with somethin like this? Thank you, Wasek

3 2

RE: [Boost-Users] Re: Why is <boost::function>::_1 not in namespa ce boost?
by Tom Matelich 29 Jan '03

29 Jan '03

> > I think I agree with Markus that it is in fact broken. It's easy to > imagine dependency scenarios like this: > > lambda <-- libA <-\ > }--> App > bind <-- libB <-/ > > The guy writing App shouldn't have to be aware of the dependency > conflict between libA and libB. Isn't this only a problem within translation units? I suppose lambda and bind might be included in headers from libA and libB into a single translation unit in App, but seems unlikely. ----------------------------------------------------------------------- DISCLAIMER: Information contained in this message and/or attachment(s) may contain confidential information of Zetec, Inc. If you have received this transmission in error, please notify the sender by return email. -----------------------------------------------------------------------

2 1

Advice on using iterator_adapter
by Rob Smallshire 28 Jan '03

28 Jan '03

Boosters, I have implemented a 'virtual container' class (i.e. it looks and behaves like it contains Objects, but it fact it doesn't) which implements an iterator interface of what are essentially bidirectional iterators. The iterators I have now are not STL compliant and I only have 'mutable' versions of the iterators, with no corresponding const iterators. Some guidance on how to make the best use of boost::iterator adapter would be appreciated to create fully STL compliant iterator/const_iterator pairs: 1) Out current iterators (i.e. the Base type which will be adapted) return smart reference objects when dereferenced, rather than true C++ references to objects. Will iterator adapter work in this situation? 2) Furthermore, can I use the iterator_adapter to adapt our home-grown mutable iterator into a STL compliant const_iterator? Do I simply modify the Reference of the iterator_adapter template type to be a const version of my smart-reference class? Many thanks, Rob

2 1

Why is <boost::function>::_1 not in namespace boost?
by Markus Werle 28 Jan '03

28 Jan '03

Hi! the placeholders in boost/bin/placeholders.hpp are put in an anonymous namespace. This is awkward, since like this they are visible even in places where I do not want them to be so, e.g.: #include <all_of_boost_lambda> #include <all_of_boost_function> // not unlikely to use both [...] { using namespace boost::lambda; std::for_each(v.begin(), v.end(), std::cerr << _1 << constant(' ')); } In such cases the "_1" is ambigous, since boost::lambda::_1 and and <anonymous>::_1 are both visible. This means that code which used to include boost_lambda stuff may change its meaning (or get ambigous) just by including another boost header. Could we not place those placeholders into namespace boost, just to get rid of the ambiguity? Any caveats I overlooked? (using my own placeholder definitions and BOOST_BIND_NO_PLACEHOLDERS is not really a satisfying solution) Markus

4 12

RE: [Boost-Users] Boost 101?
by Ferdinand Prantl 24 Jan '03

24 Jan '03

> From: Beman Dawes [mailto:bdawes@acm.org] > Sent: Thursday, January 23, 2003 02:50 > > At one time we published a header dependency chart, but as Boost grew > it became unwieldy and we no longer produce it. > > There have been past discussions on how to meet this need, and the > consensus is that usually what is required is a library (rather than > header) dependency analysis. Yes, I agree, it is almost impossible considering which files to include and which not with the current size of the boost package. It is sensible to declare explicitely library dependencies (tree) and use them. Nevertheless, I created a perl script some time ago to display header and library dependencies in my library with references to boost headers. I hacked it quickly to work with boost only; if you are interested it is at http://groups.yahoo.com/group/boost/files/hdr_depend. > This would be fairly easy to produce, using a number of Boost's own > libraries, of course:-) > > But the holdup is that it is only a medium priority task, and lately > has been pushed aside in favor of higher priority work. If > someone else wants > to take this one over, I'd be glad to hand it off. We want to do it > entirely in C++, so any Booster can maintain it. Well, currently I'm happily unpacking all boost but I like the idea; for example, for such a project I would like to use smart pointers only... I could spend some time on such a tool if anyone is interested, firstly gathering the requirements and then developing it. Ferda > > --Beman

4 3

Problem with linking to boost::signals
by René Möhring 24 Jan '03

24 Jan '03

I try to use boost signals and have the following code from the tutorial. #include <boost/signal.hpp> #include <iostream> struct HelloWorld { void operator()() const { std::cout << "Hello, World!" << std::endl; } }; int main() { // Signal with no arguments and a void return value boost::signal<void ()> sig; // Connect a HelloWorld slot HelloWorld hello; sig.connect(hello); // Call all of the slots sig(); } This compiles but gives these errors while linking: /usr/bin/ld: warning: libstdc++.so.3, needed by /usr/local/lib//libboost_signals.so, may conflict with libstdc++.so.4 test.o: In function `_ZN5boost4slotINS_8functionIFvvENS_6detail8function8unusableES5_S5_S5_ S5_S5_S5_S5_S5_S5_EEEC1I10HelloWorldEERKT_': test.o(.gnu.linkonce.t._ZN5boost4slotINS_8functionIFvvENS_6detail8funct ion8unusableES5_S5_S5_S5_S5_S5_S5_S5_S5_EEEC1I10HelloWorldEERKT_+0xac): undefined reference to `boost::signals::detail::slot_base::create_connection()' test.o: In function `_ZN5boost7signal0IvNS_10last_valueIvEEiSt4lessIiENS_8functionIFvvENS_6 detail8function8unusableES9_S9_S9_S9_S9_S9_S9_S9_S9_EEE7connectERKNS_4s lotISA_EE': test.o(.gnu.linkonce.t._ZN5boost7signal0IvNS_10last_valueIvEEiSt4lessIi ENS_8functionIFvvENS_6detail8function8unusableES9_S9_S9_S9_S9_S9_S9_S9_ S9_EEE7connectERKNS_4slotISA_EE+0x83): undefined reference to `boost::signals::detail::signal_base_impl::connect_slot(boost::any const&, boost::any const&, std::vector<boost::signals::trackable const*, std::allocator<boost::signals::trackable const*> > const&)' test.o: In function `_ZN5boost7signal0IvNS_10last_valueIvEEiSt4lessIiENS_8functionIFvvENS_6 detail8function8unusableES9_S9_S9_S9_S9_S9_S9_S9_S9_EEEclEv': test.o(.gnu.linkonce.t._ZN5boost7signal0IvNS_10last_valueIvEEiSt4lessIi ENS_8functionIFvvENS_6detail8function8unusableES9_S9_S9_S9_S9_S9_S9_S9_ S9_EEEclEv+0x15): undefined reference to `boost::signals::detail::call_notification::call_notification(boost::sh ared_ptr<boost::signals::detail::signal_base_impl> const&)' test.o(.gnu.linkonce.t._ZN5boost7signal0IvNS_10last_valueIvEEiSt4lessIi ENS_8functionIFvvENS_6detail8function8unusableES9_S9_S9_S9_S9_S9_S9_S9_ S9_EEEclEv+0x155): undefined reference to `boost::signals::detail::call_notification::~call_notification()' test.o(.gnu.linkonce.t._ZN5boost7signal0IvNS_10last_valueIvEEiSt4lessIi ENS_8functionIFvvENS_6detail8function8unusableES9_S9_S9_S9_S9_S9_S9_S9_ S9_EEEclEv+0x1c7): undefined reference to `boost::signals::detail::call_notification::~call_notification()' test.o: In function `_ZN5boost7signal0IvNS_10last_valueIvEEiSt4lessIiENS_8functionIFvvENS_6 detail8function8unusableES9_S9_S9_S9_S9_S9_S9_S9_S9_EEEC2ERKS2_RKS4_': test.o(.gnu.linkonce.t._ZN5boost7signal0IvNS_10last_valueIvEEiSt4lessIi ENS_8functionIFvvENS_6detail8function8unusableES9_S9_S9_S9_S9_S9_S9_S9_ S9_EEEC2ERKS2_RKS4_+0xa6): undefined reference to `boost::signals::detail::signal_base::~signal_base()' test.o: In function `boost::signals::connection::~connection()': test.o(.gnu.linkonce.t._ZN5boost7signals10connectionD1Ev+0x16): undefined reference to `boost::signals::connection::disconnect() const' test.o: In function `boost::signals::detail::signal_base::signal_base(boost::function2<bool, boost::any, boost::any, boost::empty_function_policy, boost::empty_function_mixin, std::allocator<boost::function_base> > const&)': test.o(.gnu.linkonce.t._ZN5boost7signals6detail11signal_baseC2ERKNS_9fu nction2IbNS_3anyES4_NS_21empty_function_policyENS_20empty_function_mixi nESaINS_13function_baseEEEE+0x41): undefined reference to `boost::signals::detail::signal_base_impl::signal_base_impl(boost::func tion2<bool, boost::any, boost::any, boost::empty_function_policy, boost::empty_function_mixin, std::allocator<boost::function_base> > const&)' test.o: In function `_ZN5boost7signal0IvNS_10last_valueIvEEiSt4lessIiENS_8functionIFvvENS_6 detail8function8unusableES9_S9_S9_S9_S9_S9_S9_S9_S9_EEED2Ev': test.o(.gnu.linkonce.t._ZN5boost7signal0IvNS_10last_valueIvEEiSt4lessIi ENS_8functionIFvvENS_6detail8function8unusableES9_S9_S9_S9_S9_S9_S9_S9_ S9_EEED2Ev+0x11): undefined reference to`boost::signals::trackable::~trackable()' test.o(.gnu.linkonce.t._ZN5boost7signal0IvNS_10last_valueIvEEiSt4lessIi ENS_8functionIFvvENS_6detail8function8unusableES9_S9_S9_S9_S9_S9_S9_S9_ S9_EEED2Ev+0x1f): undefined reference to`boost::signals::detail::signal_base::~signal_base()' test.o: In function `void boost::checked_delete<boost::signals::detail::signal_base_impl>(boost:: signals::detail::signal_base_impl*)': test.o(.gnu.linkonce.t._ZN5boost14checked_deleteINS_7signals6detail16si gnal_base_implEEEvPT_+0x19): undefined reference to`boost::signals::detail::signal_base_impl::~signal_base_impl()' My compiler is gcc 3.2 on FreeBSD. g++ -o signals test.cpp -I/usr/local/include -L/usr/local/lib/ -lboost_signals Do you have an idea what I am doing wrong?

1 0

Specifying a case-insensitive rule for a regex algorithm
by Amittai Aviram 23 Jan '03

23 Jan '03

Hi! I am using boost::regex_grep() and boost::regex_search(), and I've figured out how to set a given rule to be case-insensitive when I transform it from a string into a rule -- i.e., -- boost::regex("search[[:space:]]string", boost::regbase::normal | boost::regbase::icase) This is great. But in many regular expression processing environments, you also have the option to specify case-insensitivity (or sensitivity) within the rule string itself. For instance, in Perl, you do it with the /i switch at the end of the rule string, and in some environments you can use (?i) to make the following characters case-insensitive when processed and then (?-i) to switch case sensitivity back on. I didn't see any mention of any such thing in the regex documentation. Is this option not available in boost::regex? Thanks! Amittai Aviram amittai(a)amittai.com

4 3

Problems using date-time library
by Keno 22 Jan '03

22 Jan '03

Greets all, I'm in need of using the Date-Time library but I'm encountering a few issues. I tried compiling the following simple code: ----------------- #include <boost/date_time/gregorian/gregorian.hpp> #include <iostream> using namespace boost::gregorian; int main() { date birthday(1979,3,26); return 0; } ------------------- and got the following error(s): --------------------Configuration: boost_date - Win32 Debug-------------------- Compiling... test_date.cpp d:\softdev\vstudio6\vc98\include\new(35) : warning C4290: C++ Exception Specification ignored Linking... test_date.obj : error LNK2001: unresolved external symbol "public: static unsigned long __cdecl boost::date_time::gregorian_calendar_base<struct boost::date_time::year_month_day_base<class boost::gregorian::greg_year,class boost::gregorian::greg_mon th,class boost::gregorian::greg_day>,unsigned long>::day_number(struct boost::date_time::year_month_day_base<class boost::gregorian::greg_year,class boost::gregorian::greg_month,class boost::gregorian::greg_day> const &)" (?day_number@?$gregorian_ca lendar_base@U?$year_month_day_base@Vgreg_year@gregorian@boost@@Vgreg_month@2 3@Vgreg_day@23@@date_time@boost@@K@date_time@boost@@SAKABU?$year_month_day_b ase@Vgreg_year@gregorian@boost@@Vgreg_month@23@Vgreg_day@23@@23@@Z) Debug/boost_date.exe : fatal error LNK1120: 1 unresolved externals Error executing link.exe. boost_date.exe - 2 error(s), 1 warning(s) ------------------------------- Any clues as to what could be happening here folks?? I'm using MSVC6 with SP5 + STLPort 4.5.3 on WinXP .: Keno :.

3 7

random number lib: msvc6, prob compiling random_demo.cpp w mt19937
by Nick Toze <geeky_nick＠yahoo.co.uk> 15 Jan '03

15 Jan '03

Hello, I had a slight problem compiling random_demo.cpp after I had changed the typedef to: typedef boost::mt19937 base_generator_type; I got this error: --------------------Configuration: random_demo - Win32 Debug-------------------- Compiling... random_demo.cpp d:\c_plus_plus\boost_1_29_0\boost\random\mersenne_twister.hpp(104) : error C2064: term does not evaluate to a function d:\c_plus_plus\boost_1_29_0\boost\random\mersenne_twister.hpp(75) : see reference to function template instantiation 'void __thiscall boost::random::mersenne_twister<unsigned long,624,397,31,-1727483681,11,7,-1658038656,15,-272236544,18,-948 541730>::seed(const unsigned int &)' being compiled Error executing cl.exe. Creating browse info file... random_demo.exe - 1 error(s), 0 warning(s) I fixed it by adding an 'L' to the first line of main, i.e. base_generator_type generator(42Lu);

1 0

Boost/Python: method overloading w/out default args
by Bruce Lowery <bruce_lowery＠yahoo.com> 15 Jan '03

15 Jan '03

Using Boost/Python, how does one overload methods? For instance: struct Bar { void func( const std::string& s ); void func( int i ); }; BOOST_PYTHON_MODULE( Foo ) { class_<Bar>( "Bar" ) .def( "func", &Bar::func ) .def( ??? ) ; } The reference documentation discusses the BOOST_PYTHON_MEMBER_FUNCTION_OVERLOADS macro, but it only seems to apply to methods with default arguments. --bruce

2 2

Boost/Python: returning internal reference
by Bruce Lowery <bruce_lowery＠yahoo.com> 15 Jan '03

15 Jan '03

I have a simple C++ class that I've wrapped using Boost/Python, v1.29.0 (on Win 2K), and that has a simple accessor returning a const ref to an internal string. In the python interpreter, the accessor raises an exception (see below),whereas another access that returns a copy of the string works. Could someone explain my mistake? --bruce ------ Bar.cpp -------- #include <string> class Bar { public: Bar( const std::string& s ) : m_s( s ) {} std::string get_s_copy() const { return m_s; } const std::string& get_s_ref() const { return m_s; } std::string m_s; }; #include <boost/python.hpp> using namespace boost::python; BOOST_PYTHON_MODULE(Foo) { class_<Bar>( "Bar", init< const std::string& >( args("s") ) ) .def( "get_s_copy", &Bar::get_s_copy ) .def( "get_s_ref", &Bar::get_s_ref, return_internal_reference<>() ) .def_readonly( "m_s", &Bar::m_s ) ; } ----- Python session ---- Python 2.2.2 (#37, Oct 14 2002, 17:02:34) [MSC 32 bit (Intel)] on win32 Type "help", "copyright", "credits" or "license" for more information. >>> import Foo >>> bar = Foo.Bar("hello") >>> bar.m_s 'hello' >>> bar.get_s_copy() 'hello' >>> bar.get_s_ref() Traceback (most recent call last): File "<stdin>", line 1, in ? TypeError: bad argument type for built-in operation >>>

2 3

Boost/Python: Solaris build errors
by Bruce Lowery <bruce_lowery＠yahoo.com> 15 Jan '03

15 Jan '03

In building the boost/python library on solaris using v6.2.p2 of the sunpro compiler, two compile-time errors occur, one in 'cast.hpp' and the other in 'inheritance.hpp'. Details are below. --bruce ----------------- 1 -------------------- boost/boost/python/cast.hpp: line 65: Error: Default arguments cannot be added in later declarations of the template function in the same scope. 57 template <class Target, class Source> 58 inline Target* downcast(Source* p, yes_convertible) 59 { 60 return static_cast<Target*>(p); 61 } 62 63 template <class Target, class Source> 64 inline Target* downcast(Source* p, no_convertible, boost::type<Target>* = 0) *** 65 { *** 66 typedef typename base_type_traits<Source>::type base; 67 return (Target*)detail::downcast<base>(p, convertible<Source*>::check((base*)0)); 68 } ----------------- 2 -------------------- boost/boost/python/object/inheritance.hpp: line 134: Error: ")" expected instead of ">". 130 template <class Source, class Target> 131 inline void register_conversion( 132 // We need this parameter because CWPro7 can't determine 133 // which is the base reliably. *** 134 bool is_downcast = !cast_generator<Source,Target>::is_upcast *** 132 133 // These parameters shouldn't be used, they're an MSVC bug workaround 134 , Source* = 0, Target* = 0) 135 { 136 typedef typename cast_generator<Source,Target>::type generator; 137 138 add_cast(python::type_id<Source>() 139 , python::type_id<Target>() 140 , &generator::execute 141 , is_downcast); 142 }

2 2

RE: [Boost-Users] [BGL] missing include in boost/graph/kruskal_mi n_spanning_tree.hpp
by bjorn.karlsson＠readsoft.com 13 Jan '03

13 Jan '03

> From: Jeppe N. Madsen [mailto:yg-boost-users@m.gmane.org] > > Hi, > > kruskal_min_spanning_tree uses bgl_named_params defined in > boost/graph/named_function_params.hpp, which is not included.... Fixed in CVS. Thanks! Bjorn

1 0

BGL: filtered graphs, write_graphviz, and out of memory messages
by sa3h1n <sa3h1n＠yahoo.com> 13 Jan '03

13 Jan '03

Hi all: I am seeing a strange interaction between filtered graphs and write_graphviz. Briefly, if I try to write out a filtered graph via write_graphviz, compilation ends with an out-of-memory message. A necessary condition for this to happen is that the value_type of the property map I am filtering cannot be of a native type -- int is ok, std::string is not. Example code is appended below. I am using Codewarrior Dev Tools 8.0 on WinXP Pro. Given that the code does not even compile, I am not sure how to move beyond this point. Any ideas? TIA, - S ----------------------------------------------------------------- #include "boost/config.hpp" #include <fstream.h> #include <iostream> // for std::cout #include <utility> // for std::pair #include <algorithm> // for std::for_each #include "boost/utility.hpp" // for boost::tie #include "boost/graph/graph_traits.hpp" // for boost::graph_traits #include "boost/graph/adjacency_list.hpp" #include "boost/graph/graphviz.hpp" #include "boost/graph/filtered_graph.hpp" using namespace boost; namespace boost { enum edge_prop_t { edge_prop = 389457 }; BOOST_INSTALL_PROPERTY(edge, prop); } template <typename T> struct positive_edge_prop { positive_edge_prop() {}; positive_edge_prop(T t) : m_prop(t) {}; template <typename Edge> bool operator() (const Edge& e) const { return "foo" == get(m_prop, e); } T m_prop; }; int main(void) { // create an edge property // if std::string is replaced with int, and some attendant changes are made, the code compiles typedef property<edge_prop_t, std::string> EdgeProperty; // create a simple graph type typedef adjacency_list<vecS, vecS, bidirectionalS, no_property, EdgeProperty> Pgraph; enum { A, B, C, D, E, F, N }; const int num_vertices = N; typedef std::pair<int, int> Edge; Edge edge_array[] = { Edge(A, B), Edge(B, C), Edge(C, D), Edge(D, E), Edge(E, F), Edge(F, A), }; const int num_edges = sizeof(edge_array)/sizeof(edge_array[0]); Pgraph pg(edge_array, edge_array + num_edges, num_vertices); // Add edge properties typedef boost::property_map<Pgraph, edge_prop_t>::type ewt; ewt ew = get(edge_prop_t(), pg); boost::graph_traits<Pgraph>::edge_iterator ei, ei_end; boost::tie(ei, ei_end) = edges(pg); for (ei; ei != ei_end; ei++) put(ew, *ei, "bar"); std::map<std::string,std::string> graph_attr, vertex_attr, edge_attr; boost::write_graphviz(std::cout, pg); positive_edge_prop<ewt> filter(get(edge_prop_t(), pg)); filtered_graph<Pgraph, positive_edge_prop<ewt> > fg(pg, filter); boost::write_graphviz(std::cout, fg); // if this line is commented out, the code compiles just fine return 0; } -----------------------------------------------------------------

1 0

Next release date???
by Alejandro Escalante Medina 13 Jan '03

13 Jan '03

Does anyone have an idea for the date of the next release??? Thanks, Alex _________________________________________________________ Do You Yahoo!? La mejor conexión a internet y 25MB extra a tu correo por $100 al mes. http://net.yahoo.com.mx

2 1

Problem with g++ 2.95.3 and boost-example file_dependencies.cpp
by Michael Brinkmann 12 Jan '03

12 Jan '03

Hi, I am trying to get the "file_dependencies"-example compiled and running with the g++-version mentioned above. But, in the function "dijkstra_shortest_paths_no_init()" an exception is thrown and the program terminates prematurely. When hunting the bug with gdb, I get the message "Cannot access memory at address ..." for the parameter "weight", perhaps this is the reason for the exception. Is this a known problem? Will an upgrade to a higher g++-version help me, or has someone another hint? Thanks in advance Michael

1 0

threads - using boost.threads in win32 without use of dll?
by Phil Nash 12 Jan '03

12 Jan '03

Hi, I understand that on the win32 platforms boost.threads requires a dll to be build - threadmon. I seem to remember that this is there as a workaround to help support thread local storage (but I can't seem to find mention of it in the docs now, so I am not 100% sure of this). If this is the case, is it possible to use boost.threads *without* the dll if thread local storage is not being used? Regards, [)o IhIL..

1 0

boost::signal and thread-safety
by svanechel <svanechel＠yahoo.com> 11 Jan '03

11 Jan '03

Hi, The signal FAQ says that currently boost::signals aren't thread-safe. I'm guessing this applies to connecting/disconnecting slots from different threads and calling a signal from multiple threadsl, right? If synchronize access when connecting slots and make sure that only a single thread calls a certain signal, can I then use boost::signal to fire events from objects that do processing in a worker thread? As a secondary question: how far in the future will this thread- safety support be added? Thanks in advance, Sven

2 2

[BGL] missing include in boost/graph/kruskal_min_spanning_tree.hpp
by Jeppe N. Madsen 11 Jan '03

11 Jan '03

Hi, kruskal_min_spanning_tree uses bgl_named_params defined in boost/graph/named_function_params.hpp, which is not included.... rgds Jeppe

1 0

(no subject)
by Wilken, Marc 11 Jan '03

11 Jan '03

Hello, does anyone knows, if a construct like out(blitz::Range(0,2))=inp(blitz::Range(0,2)); (that means accesing and assigning of matrix/vector ranges) exist for boost::numeric::ublas too? Marc This e-mail contains confidential information for the exclusive attention of the intended addressee. Any access of third parties to this e-mail is unauthorised. Any use of this e-mail by unintended recipients such as copying, distribution, disclosure etc. is prohibited and may be unlawful. When addressed to our clients the content of this e-mail is subject to the General Terms and Conditions of GL's Group of Companies applicable at the date of this e-mail. GL's Group of Companies does not warrant and/or guarantee that this message at the moment of receipt is authentic, correct and its communication free of errors, interruption etc.

2 1

BGL, vertex_iterators, listS vs vecS
by sa3h1n <sa3h1n＠yahoo.com> 11 Jan '03

11 Jan '03

BGL gurus: I would like to use listS for vertex and edge lists because of iterator stability. However, if I run the following code: ------------ // use listS typedef adjacency_list<listS, listS> Pgraph; // use vecS typedef adjacency_list<vecS, vecS> Pgraph2; int num_vertices = 2; Pgraph pg (num_vertices); boost::graph_traits<Pgraph>::vertex_iterator vi, vi_end; boost::tie(vi, vi_end) = vertices(pg); Pgraph2 pg2 (num_vertices); boost::graph_traits<Pgraph2>::vertex_iterator vi2, vi_end2; boost::tie(vi2, vi_end2) = vertices(pg2); for (vi, vi2; vi != vi_end && vi2 != vi_end2; vi++, vi2++) { std::cout << *vi << std::endl; std::cout << *vi2 << std::endl; } ------------ I get the following output: ------------ 0x00146e60 <- this is *vi which uses listS 0 <- this is *vi2 which uses vecS 0x00146e78 <- this is *vi which uses listS 1 <- this is *vi2 which uses vecS ------------ One of the problems with this is when I try to display the graph using write_graphviz. Vertex indices turn out be 0x0014e60 etc instead of nice numbers like 0, 1, etc. Tracing execution through the debugger, it appears that vi and vi_end are of type generic_iterator(), whereas vi2 and vi2_end are of type iterator_adaptor() defined in boost/iterator_adaptors.hpp . Unfortunately, this is where my c++ expertise peters out. So what is happenning here? And is there anything I can do to ensure iterator stability while using vecS? Any help will be much appreciated, S

2 2

Re: [boost] boost version 1.27 and 1_25_0
by David Abrahams 10 Jan '03

10 Jan '03

Chunhui Pan <chpan(a)milkyway.gsfc.nasa.gov> writes: > Hi, > > If anyone can give an instruction on how to build > boost 1.27 lib and boost_1_25_0 lib on Red Hat7.2? 1.27 instructions are at: http://cvs.sourceforge.net/cgi-bin/viewcvs.cgi/*checkout*/boost/boost/tools… And 1.25 instructions are at: http://cvs.sourceforge.net/cgi-bin/viewcvs.cgi/*checkout*/boost/boost/tools… HTH, -- David Abrahams dave(a)boost-consulting.com * http://www.boost-consulting.com Boost support, enhancements, training, and commercial distribution

1 0

multi_array dinamically built
by Ramón Casero Cañas 10 Jan '03

10 Jan '03

Hi, I would like to dinamically create a multi_array. Something like typedef boost::multi_array<bool, 2> BoolMatrix; typedef boost::multi_array<bool, 2>::index indexBoolMatrix; boost::array<BoolMatrix::index, 2> shape = {{ 5, 8 }}; BoolMatrix *A; // NO MEMORY ALLOCATED ... A = new BoolMatrix(shape); // MEMORY ALLOCATED A.assign(data,data+40); // ELEMENTS INITIALIZED The compiler (gcc 2.95.4) says that this kind of things are wrong. Or didn't I understand how the memory allocation in multi_array works? Cheers, Ramón.

1 0

NEWBIE - building on linux
by gcs_99 <gregsteele＠woodsidenet.com> 10 Jan '03

10 Jan '03

Can anyone point me to directions for building boost under linux. I have installed bjam. When I type bjam -sTOOLS=gcc in my boost root dir I get tons of errors. I understand that much of boost is implemented as templates in header files. Is there an install script to place these in the correct location for gcc? Thanks. G

4 6

request for solaris/forte toolset jamfile
by Dr Bruce Lowery <bruce_lowery＠yahoo.com> 10 Jan '03

10 Jan '03

If anyone has a somewhat working jamfile for solaris/forte, could you post a copy of it? --bruce

4 3

Installing boost regex
by Alisdair Meredith 10 Jan '03

10 Jan '03

We're (still!) in the process up upgrading our software to the latest borland compiler, BCB6 I am now at the stage of trying to get the boost libraries integrated into our builds. After running bjam I get some libboost_regex libs, but these don't seem to be used. Reading the separate install procedure under the regex library, I find another install procedure that produces a set of libraries targetted at BCB that do solve my problems. Is this the correct procedure? Is there any way to integrate this with the jam scripts in the future? If not, should we add a link under installing boost to the separate procedure for the regex libraries? -- AlisdairM Team Thai Kingdom

2 2

stop a thread
by hlh771 <hlh771＠yahoo.com.cn> 10 Jan '03

10 Jan '03

How to stop(or cancel) a thread? Suppose I have a main thread and a worker thread, the worker thread does something looply,and when the main thread get a message indicating the process should exit, I want stop the worker thread,how to do that?I have read other thread library,their thread class supply a method called stop or cancel which stops the thread. Their Implementation is:the thread class has a data member indicating whether the thread should stop,and the thread function looply check the data member whether the thread should stop.Generally the data member's type is a bool or Event. Is it the best way to stop the thread? How the boost to deal with this? Thanks in advance.

1 0

Starting to use unit tests
by Scott Nesbitt 10 Jan '03

10 Jan '03

I am in to process of adding the boost test framework to our project, and am looking for advice or samples for how best to do this. Specifically, we have multiple executables sharing several libraries. I was thinking about having a different test suite for each library, but don't see much documentation on using multiple test suites. Any advice or pointers? Thanks, Scott Nesbitt scott.nesbitt(a)yahoo.com __________________________________________________ Do you Yahoo!? Yahoo! Mail Plus - Powerful. Affordable. Sign up now. http://mailplus.yahoo.com

2 1

bind() and argument_type?
by Duane Murphy 09 Jan '03

09 Jan '03

I am trying to use bind() with std::not1() (I think this is what I want to do :-). But std::not1() gets a compiler error because bind doesnt have an argument_type typedef. bool is_something( int x, int y ) { return something( x, y ); } { ... if ( std::not1( boost::bind( is_something, 3, _1 ) ) ) { // not something } } Am I approaching this incorrectly? Does bind support argument_type? Thanks for the advice, ...Duane

2 1

RE: [Boost-Users] Problem with BOOST_STATIC_ASSERT
by Randy Bowen 09 Jan '03

09 Jan '03

Thanks to John and Richard Cox for investigating this. > You will be pleased to know that it works correctly with VC7.1. > > John Maddock.

1 0

RE: [Boost-Users] Problem with BOOST_STATIC_ASSERT
by Randy Bowen 09 Jan '03

09 Jan '03

Thanks for investigating this, John. I don't have access to the beta VC7.1 compiler. Do you (or does anyone else) know if this will still be a problem with the forthcoming version of the Microsoft compiler? Randy Bowen -----Original Message----- From: John Maddock [mailto:john_maddock@compuserve.com] Sent: Wednesday, January 08, 2003 4:17 AM To: Boost-Users(a)yahoogroups.com Subject: Re: [Boost-Users] Problem with BOOST_STATIC_ASSERT > Some further follow-up on this. I've created a minimal program that > exhibits the new problem behavior: > > /******************************************/ > #include <boost/type_traits.hpp> > #include <boost/static_assert.hpp> > > template < typename ValueType > > struct A > { > BOOST_STATIC_ASSERT(boost::is_integral<ValueType>::value); > }; > > extern "C" int main(int argc, char* argv[]) > { > return 0; > } > /******************************************/ > > The fact that this assertion fails is counter-intuitive to me, since > there is no instantiation to compile. Note that this behavior seems > limited to the use of "is_integral" as the type predicate. Constructing > a similar program using "is_POD" behaves as expected: I can reproduce that with both VC6 and VC7, all the other compilers I have access to are OK with it. It's clearly a compiler bug, because the assert is triggered even though no instance of the template class is ever created. Unfortunately I don't have any kind of workaround :-( John Maddock http://ourworld.compuserve.com/homepages/john_maddock/index.htm Yahoo! Groups Sponsor ADVERTISEMENT Info: <http://www.boost.org> Wiki: <http://www.crystalclearsoftware.com/cgi-bin/boost_wiki/wiki.pl> Unsubscribe: <mailto:boost-users-unsubscribe@yahoogroups.com> Your use of Yahoo! Groups is subject to the Yahoo! Terms of Service.

2 1

RE: [Boost-Users] Problem with BOOST_STATIC_ASSERT
by richard_cox＠dell.com 09 Jan '03

09 Jan '03

On Wednesday, January 08, 2003 5:29 PM, Randy Bowen [mailto:rbowen@stamps.com] wrote: > -----Original Message----- > From: > Sent: > To: Boost-Users(a)yahoogroups.com > Subject: RE: [Boost-Users] Problem with BOOST_STATIC_ASSERT > > > Thanks for investigating this, John. I don't have access to > the beta VC7.1 compiler. Do you (or does anyone else) know > if this will still be a problem with the forthcoming version > of the Microsoft compiler? > > Randy Bowen > > > > -----Original Message----- > From: John Maddock [mailto:john_maddock@compuserve.com] > Sent: Wednesday, January 08, 2003 4:17 AM > To: Boost-Users(a)yahoogroups.com > Subject: Re: [Boost-Users] Problem with BOOST_STATIC_ASSERT > > > > Some further follow-up on this. I've created a minimal > program that > > exhibits the new problem behavior: > > > > /******************************************/ > > #include <boost/type_traits.hpp> > > #include <boost/static_assert.hpp> > > > > template < typename ValueType > > > struct A > > { > > BOOST_STATIC_ASSERT(boost::is_integral<ValueType>::value); > > }; > > > > extern "C" int main(int argc, char* argv[]) > > { > > return 0; > > } > > /******************************************/ > > > > The fact that this assertion fails is counter-intuitive to > me, since > > there is no instantiation to compile. Note that this > behavior seems > > limited to the use of "is_integral" as the type predicate. > Constructing > > a similar program using "is_POD" behaves as expected: > > I can reproduce that with both VC6 and VC7, all the other > compilers I have access to are OK with it. It's clearly a > compiler bug, because the assert is triggered even though no > instance of the template class is ever created. Unfortunately > I don't have any kind of workaround :-( > > John Maddock > http://ourworld.compuserve.com/homepages/john_maddock/index.ht m The code as quoted built without error OK on VC++7.1 RC2 (Final Beta) here... using snapshot of boost taken last weekend. Richard Cox Senior Software Developer Dell Technology Online All opinions and statements mine and do not in any way (unless expressly stated) imply anything at all on behalf of my employer

1 0

Re: [Boost-Users] Re: Re: Slight Alteration to Boost Directory Structure
by William E. Kempf 09 Jan '03

09 Jan '03

> From: "Simon Bailey <conic(a)bigpond.com>" <conic(a)bigpond.com> > --- In Boost-Users(a)yahoogroups.com, "Paul Mensonides" <yg-boost- > users@m...> wrote: > > "William E. Kempf" <wekempf@c...> wrote in message > > > > > That still misses the point. Boost currently won't help you to > install > > the library, but installation certainly wouldn't entail copying the > entire > > $BOOST_ROOT tree to the $INCLUDE directory on your system. You'd > only copy > > the $BOOST_ROOT/boost tree there. > > > > Yes, and this is why I said that it hasn't been a problem for me. > However, > > I see where the OP is coming from as well. > > > > Paul Mensonides > > Paul understands me correctly. I confused the issue by making two > requests at once. > > Mainly, I want to include other libraries alongside boost and provide > a single include path. That is the first wish. Bill has pointed out > that I could move BOOST_ROOT/boost into my INCLUDE directory. I > didn't know that - I thought it might break things. > > The second thing I was suggesting (and this is quite minor) was that > we make the change to include smart_ptr.hpp (etc.) and thread.hpp > (etc.) at the same directory level. I thought that the > subdirectories (like /thread) were the way of the future. I didn't > realise that the plan was to move all the headers into > BOOST_ROOT/boost. > > So it seems my first wish is not needed and my second wish is already > on the drawing board. :-) No, it's never been planned to move all headers into $BOOST_ROOT/boost. That would be problematic at best. The headers are in subdirectories to avoid name clashes, and that's how it should be for large libraries and/or libraries that use common names. What many of these libraries do, and what Boost.Threads will do with the next release, however, is to have a *single* include header in $BOOST_ROOT/boost that includes all the other headers from the specific sub-library. This (mostly) avoids the dangers of name clashes, gives a simple single include file for many uses, yet retains the ability for users to include only what they need to in order to reduce compilation time, etc. For example, both of the following would be legal: #include <boost/thread/thread.hpp> #include <boost/thread/mutex.hpp> #include <boost/thread/condition.hpp> // ... code using boost::thread, boost::mutex and // boost condition - or - #include <boost/thread.hpp> // ... code using boost::thread, boost::mutex and // boost condition William E. Kempf wekempf(a)cox.net

1 0

Re: [Boost-Users] Re: Re: Slight Alteration to Boost Directory Structure
by William E. Kempf 09 Jan '03

09 Jan '03

> From: "Simon Bailey <conic(a)bigpond.com>" <conic(a)bigpond.com> > Suddenly I remember why I wanted the directory structure change... > > I could achieve what I want by moving BOOST_ROOT/boost into my > INCLUDE directory. But the disadvantage with that is that I need to > perform a move operation. > > *** In my current development environment, my source control > directory structure mirrors my release directory structure exactly. > *** > > So, in source control I have: > > /include > --/BOOST_ROOT > ----/boost <- headers are here > --/my_lib <- headers are here > > But, in my release I want to have: > /include > ----/boost <- headers are here > ----/my_lib <- headers are here > > I was suggesting the change so that source control and relase could > be the same. This would necessitate that documentation and source > for boost would have to be within the boost (include) directory. > Some might find that a worse alternative I suppose. I think most would. I know I do. William E. Kempf wekempf(a)cox.net

1 0

which boost lib to use for generic list sort by member function ?
by Philippe Guglielmetti 09 Jan '03

09 Jan '03

I try to define a "sortby" method that would sort a list using any member function that fits some constraints (taking no parms, returning a result that can be compared using the > operator and being const) I wrote this: template <class T> class mylist:public std::list<T> { // ... template <typename T2, typename R, R (T2::*F)() const> struct cmp: public std::binary_function<T,T,bool> { bool operator()(const T& f1, const T& f2) {return (f1.F)() > (f2.F)();}; }; template <typename T2, typename R> bool sortby(R (T2::*F)() const) { sort(cmp<T,R,F>()); return true; }; }; // class mylist using VC6 (I know...) I get these 3 errors: algos.h(356) : error C2440: 'specialization' : cannot convert from 'double (__thiscall cadoo::cadFace::*)(void) const' to 'double (__thiscall cadoo::cadFace::*)(void)' algos.h(356) : error C2975: 'cmp' : invalid template argument for 'F', constant expression expected algos.h(356) : error C2514: 'dyl::List<class cadoo::cadFace>::cmp<class cadoo::cadFace,double,0>' : class has no constructors any idea ? I also checked boost.org for help on functors, lambda ... but could not figure out how to use them simply for my application. thanks! Philippe Guglielmetti

1 0

Boost::python + vc + sgistl problem
by yg-boost-users＠m.gmane.org 09 Jan '03

09 Jan '03

I use Boost::Python module with VC and SGI stl. But I got some compile errors like mix\max redefine.... Who know how to solve this problem??? -- CNews 0.4

1 0

HELP: Command line sample.
by Lichun Bao 09 Jan '03

09 Jan '03

Hi, Can someone please provide a sample single command line compile using g++ that generates the final executable binary for a test example? For example, in the unpacked BOOST directory, how to do something like: g++ <options> libs/graph/example/bfs-example.cpp -o \ /tmp/bfs-example for the final executable? Thanks, Lichun

1 0

Re: [Boost-Users] Boostifying GluCat
by William E. Kempf 08 Jan '03

08 Jan '03

Sorry folks, this posting never should have made the list. As one of the moderators, I hit the wrong button here. Sorry for any inconvenience. William E. Kempf wekempf(a)cox.net

1 0

Boostifying GluCat
by Paul C. Leopardi 08 Jan '03

08 Jan '03

Bjorn, GluCat is a new numeric template class library, hopefully destined for promotion to Boost. I would like to know where there is some documentation on how to: 1. Add concepts. Is there such a thing as a non-commutative algebra concept? How do I check concepts? 2. Add numeric limits. Limits of eg. framed_multi<T> would default to limits of T. 3. Add numeric promotion_traits. What are promotion_traits ? There is currently a discussion on ublas-dev mailing list about promotion_traits within uBLAS. 3. Add numeric type_traits. Thanks On Thu, 9 Jan 2003 00:18, you wrote: > Paul, > unfortunately this post is off topic for Boost-Users, and is thus > rejected. While it may well be of interest to some of our subscribers, > and the usage of uBLAS is a nice success story, we do not allow > announcement postings - only topics related to usage of Boost libraries. > > Regards, > Bjorn Karlsson > Moderator

1 0

Re: [Boost-Users] Re: Re: Slight Alteration to Boost Directory Structure
by William E. Kempf 08 Jan '03

08 Jan '03

> From: "Paul Mensonides" <yg-boost-users(a)m.gmane.org> > "William E. Kempf" <wekempf(a)cox.net> wrote in message > news:20030108172855.VIMZ4411.lakemtao01.cox.net@smtp.east.cox.net... > > > > I believe that he is talking about the root Boost directory which is the > > > parent directory of the Boost headers. If, for instance, I have a > directory > > > "include" that I want to put all my library headers in and have a single > > > environment variable to specify the header search paths. This cannot be > > > done with Boost because the Boost headers are located at > > > "boost/boost/*.hpp", not "boost/*.hpp". Furthermore, you can't even put > > > Boost in the same directory as other includes, because you might get a > > > conflict between "include/boost" and "include/boost/boost". It doesn't > > > bother me personally, but I see the point that he is getting at. > > > > If that's the point (which doesn't seem to be, since he talks about "new > boost libraries" doing the right thing), I can't agree. Moving Boost > headers out of a boost subdirectory and into the top-level include directory > (so <boost/shared_ptr.hpp> becomes <shared_ptr.hpp>) would only result in > name clashes with other headers. > > That is not what he (or I) means. He is suggesting that the path to > "shared_ptr.hpp" should be "$INCLUDE/boost/shared_ptr.hpp" rather than > "$INCLUDE/boost/boost/shared_ptr.hpp", which is the way it is now. > Everything would still be the same for including headers except that only a > single search path is necessary to setup with the compiler. That's the way it is now *only* if you set your include directories incorrectly. When they are properly set to $BOOST_ROOT then it's only <boost/shared_ptr.hpp>, not <boost/boost/shared_ptr.hpp>. William E. Kempf wekempf(a)cox.net

2 1

Re: [Boost-Users] Re: Slight Alteration to Boost Directory Structure
by William E. Kempf 08 Jan '03

08 Jan '03

> From: "Paul Mensonides" <yg-boost-users(a)m.gmane.org> > Date: 2003/01/08 Wed PM 12:18:35 EST > To: boost-users(a)yahoogroups.com > Subject: [Boost-Users] Re: Slight Alteration to Boost Directory Structure > > "William E. Kempf" <wekempf(a)cox.net> wrote in message > > > > 2. Each library bundled with boost means another #include search > > > path needs to be added. > > > > No, there's only one path added for all Boost libraries, > $BOOST_ROOT/boost. > > Here I think he is talking about libraries other than Boost. How can anything we do to the Boost directory structure effect this? I don't really think that's what he meant. > > > Proposed solution: > > > 1. Create a single boost include directory instead of having the two > > > nested ones that there are now. > > > > What two nested ones? > > I believe that he is talking about the root Boost directory which is the > parent directory of the Boost headers. If, for instance, I have a directory > "include" that I want to put all my library headers in and have a single > environment variable to specify the header search paths. This cannot be > done with Boost because the Boost headers are located at > "boost/boost/*.hpp", not "boost/*.hpp". Furthermore, you can't even put > Boost in the same directory as other includes, because you might get a > conflict between "include/boost" and "include/boost/boost". It doesn't > bother me personally, but I see the point that he is getting at. If that's the point (which doesn't seem to be, since he talks about "new boost libraries" doing the right thing), I can't agree. Moving Boost headers out of a boost subdirectory and into the top-level include directory (so <boost/shared_ptr.hpp> becomes <shared_ptr.hpp>) would only result in name clashes with other headers. William E. Kempf wekempf(a)cox.net

2 1

Re: [Boost-Users] Slight Alteration to Boost Directory Structure
by William E. Kempf 08 Jan '03

08 Jan '03

> From: "Simon Bailey <conic(a)bigpond.com>" <conic(a)bigpond.com> > > I've been using boost for about a year now. I upgraded from 27 to > 28 and appreciated the backwards compatability. But I'm wondering > if a slight modification to the directory structure is in order? > > I have written a cross-platform class library. The library provides > high-level business objects and comes bundled with boost 1.28.0 to > provide support for threads, CRC, shared_ptr, and function (or > whatever else the end user might want). > > Currently, users need to add two directories to their #include > search path: > include/boost > include/sb_library > > Then, in their source code: > #include <boost/shared_ptr.hpp> > #include <boost/thread/thread.hpp> > #include <sb_library/business_facet/business_object.h> > > Two problems (albeit very minor ones!) > 1. Boost has two levels of includes (due to it's history of course). Two levels? There's only one, $BOOST_ROOT/boost. To avoid name clashes, many libraries are located in library specific directories off of this, as you see with <boost/thread/thread.hpp>. Something that many libraries do (and Boost.Threads will be updated to do shortly) is to have a single header in the root include directory that includes all of the headers from the library directory, thus for Boost.Threads you'd include only <boost/thread.hpp> instead of <boost/thread/thread.hpp>, <boost/thread/mutex.hpp> etc. > 2. Each library bundled with boost means another #include search > path needs to be added. No, there's only one path added for all Boost libraries, $BOOST_ROOT/boost. > Proposed solution: > 1. Create a single boost include directory instead of having the two > nested ones that there are now. What two nested ones? > 2. Put all boost headers into a sub-directory of boost > eg <boost/ptr/shared_ptr.hpp>, <boost/thread/thread.hpp>. (It's a > small change for existing users to change their includes). This is mostly how it is now... though many headers don't live in library specific subdirectories (which is probably how it should be). > 3. Place all src, documentation etc. into a directory within boost > called "private" or something. > > Suggested directory structure: > > --include // This is the only directory needed to be specified in > the #include search path > ----other_library1 > ----other_library2 > ----boost > -------ptr > -------thread > -------private > ----------documentation > ----------src > ----------thread > > In other words, you will have <boost/ptr/...> and <boost/thread/...> > but never <boost/private/...>. This is just a variation on what's currently done, and I can't see how it buys you anything, and actually is a more confusing layout IMHO. > I think it's worth breaking the backwards compatability a little in > order to bring the older boost headers in line with the directory > structure of the newer ones like thread. I also think enabling a > single include path for multiple libraries is useful. Boost.Threads followed existing directory guidelines (as do all libraries). There's no difference between it and "older" libraries (note that Boost.Threads is not that new). William E. Kempf wekempf(a)cox.net

2 1

is_convertible and array size question
by globeordi＠free.fr 08 Jan '03

08 Jan '03

Hello all, Just perhaps a silly question but who knows ? I have a small class that encapsulates a char[] type such as: template <const unsigned int n> struct C_Array { enum { size = n }; typedef char (&Array)[size]; typedef const char (&ConstArray)[size]; C_Array() { memset(&v[0], 0, 30); v[0] = '6'; v[1] = '7'; v[2] = '8'; v[3] = '9'; } char v[n]; int toto; char & operator[](const unsigned int i) { return v[i]; } const char & operator[](const unsigned int i) const { return v[i]; } operator Array () { return v; } operator ConstArray () const { return v; } }; Thanks to the operator Array, the objects of this class can be converted to references to arrays and the following pseudo-code works fine as expected: C_Array<30> obj; cout << boost::is_convertible<obj, char[]>::value << endl; But what I want to get is the actual size of the converted array. In the example the size is 30. It is not very important (workaround exists), just curious to see how it can be done without telling to compiler more information. Best regards Joseph Canedo

1 0

RE: [Boost-Users] Problem with BOOST_STATIC_ASSERT
by Randy Bowen 08 Jan '03

08 Jan '03

Some further follow-up on this. I've created a minimal program that exhibits the new problem behavior: /******************************************/ #include <boost/type_traits.hpp> #include <boost/static_assert.hpp> template < typename ValueType > struct A { BOOST_STATIC_ASSERT(boost::is_integral<ValueType>::value); }; extern "C" int main(int argc, char* argv[]) { return 0; } /******************************************/ The fact that this assertion fails is counter-intuitive to me, since there is no instantiation to compile. Note that this behavior seems limited to the use of "is_integral" as the type predicate. Constructing a similar program using "is_POD" behaves as expected: /******************************************/ #include <boost/type_traits.hpp> #include <boost/static_assert.hpp> template < typename ValueType > struct B { BOOST_STATIC_ASSERT(boost::is_POD<ValueType>::value); }; struct C {}; extern "C" int main(int argc, char* argv[]) { // B<int> Bint; /* OK to add */ // B<C> BC; /* causes error */ return 0; } /******************************************/ Here, no error occurs until there is an attempt to instantiate B<C>. Randy Bowen -----Original Message----- From: Randy Bowen Sent: Tuesday, January 07, 2003 1:40 PM To: Boost-Users(a)yahoogroups.com Subject: [Boost-Users] Problem with BOOST_STATIC_ASSERT I'm in the process of updating some applications from use of Boost 1.28.0 to 1.29.0, using MSVC7. I've noticed that a number of static assertions that used to compile now generate errors. The assertions all use is_integral to perform a type validation - e.g.: BOOST_STATIC_ASSERT(::boost::is_integral<value_type>::value); This produces an error of the form: c:\Perforce\Development\Web\cppinclude\domestic\Application Framework\State Machine Engine\StringSerializer.h(145) : error C2027: use of undefined type 'boost::STATIC_ASSERTION_FAILURE<x>' with [ x=false ] c:\Perforce\Development\Web\cppinclude\domestic\Application Framework\State Machine Engine\StringSerializer.h(143) : see reference to class template instantiation 'sme::StringSerializer<_StringType,>::StringWriter<<template parameter>>' being compiled c:\Perforce\Development\Web\cppinclude\domestic\Application Framework\State Machine Engine\StringSerializer.h(169) : see reference to class template instantiation 'sme::StringSerializer<_StringType,>' being compiled I believe that the data type is integral, and is complete at compile time. Any known reason why this would happen? Randy Bowen

2 1

Severe bugs in lambda/algorithm.cpp
by Markus Werle 08 Jan '03

08 Jan '03

Hi! I wonder whether this file ever was included by anyone else but me and which compiler did let this pass without complaint. Besides the #include <utilities> which never ever happens we have 2 errors in struct count struct count { template <class Args> struct sig { typename ::std::iterator_traits< // **************************************************** // ERROR: typedef is not allowed here typedef typename boost::remove_const< typename boost::tuples::element<1, Args>::type >::type >::difference_type type // ERROR: semicolon missing }; Same errors in count_if beginning at line 172. ----------------------------------------------------- In line 613: a duplicate parameter name c operator()(A a, B b, C c, D c) const Same on line 681 best regards, Markus

1 0

lambda/algorithm.hpp: #include <utilities>?
by Markus Werle 08 Jan '03

08 Jan '03

Hi! Line 23 in lambda/algorithm.hpp reads #include <utilities> // for std::pair IMHO (and that of intel's C++ and SGI's documentation) this line should be #include <utility> // for std::pair right? best regards, Markus

1 0

Slight Alteration to Boost Directory Structure
by Simon Bailey <conic＠bigpond.com> 07 Jan '03

07 Jan '03

I've been using boost for about a year now. I upgraded from 27 to 28 and appreciated the backwards compatability. But I'm wondering if a slight modification to the directory structure is in order? I have written a cross-platform class library. The library provides high-level business objects and comes bundled with boost 1.28.0 to provide support for threads, CRC, shared_ptr, and function (or whatever else the end user might want). Currently, users need to add two directories to their #include search path: include/boost include/sb_library Then, in their source code: #include <boost/shared_ptr.hpp> #include <boost/thread/thread.hpp> #include <sb_library/business_facet/business_object.h> Two problems (albeit very minor ones!) 1. Boost has two levels of includes (due to it's history of course). 2. Each library bundled with boost means another #include search path needs to be added. Proposed solution: 1. Create a single boost include directory instead of having the two nested ones that there are now. 2. Put all boost headers into a sub-directory of boost eg <boost/ptr/shared_ptr.hpp>, <boost/thread/thread.hpp>. (It's a small change for existing users to change their includes). 3. Place all src, documentation etc. into a directory within boost called "private" or something. Suggested directory structure: --include // This is the only directory needed to be specified in the #include search path ----other_library1 ----other_library2 ----boost -------ptr -------thread -------private ----------documentation ----------src ----------thread In other words, you will have <boost/ptr/...> and <boost/thread/...> but never <boost/private/...>. I think it's worth breaking the backwards compatability a little in order to bring the older boost headers in line with the directory structure of the newer ones like thread. I also think enabling a single include path for multiple libraries is useful. Cheers Simon

1 0

Problem with BOOST_STATIC_ASSERT
by Randy Bowen 07 Jan '03

07 Jan '03

I'm in the process of updating some applications from use of Boost 1.28.0 to 1.29.0, using MSVC7. I've noticed that a number of static assertions that used to compile now generate errors. The assertions all use is_integral to perform a type validation - e.g.: BOOST_STATIC_ASSERT(::boost::is_integral<value_type>::value); This produces an error of the form: c:\Perforce\Development\Web\cppinclude\domestic\Application Framework\State Machine Engine\StringSerializer.h(145) : error C2027: use of undefined type 'boost::STATIC_ASSERTION_FAILURE<x>' with [ x=false ] c:\Perforce\Development\Web\cppinclude\domestic\Application Framework\State Machine Engine\StringSerializer.h(143) : see reference to class template instantiation 'sme::StringSerializer<_StringType,>::StringWriter<<template parameter>>' being compiled c:\Perforce\Development\Web\cppinclude\domestic\Application Framework\State Machine Engine\StringSerializer.h(169) : see reference to class template instantiation 'sme::StringSerializer<_StringType,>' being compiled I believe that the data type is integral, and is complete at compile time. Any known reason why this would happen? Randy Bowen

1 0

uBLAS - Calculation of Inverse and/or Least Squares
by jonny5cents <jon_nichols95＠hotmail.com> 07 Jan '03

07 Jan '03

Hello All, Can anyone suggest the best way to use uBLAS to calculate the inverse of a matrix?? One of the reasons I want to do this is to solve a linear least squares problem, which I know can be done without explicitly solving for the inverse. Does anyone know if this is possible?? thanks, jon

3 2