- Boost - lists.boost.org

[asio] select_reactor questions
by Andrew Schweitzer 03 Jan '06

03 Jan '06

Hi Chris, just trying to understand how select_reactor works. Can you confirm/correct? select_reactor's job is to wait for socket operations and timeouts. When any of these occur, select_reactor calls demuxer::post so that demuxer_service to can actually issue the callbacks for these events. select_reactor does its job by calling select. Select is passed a list of all active sockets plus a special, "dummy" socket. Select_reactor, via a socket_select_interrupter object, is both client and server on this special dummy socket. When the user wants to add a new event to react to, select must be interrupted so it can start waiting on the new event. This is accomplished by issuing a write on the dummy socket. If the user has scheduled any timers, select is passed a timeout of the smallest existing timer. On Windows select_reactor constructor starts its own thread which executes the select wait in a loop. Assuming all of that is correct, can you explain what the code in socket_ops::select() does that checks if all sockets are null and if so calls ::Sleep()? I sort of expect I've missed something, but it looks like dead code. with the prescence of the interrupter and its dummy socket, the socket list will never be null. Also, if they were null, I'm not sure Sleep would work very well. For one thing, it's not interruptable, so if the only thing is a large timer, any new additions of sockets or timers would have to wait for the large timer to complete before they could be added. For another, it looks like if there were no timers running, it would essentially be polling 1/ms for sockets + timers. Thanks, Andrew

2 2

[Boost-bugs] [ boost-Bugs-1395924 ] boost.build needs fixes for HP/UX
by SourceForge.net 03 Jan '06

03 Jan '06

Bugs item #1395924, was opened at 2006-01-03 11:27 Message generated for change (Tracker Item Submitted) made by Item Submitter You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=107586&aid=1395924&group_… Please note that this message will contain a full copy of the comment thread, including the initial issue submission, for this request, not just the latest update. Category: None Group: None Status: Open Resolution: None Priority: 5 Submitted By: Oliver Kowalke (oliver_kowalke) Assigned to: Nobody/Anonymous (nobody) Summary: boost.build needs fixes for HP/UX Initial Comment: Hello, I'm using boost on HP/UX (11.11 PA-RISC) with gcc (3.4.4). The default build for boost needs to be fixed for this platform! 1.) -fPIC: Shared libraries must be build as position-independent code (HP/UX linker requires this) on HP/UX! 2.) On HP/UX the extension *.sl is common for shared libraries -> boosts default is *.so (my UNIX-Admin was m irritated be *.so). Do not use "-sSUFDLL=.sl"!!! This generates libraries which will fail during load (for instance signals library). I was able to reproduce this behaviour. Oliver ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=107586&aid=1395924&group_… ------------------------------------------------------- This SF.net email is sponsored by: Splunk Inc. Do you grep through log files for problems? Stop! Download the new AJAX search engine that makes searching your log files as easy as surfing the web. DOWNLOAD SPLUNK! http://ads.osdn.com/?ad_id=7637&alloc_id=16865&op=click _______________________________________________ Boost-bugs mailing list Boost-bugs(a)lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/boost-bugs

1 0

[Boost-bugs] [ boost-Bugs-1395874 ] Proogram_options: Fails to parse options with a common root
by SourceForge.net 03 Jan '06

03 Jan '06

Bugs item #1395874, was opened at 2006-01-03 09:56 Message generated for change (Tracker Item Submitted) made by Item Submitter You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=107586&aid=1395874&group_… Please note that this message will contain a full copy of the comment thread, including the initial issue submission, for this request, not just the latest update. Category: None Group: None Status: Open Resolution: None Priority: 5 Submitted By: Roger Leigh (rleigh) Assigned to: Nobody/Anonymous (nobody) Summary: Proogram_options: Fails to parse options with a common root Initial Comment: Given the following testcase: #include <iostream> #include <boost/program_options.hpp> namespace opt = boost::program_options; int main(int argc, char *argv[]) { try { bool all, all_chroots, all_sessions; opt::options_description chroot("Chroot selection"); chroot.add_options() ("all,a", "Select all chroots and active sessions") ("all-chroots", "Select all chroots") ("all-sessions", "Select all active sessions"); opt::variables_map vm; opt::store(opt::parse_command_line(argc, argv, chroot), vm); opt::notify(vm); if (vm.count("all")) all = true; if (vm.count("all-chroots")) all_chroots = true; if (vm.count("all-sessions")) all_sessions = true; return 0; } catch (std::exception const& e) { std::cerr << e.what() << std::endl; return 1; } } $ g++ -o opt opt.cc -lboost_program_options $ ./opt --help unknown option help $ ./opt --all-chroots $ ./opt --all-sessions $ ./opt --all ambiguous option all --all is not abiguous, since it's an exact match. It's not ambigous for GNU getopt or GLib GOption, so this is IMHO a bug. Regards, Roger ii libboost-dev 1.33.0-6 Boost C++ Libraries development fi ii libboost-program-options1.33. 1.33.0-6 program options library for C++ ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=107586&aid=1395874&group_… ------------------------------------------------------- This SF.net email is sponsored by: Splunk Inc. Do you grep through log files for problems? Stop! Download the new AJAX search engine that makes searching your log files as easy as surfing the web. DOWNLOAD SPLUNK! http://ads.osdn.com/?ad_id=7637&alloc_id=16865&op=click _______________________________________________ Boost-bugs mailing list Boost-bugs(a)lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/boost-bugs

1 0

[Boost-bugs] [ boost-Bugs-1395857 ] wave redefines BSIZE
by SourceForge.net 03 Jan '06

03 Jan '06

Bugs item #1395857, was opened at 2006-01-03 09:18 Message generated for change (Tracker Item Submitted) made by Item Submitter You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=107586&aid=1395857&group_… Please note that this message will contain a full copy of the comment thread, including the initial issue submission, for this request, not just the latest update. Category: None Group: None Status: Open Resolution: None Priority: 5 Submitted By: Oliver Kowalke (oliver_kowalke) Assigned to: Nobody/Anonymous (nobody) Summary: wave redefines BSIZE Initial Comment: Hello, on HP/UX 11.11 wave library redefines BSIZE in libs/wave/src/cpplexer/re2clex/cpp.re:46. It is already defined by the os in /usr/include/sys/param.h:215. Oliver ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=107586&aid=1395857&group_… ------------------------------------------------------- This SF.net email is sponsored by: Splunk Inc. Do you grep through log files for problems? Stop! Download the new AJAX search engine that makes searching your log files as easy as surfing the web. DOWNLOAD SPLUNK! http://ads.osdn.com/?ad_id=7637&alloc_id=16865&op=click _______________________________________________ Boost-bugs mailing list Boost-bugs(a)lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/boost-bugs

1 0

Bug in Boost.Format_options
by Roger Leigh 02 Jan '06

02 Jan '06

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Given the following testcase: #include <iostream> #include <boost/program_options.hpp> namespace opt = boost::program_options; int main(int argc, char *argv[]) { try { bool all, all_chroots, all_sessions; opt::options_description chroot("Chroot selection"); chroot.add_options() ("all,a", "Select all chroots and active sessions") ("all-chroots", "Select all chroots") ("all-sessions", "Select all active sessions"); opt::variables_map vm; opt::store(opt::parse_command_line(argc, argv, chroot), vm); opt::notify(vm); if (vm.count("all")) all = true; if (vm.count("all-chroots")) all_chroots = true; if (vm.count("all-sessions")) all_sessions = true; return 0; } catch (std::exception const& e) { std::cerr << e.what() << std::endl; return 1; } } $ g++ -o opt opt.cc -lboost_program_options $ ./opt --help unknown option help $ ./opt --all-chroots $ ./opt --all-sessions $ ./opt --all ambiguous option all - --all is not abiguous, since it's an exact match. It's not abigous for GNU getopt or GLib GOption, so this is IMHO a bug. Regards, Roger Versions of packages libboost-program-options-dev depends on: ii libboost-dev 1.33.0-6 Boost C++ Libraries development fi ii libboost-program-options1.33. 1.33.0-6 program options library for C++ - -- Roger Leigh Printing on GNU/Linux? http://gimp-print.sourceforge.net/ Debian GNU/Linux http://www.debian.org/ GPG Public Key: 0x25BFB848. Please sign and encrypt your mail. -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.2 (GNU/Linux) Comment: Processed by Mailcrypt 3.5.8+ <http://mailcrypt.sourceforge.net/> iD8DBQFDubyOVcFcaSW/uEgRAv/1AJ0bMx0Me4owHITYjdQqjJz+N+yNOACggbU7 rTykg0EABvU1XIiX+LZ5KjY= =QqE/ -----END PGP SIGNATURE-----

1 0

bug in boost.build?
by Oliver.Kowalke＠infineon.com 02 Jan '06

02 Jan '06

Hello, I try to use boost-1.33.1 on HP/UX ('bjam -prefix=/opt/boost "-sTOOLS=gcc" install') with gcc. For me it seams that the libraries are NOT tested on HP/UX! Uname -a: HP-UX drssx248 B.11.11 U 9000/800 888950691 unlimited-user license Gcc -v: Reading specs from /opt/hp-gcc/3.4.4/ilp32/lib/gcc/hppa1.1-hp-hpux11.11/3.4.4/specs Configured with: /proj/oscp/nightly/gcc/gcc_3_4_4_release/gcc/configure --enable-languages=c,c++ --enable-threads=posix --with-ld=/usr/ccs/bin/ld --without-gnu-ld --build=hppa1.1-hp-hpux11.11 --host=hppa1.1-hp-hpux11.11 --target=hppa1.1-hp-hpux11.11 --prefix=/opt/hp-gcc/3.4.4/ilp32 --with-as=/proj/oscp/nightly/gcc/gcc_3_4_4_release/HP-UX/hppa//opt/hp-gc c/3.4.4/ilp32/bin/as Thread model: posix gcc version 3.4.4 1.) In order to be platform-comnpliant the libraries should have the extension. 2.) I get linker errors like (please see my email [boost] linker problem ob HP/UX - wrong extension) /usr/ccs/bin/ld: Invalid loader fixup in text space needed in output file for symbol "__gxx_personality_sj0" in input file "/opt/boost/lib/libboost_regex-gcc." 3.) I get runtime segmentation faults if the system tries to load boost libraries (please see my email [boost] bug in signals library? - segmentation fault(onlylinked) -> exit(11) called) Those problems should be fixed if I try boost to compile with: bjam --prefix=/opt/boost "-sTOOLS=gcc" "-sSUFDLL=.sl" "-sBUILD=debug release <runtime-link>dynamik <threading>multi <cxxflags>-D_REENTRANT <cxxflags>-fpic <release><optimization>speed" install But I get only libboost__<xxx>-gcc-mt-1_33_1.a and libboost_<xxx>-gcc-mt-d-1_33_1.a libraries. No libboost__<xxx>-gcc-mt-1_33_1.sl libboost__<xxx>-gcc-mt-d-1_33_1.sl? Oliver

2 1

new smart pointer suggestion
by Preston A. Elder 02 Jan '06

02 Jan '06

Hey, What does everyone think of a new smart pointer called deep_ptr. This pointer would hold a unique pointer (only one copy at a time), but provide convenient syntax to have a deep copy of the underlying data on an operator= call. A sample implementation would be something like: template<typename T> class deep_ptr { typedef deep_ptr<T> this_type; public: typedef T element_type; typedef T value_type; typedef T * pointer; typedef T & reference; deep_ptr() : px(NULL) {} deep_ptr(T * ptr) : px(ptr) {} deep_ptr(deep_ptr const & in) : px(new T(*in.px)) {} deep_ptr &operator=(deep_ptr const & in) { delete px; px = new T(*in.px); } ~deep_ptr() { delete px; } void reset() { this_type().swap(*this); } template<typename U> void reset(U * p) { BOOST_ASSERT(p == 0 || p != px); this_type(p).swap(*this); } reference operator* () const { BOOST_ASSERT(px != 0); return *px; } T * operator->() const { BOOST_ASSERT(px != 0); return px; } T * get() const { return px; } operator bool () const { return px != 0; } bool operator! () const { return px == 0; } void swap(deep_ptr<T> & other) { std::swap(px, other.px); } private: T *px; }; template<typename T, typename U> inline bool operator==(deep_ptr<T> const & a, deep_ptr<U> const & b) { return a.get() == b.get(); } template<typename T, typename U> inline bool operator!=(deep_ptr<T> const & a, deep_ptr<U> const & b) { return a.get() != b.get(); } template<typename T, typename U> inline bool operator<(deep_ptr<T> const & a, deep_ptr<U> const & b) { return a.get() < b.get(); } template<class T> inline void swap(deep_ptr<T> & a, deep_ptr<T> & b) { a.swap(b); } This way, as I said, each copy would have its own version of the data, but in a class that uses it, you would not then have to write your own operator= for that class to handle the deep copy, you just make it use a deep_ptr instead of a regular pointer. I'll admit, its simply a case of adding syntactical sugar, but I could see how this would become extremely useful :) -- PreZ :) Death is life's way of telling you you've been fired. -- R. Geis

2 1

[range] boost::end("hello") is sub-optimal
by Eric Niebler 02 Jan '06

02 Jan '06

There is a bug in Boost.Range that is causing boost::end() to execute a slower code path when called with a string literal. The type of "hello" is char const [6], and end() should be O(1). Indeed, Boost.Range has code to handle just this case, but it never gets called. The problem is the way overload resolution happens between these two functions template< typename T, std::size_t sz > const T* boost_range_end( const T (&array)[sz] ) const char* boost_range_end( const char* s ) The first is O(1) and the second is O(N) becuase it calls strlen. But the second will always be chosen because non-templates are preferred over templates. -- Eric Niebler Boost Consulting www.boost-consulting.com

7 10

[ptr_map] Boost pointer containers, where is the working map type?
by axter 02 Jan '06

02 Jan '06

http://www.boost.org/libs/ptr_container/doc/reference.html I'm trying to use the boost::ptr_map class, and either I'm misunderstanding it's usage or syntax. I originally assumed that a ptr_map class is a map container of pointers and/or container of abstract objects via pointers. However, the example code in the following link, does not seem to be using pointers at all. http://www.boost.org/libs/ptr_container/doc/reference.html Moreover, if I attempt to use an abstract pointer in either the first or second type, I get compile errors. I also tried creating a ptr_map to a base class with a normal virtual function, but when I ran my test, I can see that it slices the derived class. So exactly what is the ptr_map class for, and does anyone have example code that will work with an abstract pointer? I'm trying to compare using the boost pointer containers to using a clone smart pointer. Which boost container would be able to give me functionallity like the following: map<cow_ptr<Shape>, int> mShapeToInt; map<int, cow_ptr<Shape> > mIntToShape; Where Shape is an abstract base class, and cow_ptr is the following: http://code.axter.com/cow_ptr.h

2 1

Re: [boost] ptr_container: ptr_map_adapter interface
by axter 02 Jan '06

02 Jan '06

"Slawomir Lisznianski" <public(a)paramay.com> wrote in message news:<donook$li2$1(a)sea.gmane.org>... > Hello, > > I've been using ptr_map lately and have a few questions regarding its > interface. > > Below is a signature of `insert' function as declared in > ptr_map_adapter: > > std::pair<iterator,bool> insert( key_type& k, value_type x ); > > Why is `k' a non-const reference? > > To imply ownership transfer, couldn't `x' be of > `std::auto_ptr<value_type>' type instead? > > What happend to std::pair as an argument of insert? Was symmetry with > std::map dropped for a reason here? > >From my test, it doesn't seem as though ptr_map has consistent and reliable implementation. I recommend that boost replace the pointer containers with the following smart pointers: http://code.axter.com/copy_ptr.h http://code.axter.com/cow_ptr.h Example usage: std::map<int, cow_ptr<foo> > MyPtrMap; The above type gives you more functionality then the boost::ptr_map type, and of course, it's interface matches the std::map interface. Where as the boost::ptr_map type has the following problems: 1. Doesn't allow you to use an abstract pointer 2. Requires a non-constant argument for the insert function 3. Requires value semantics for the operator[] function 4. When assigning value via operator[] from one container to another, you get object splicing for derived types 5. It doesn't seem to compile on VC++ 6.0, and when it compiles on VC++ 7.1 you get compiler warnings If you use cow_ptr instead, you don't have the above problems. There are similar problems in the boost::ptr_set class, which also can be replaced with cow_ptr or copy_ptr std::set<cow_ptr<foo> > MyPtrSet; IMHO, the entire boost pointer container set of classes can be replace with either cow_ptr and/or copy_ptr, and the result would be a more reliable, and consistent implementation.

2 1