- Boost - lists.boost.org

[ptr_container] clone_ptr library
by raffimoni 25 May '10

25 May '10

Hello everybody, I did not find any boost library that handle with a situation described below, so that I request for your comments. Is there any interest in such a library? ===================== MOTIVATION ===================== Every time we need a class that owns some pointer (so that class is responsible for freeing it), we always need to define destructor, copy constructor and assignment operator. This is example implementation: class SomeClass { Base* ptr_; public: SomeClass( Base* ptr ): ptr_(ptr) {} ~SomeClass() { delete ptr_; } SomeClass(const SomeClass& s); //must be implemented const SomeClass& operator =(const SomeClass& s); //must be implemented }; In order to avoid a need of defining destructor we can use auto_ptr: class SomeClass { std::auto_ptr<Base> ptr_; public: SomeClass( Base* ptr ): ptr_(ptr) {} SomeClass(const SomeClass& s); //must be implemented const SomeClass& operator =(const SomeClass& s); //must be implemented }; But we need still implement copy constructor and assignment operator. Of course we can declare SomeClass noncopyable by deriving from boost::noncopyable to avoid this, but suppose this is not special case class and copying should be allowed. The best elegant solution to this problem is using a clone_ptr: class SomeClass { clone_ptr< Base > ptr_; public: SomeClass( Base* ptr ): ptr_(ptr) {} }; We can think about clone_ptr in two different ways: -as auto_ptr with different copy semantics (this is exactly what clone_ptr is) -as one-element ptr_container clone_ptr frees programmer from managing pointer the same way ptr_container for container of pointers does. So during copying the above class with clone_ptr behaves exactly like this: class SomeClass { //suppose this contains only one pointer boost::ptr_vector< Base > ptr_; public: SomeClass( Base* ptr ) {ptr_.push_back( ptr );} }; Notice that we do not need longer to define destructor, copy constructor and assignment operator. The default behaviour of these methods makes a deep copy of pointer and we should not be bothered by this anymore. clone_ptr uses the same concept of cloning that ptr_containers do, so new_clone function should be defined for Base class, if it is not and clone_ptr< Base > points to some type that derives from Base, an appropriate assert will fail the same way it does with ptr_container. Defining new_clone for Base makes it cloneable for ptr_pointer and clone_ptr simultaneously. ===================== SAMPLE IMPLEMENTATION ===================== class clone_ptr { private: T* ptr_; public: explicit clone_ptr(T* ptr) throw(): ptr_( ptr ) {} ~clone_ptr() { delete ptr_; }; clone_ptr(const clone_ptr<T>& ptr) { ptr_ = new_clone( *ptr.ptr_ ); //default is defined in ptr_container library assert( typeid( *ptr_ ) == typeid( *ptr.ptr_ ) ); } public: void reset( T* ptr = 0) throw() { if (ptr != ptr_) { delete ptr_; ptr_ = ptr; } } T* operator -> () const throw() { return ptr_; } }; Of course this implementation is not complete. Notice the way copy constructor is implemented. ===================== REMARKS ===================== clone_ptr should definitely not be used with stl containers (e.g. stl::vector< clone_ptr< Base > >) because of performance, ptr_containers are the best way to achieve such functionality. I'm looking forward to your feedback. Best Regards, Rafal Moniuszko

5 5

[test] [Patch] Unconditional use of non-standard functions.
by John Maddock 25 May '10

25 May '10

Boost.Test has just started using ::fedisableexcept and ::feenableexcept unconditionally - even though these are GLIBC specific extensions - which presumably is why Boost.Test is currently failing so many tests :-( I've just committed a patch to restrict the use of these functions to GLIBC platforms that have the GNU extensions enabled: see https://svn.boost.org/trac/boost/changeset/62194/trunk/boost/test/impl/exec…. Please feel free to change the logic for any other platforms that may have these functions.... but for now, hopefully we can get back to testing Boost again! Cheers, John.

1 0

[fusion] flat fusion sequence of all the node of a type
by vicente.botet 24 May '10

24 May '10

Hi, I want to create a fusion sequence that give access to all the nodes of a struct/class/array that inherit from a given class TO. If we remove this constraint, the fusion sequence can be seen as flattening the tree of nodes in a given order. Once I will have this I can use filter_if<is_base_of<TO,_> > Given class TO; struct W { TO t; }; struct X : TO { // direct TO* to_ptr; char c; }; struct Y : TO { // direct with embeddeds X x; int i; }; struct Z { // with embeddeds Y y[2]; TO t; W w; }; Z z; the flat view corrsponds to z.w.t, z.w, z.t, z.y[0].x.to_ptr, z.y[0].x.c, z.y[0].x, z.y[0], z.y[1].x.to_ptr, z.y[1].x.c, z.y[1].x, z.y[1], z.y, z The filtered view corrsponds to Y z.y[0] X z.y[0].x Y z.y[1] X z.y[1].x TO z.t TO z.w.t If I have fusion sequence struct adaptors for W, X, Y, Z. Is there a simple and run-time efficient way to get this fusion sequence adaptor or I need to do something similar to the struct adaptor? Have others had a similar need already? Best, Vicente P.S. For my personal usage I need only a forward sequence.

1 0

[MSM] compiling MSM with GCC 4.4.1 and -fno-rtti
by Richard Szabo 24 May '10

24 May '10

Hi All I try to compile boost msm in an embedded target with gcc 4.4.1 for performance reasons I want to disable RTTI. I'm having problems because backend is using boost::any which is uses RTTI. Is it possible to remove this dependency ?. or introduce some compile switch like in shared pointers .... BOOST_NO_STD_TYPEINFO Thanks in advance. Richie

1 0

Maintenance Patches and cleaning up trac
by Christopher Jefferson 24 May '10

24 May '10

Having recent been browsing through boost trac I have noticed that it seems to be getting a little clogged with some very simple patches. Examples of simple patches: https://svn.boost.org/trac/boost/ticket/3402 - simple documentation patch, 6 months old, duplicated. https://svn.boost.org/trac/boost/ticket/3669 - add stdint.h to thread, fixes comeau, 6 months old. https://svn.boost.org/trac/boost/ticket/4060 - boost thread support for BOOST_NO_IOSTREAM I notice these 3 are in 'thread', and several others are in 'filesystem'. I do not wish to maintain these libraries myself, and it appears no-one else is coming forward to keep these libraries ticking over on a day-to-day basis. I would however personally be happy (possibly with help) to look at applying these patches, and applying the to trunk, and then release in time, assuming they pass regression testing. I would suggest a policy something like the following: 1) If a patch is considered to be "trivially obvious", mark the patch in trac, by leaving a comment, as a 'maintenance patch'. 2) If no-one disagrees within some period of time (2 weeks? 3?), then the patch is applied to trunk. If it passes regression testing it is then moved to release. Any thoughts on the topic? Chris

8 14

Re: [boost] [Boost-docs] Maintaining boost.org
by David Abrahams 24 May '10

24 May '10

At Tue, 25 May 2010 00:32:02 +0800, Joel de Guzman wrote: > > On 5/24/10 10:41 PM, Rene Rivera wrote: > > > > One idea I had, before Boostcon when the whole where should Boost go > > discussion came up, was to move to using something like the Drupal model > > for sub-projects <http://drupal.org/project/Modules>. Particularly how > > the projects each have their own web area. And also how the groups > > section area of Drupal works <http://groups.drupal.org/>. > > > > Note, I'm not suggesting Drupal (even if I do currently use it). Just > > some of the structure they employ for projects since it does mirror the > > conglomerate project structure Boost has. > > Last November, we switched the Spirit site to Wordpress. > > http://boost-spirit.com/home/ > > After having tried Drupal, Joomla, and a couple of other > CMS, we settled on Wordpress because of its simplicity, > low entry barrier and low maintenance. It's just perfect for > people on the go, like us. Agreed; WP rocks. Consider WPMU (or WP3.0) and/or BuddyPress also. I am using those at http://techarcana.net, although I haven't had time to make everything really nice yet. -- Dave Abrahams Meet me at BoostCon: http://www.boostcon.com BoostPro Computing http://www.boostpro.com

1 0

How to build Boost with C++0x compilers
by vicente.botet 24 May '10

24 May '10

Hi, I don't know how to build Boost with C++0x compilers. Which option must be used with bjam? Should the user.config file be modified? Can someone help me? Best, Vicente

5 6

Testing macros..
by Rene Rivera 24 May '10

24 May '10

Once upon a time I did a bunch of work towards a Boost Predef library. The goal was to have a consistent set of version and feature definitions as could be determined from the predefined macros of toolsets <https://svn.boost.org/svn/boost/sandbox/predef/>. But the work stalled for one basic reason.. How do I write tests for such a library? The only other library that approaches the domain of this one is Boost Config and, AFAICT, it doesn't test the macros themselves. As I'm looking to finish of this lib for eventual submission I'm looking for ideas on this. -- -- Grafik - Don't Assume Anything -- Redshift Software, Inc. - http://redshift-software.com -- rrivera/acm.org (msn) - grafik/redshift-software.com -- 102708583/icq - grafikrobot/aim,yahoo,skype,efnet,gmail

3 2

Is there a plan to provide a C++0x std library repository in Boost?
by vicente.botet 24 May '10

24 May '10

Hi, Boost provides a lot of C++0x accepted libraries that could be in the std namespace. As it was done for std::tr1, is there a plan to provide these libraries on namespace std? Best, Vicente

4 5

[challenge!] find a common domain
by Eric Niebler 23 May '10

23 May '10

I recently solved a devilish C++ coding problem related to the new sub-domain feature of Proto. I don't care for my solution, so I thought I'd throw it out there and see if some smart person could do better. I'm sending along the skeleton of the solution and some tests that it needs to satisfy. The rest is up to you. The problem goes like this. Every Proto expression has a "domain", which is a type defined something like this: struct my_domain : domain<> {}; A domain can be a sub-domain of another domain, defined like this: struct my_sub_domain : domain<my_domain> {}; When combining sub-expressions into larger expressions, the new expression has some domain that depends on the domains of its children. This domain is calculated according to some rules. - Domains that share no common super-domain are incompatible with each other. The result should be not_a_domain. - A sub-domain is "stronger" than its super-domain. - For a set of domains, the common domain is the strongest domain that is a super-domain of every domain in the set. - default_domain is special; it is compatible with (and dominated by) all other domains. - A sub-domain of default_domain is stronger than default_domain and still compatible with and dominated by all other domains. Your job: implement the deduce_domain3 template that finds the common domain of 3 domains. You're allowed to use decltype, but you get bonus points for a solution that doesn't. Bonus also for instantiating fewest templates. The challenge is for the ternary case, but your solution should scale to N domains. If your solution is better than mine, I'll use it in Proto and credit you! Good luck. P.S. I haven't checked my solution in yet, so you can't look in trunk and cheat. ;-) -- Eric Niebler BoostPro Computing http://www.boostpro.com

8 24