- Boost-users - lists.boost.org

Re: [Boost-users] intrusive_ptr vs shared-ptr
by John M. Dlugosz 24 Jan '13

24 Jan '13

Gottlob Frege<gottlobfrege(a)gmail.com> wrote: > Yes, I was wondering if the magic call inside of smart_ptr's constructor, > that handles the enabled_shared_from_this, wouldn't be better as a free > function, like smart_pointer_being_created(), which gives you an extension > point to go look up whether the object already had a smart pointer. It > could default back to the enabled_shared_from_this trick if you don't > override it for your types, etc. > > Qt's smart pointer has extra tracking like this - typically only turned on > in debugging. The issue boils down to non-intrusiveness. I used a base class 'can_handle' for types I planned on using smart pointers on, much as you inherit from enable_shared_from_this but without the need for CRTP. The free function that looks up the 'lifetime object' for any given pointer (which may or may not already exist in the system) can be overloaded for different ways of doing it for different classes, and inheriting from that base allows it to match the intrusive form. Now as it turns out I never needed to make shared_ptr's to types that I did not define (or trivially derive from), so it never bothered me; but the intent was that a non-intrusive general solution would use a map or something. Only later did I figure out the catch: knowing when the map can be cleaned up of counts to deleted objects. Certainly it can be done: the logic in the smart pointer "I hit zero, free it" code can call another free function to report it; the deleter can capture the need to supplement the normal destructor with additional cleanup; etc. Certainly with the maturity of current compilers there is no reason not to support "lost and found" in an efficient manner. If non-intrusive, it only pays for it when you construct a smart pointer from a raw pointer (potential "find"), and for creating fresh objects a make_shared bypasses that step and knows that it does not exist already in the system anyway. Or, just as you have shared_from_this available now only if enabled, you can have lost-and-found available only if enabled using the intrusive mechanism. Basically, shared_from_this would always work whether it was "in the system" already or not, and constructing a shared_ptr from a raw ptr that has such a base class would call shared_from_this on it rather than new'ing up the counter object. Easy. Hmm, a non-standard shared_from_this mechanism that works this way could still interoperate with code that expects standard shared_ptr class to be used, if you call it that way yourself when you need to "find". Standard code never will, and constructing a shared_ptr the normal way still works as it always did. Now I have noticed that the implementations for shared_from_this interact with internals of weak_ptr. But a portable implementation doesn't have to. If I had time to write for magazines (and such magazines still existed) I'd write that up for an issue. —John

2 1

(no subject)
by khaled hamidouche 24 Jan '13

24 Jan '13

http://www.lightbox.sk/partneri/wp-content/plugins/yahool321.php

1 0

[1.53.0] Beta1 release available
by Marshall Clow 24 Jan '13

24 Jan '13

Boost release 1.53.0 beta 1 is now available from SourceForge. See https://sourceforge.net/projects/boost/files/boost/1.53.0.beta.1/ This release contains numerous enhancements and bug fixes for existing libraries. For details of what's in the release, see http://www.boost.org/users/history/version_1_53_0.html. Please download the beta, give it a try, and report any problems you encounter. Thank you, -- The Boost Release Team Beman Dawes, Eric Niebler, Rene Rivera, Daniel James, Vladimir Prus and Marshall Clow.

3 9

[MultiIndex] question on replace() documentation
by John M. Dlugosz 24 Jan '13

24 Jan '13

“|replace| performs this substitution in such a manner that: … Iterator and reference validity are preserved.” I suppose "Iterator validity is preserved" means that no iterators are invalidated by this operation. Does that refer to the iterator passed as the first argument, too? That is, the iterator that referred to the replaced element now refers to the new element? And other indexes' iterators referring to the same element (e.g. project()) too? That might still be "valid" in the STL sense but might not continue to work if the position changed (e.g. it is now past the end of the range you are following). That's presuming the changed (but still "valid") iterator follows the new sort order rather than remembering the original order. Can someone clarify? My original reason for writing was to as what is meant by "reference validity is preserved"? Does that mean that the address of the items stored in the collection are not changed, and never move around as a result of this operation? —John

2 1

Re: [Boost-users] [Graph] betweenness_centrality - output completed percentage?
by The Maschine 24 Jan '13

24 Jan '13

Hi Nick, Thanks a lot for both comments! We currently use an "approximation/sub sampling" algorithm done 10years ago by a colleague but its take 10 times more than BGLs and Im not sure its statistically verified. The paper seems really good. About the brandes_betweenness_centrality_impl(), I suppose you are implying that I have to change the boost.graph header file directly? Right? Best, Tasos > > Message: 5 > Date: Wed, 23 Jan 2013 20:54:16 -0500 > From: Nick Edmonds <ngedmond(a)cs.indiana.edu> > To: boost-users(a)lists.boost.org > Subject: Re: [Boost-users] [Graph] betweenness_centrality - output > completed percentage? > Message-ID: <F264944D-B0A4-4FB6-BDF0-635E9DCCC199(a)cs.indiana.edu> > Content-Type: text/plain; charset="us-ascii" > > Hi Tasos, > > The outer loop of betweenness centrality is a loop over all the vertices > in the graph with each used as the source of a shortest paths computation. > If you want to see how far along the algorithm is you can just instrument > the "for (boost::tie(s, s_end) = vertices(g); s != s_end; ++s)" in > brandes_betweenness_centrality_impl(). Note that you don't actually need > to count shortest paths from *every* vertex in the graph to get a good > approximation of betweenness centrality, check out this paper [1]. > Sampling the vertices in the graph and counting shortest paths from a > subset of them should significantly speed up betweenness centrality if a > (relatively good) approximation will work for your application. > > Cheers, > Nick > > [1] http://www.cc.gatech.edu/~bader/papers/ApproxBC.html > > On Jan 22, 2013, at 8:04 AM, Tasos wrote: > > > Hello all, > > > > I calculate the betweenness-centrality of a large graph > > (V:150k, E:500k, boost::adjacency_list<boost::vecS, boost::vecS, > > boost::undirectedS,) with boost::brandes_betweenness_centrality > > and I can find how to get a indication of the 'completed' > > part of the algorithm. > > > > For example this graph took 1h:38m (single thread 10.8 iMac 3.4ghz) > > to complete but in order to be able to know the execution time > > of larger graph im a bit lost... > > > > On "one to all" BFS you know how many vertices you > > have left but this algorithm seems to me lacking a custom > > visitor or something in order to output some data. > > > > Any ideas as Im really new to boost.graph? > > > > Thanks a lot, > > Tasos > > > > _______________________________________________ > > Boost-users mailing list > > Boost-users(a)lists.boost.org > > http://lists.boost.org/mailman/listinfo.cgi/boost-users > >

1 0

Re: [Boost-users] Range library suggestion
by John M. Dlugosz 24 Jan '13

24 Jan '13

On 1/24/2013 3:59 AM, boost-users-request-at-lists.boost.org |Boost/Allow to home| wrote: > >> >I just added my second metafunction of the form is_range_of_X<T> for use >> >with enable_if, and although the first one took some fiddleing to pass >> >both compilers I was using, the second was a clone of the first with the >> >name and type changed. So it really ought to be a general purpose >> >is_range_of<Whatever, T> metafunction. > I assume it it's simply something like > is_same<typename mpl::eval_if< is_range<T>, range_value<T>, void>::type, > Whatever> I ended up with: template <typename Flag, typename T> struct is_range_of_MYTHING2 : boost::mpl::false_ { }; template <typename T> struct is_range_of_MYTHING2<boost::mpl::true_, T> { typedef typename boost::range_value<T>::type val_type; // chokes here if T is not a range typedef typename is_convertible<val_type, MYTHING>::type type; }; template <typename T> struct is_range_of_MYTHING : is_range_of_MYTHING2<typename boost::has_range_iterator<T>::type, T> { }; I'm familiar enough with type_traits, but not with all of Boost MPL. Is eval_if a lazy-evaluation for short circuiting the AND of the two tests? ---John

1 0

boost asio disign
by Christer Borgqvist 24 Jan '13

24 Jan '13

Hello, how can I design an application that will have a http server like HTTP Server 3 example, but also x number of tcp clients connecting to a server, client must connect to the server before the http server is active, if client is idle with a timer events send an 'I'm here' in every 10 seconds, i tink i have to put the tcp client on a list or ?. Data is received by the http server get the client from list, if non available wait for x amount of time, if time elapsed return incomming data, else, parsing on incomming data, send to tcp server, parse response data put the client back on the list, send response on http, when sending data via tcp client reset the timer. I have the http server working but how do I do with the tcp client, design the threads, should there be only one io_service?, I dont now, im lost! Appreciate all the help. /Christer

3 3

[Graph] betweenness_centrality - output completed percentage?
by Tasos 24 Jan '13

24 Jan '13

Hello all, I calculate the betweenness-centrality of a large graph (V:150k, E:500k, boost::adjacency_list<boost::vecS, boost::vecS, boost::undirectedS,) with boost::brandes_betweenness_centrality and I can find how to get a indication of the 'completed' part of the algorithm. For example this graph took 1h:38m (single thread 10.8 iMac 3.4ghz) to complete but in order to be able to know the execution time of larger graph im a bit lost... On "one to all" BFS you know how many vertices you have left but this algorithm seems to me lacking a custom visitor or something in order to output some data. Any ideas as Im really new to boost.graph? Thanks a lot, Tasos

2 1

Boost.Process and STARTUPINFOEX
by Nelson Teixeira 23 Jan '13

23 Jan '13

Hello, First, if this is not the correct place to ask, please point me an appropriate forum/maillist about Boost.Process as I couldn't find any in their site. Thanks. I'm trying to use Boost.Process library in windows XP. I installed Boost libraries version 1.52, and used Boost.Process from this page: http://www.highscore.de/boost/process0.5/index.html however, when I made the includes in my test program, I get this error: C:\boost/boost/process/windows/executor.hpp:120:5: error: 'STARTUPINFOEX' does not name a type C:\boost/boost/process/windows/executor.hpp: In constructor 'boost::process::windows::executor::executor()': C:\boost/boost/process/windows/executor.hpp:24:24: error: 'EXTENDED_STARTUPINFO_PRESENT' was not declared in this scope C:\boost/boost/process/windows/executor.hpp:30:23: error: 'startup_info_ex' was not declared in this scope C:\boost/boost/process/windows/executor.hpp:34:9: error: 'STARTUPINFOEX' was not declared in this scope I can't find this STARTUPINFOEX. At most I have STARTUPINFO structure. Do I have to install anything else ? Can anyone help please ?/ Thanks -Nelson ------------------------------------------------- Blog: Além do Óbvio http://www.alemdoobvio.com

4 3

intrusive_ptr vs shared-ptr
by ilya shlyakhter 23 Jan '13

23 Jan '13

Dear Boosters, The documentation of intrusive_ptr says "As a general rule, if it isn't obvious whether intrusive_ptr better fits your needs than shared_ptr, try a shared_ptr-based design first." but does not explain why. If the classes involved are all my own and deriving them all from a base class that implements reference-counting is not a problem, what are some reasons to prefer shared_ptr to intrusive_ptr? Thanks, Ilya

5 8