June 2006 - Boost-users - lists.preview.boost.org

[posix_time][date_time] Another I/O error?
by Bogdan M. 03 Jun '06

03 Jun '06

Hi all, I upgraded to boost 1.33.1 and I started experiencing problems with the following simple program while building with Visual Studio 2005. #include <iostream> #include <boost/date_time/posix_time/posix_time.hpp> void main() { { std::cout << boost::posix_time::second_clock::universal_time(); } std::cout << std::endl; } While printing the current time, the program crashes in Microsoft's 'xutility' header file (line 197) while trying to clean up some temporary strings created in time_facet.hpp (line 308), immediately after running boost::algorithm::replace_all(format, ss.str(), ""); I suspect the third parameter is the issue here (should it be surrounded explicitly with std::string("")?!). Has anyone else encountered this problem. Thanks, Bogdan

2 3

updated m4 macros for boost at Autoconf Macro Archive
by gelbeiche 03 Jun '06

03 Jun '06

Hi, for autoconf/automake users: The boost macros at http://autoconf-archive.cryp.to/index.html are updated. I also wrote a tiny cooking recipe how to use them, see http://randspringer.de/boost/index.html. Thomas

1 0

Re: [Boost-users] Boost Serialization & cross platform operability.
by Wu Yinghui, Freddie 03 Jun '06

03 Jun '06

> From: > "John Spicer" <jspicer(a)cabcds.org> > Date: > Fri, 2 Jun 2006 16:52:25 -0400 > > Has anyone had any luck getting a serialized object (created using boost > binary serialization) between two different platforms? I am currently > using binary serialization on 32 bit & 64 bit machines without issue. > However, passing the serialized objects between the two platform types > fail. The serialized classes themselves support interoperability, and as > such it appears that the serialization libraries themselves do not. I am > assuming they are storing some type of meta data along with the payload? > > Any thoughts on how to do this and keep with binary serialization? > > -- John Based on my own experience, boost binary archives are _not_ guaranteed to be portable between different hardware platforms. There are also some hints in the documentation, though I do not immediately recall where it is mentioned. (The problem is mostly due to endian mismatch and datatype size difference, if my memory is right.) What I did to counter this problem is to write a very simple subclass of binary archives and handle specifically the endian problem. The datatype size problem can be handled by serialization only types with specific size no matter on which platform, e.g. boost::int32_t, etc. Just 5 cents... Cheers, Freddie

3 2

Boost Serialization & cross platform operability.
by John Spicer 02 Jun '06

02 Jun '06

Has anyone had any luck getting a serialized object (created using boost binary serialization) between two different platforms? I am currently using binary serialization on 32 bit & 64 bit machines without issue. However, passing the serialized objects between the two platform types fail. The serialized classes themselves support interoperability, and as such it appears that the serialization libraries themselves do not. I am assuming they are storing some type of meta data along with the payload? Any thoughts on how to do this and keep with binary serialization? -- John

3 2

[BGL] kruskal_minimum_spanning_tree
by Cromwell Enage 02 Jun '06

02 Jun '06

The following code used to work, but no longer does. typedef boost::adjacency_list<> G; typedef boost::graph_traits<G>::edge_descriptor E; G graph; std::vector<E> spanning_tree; boost::kruskal_minimum_spanning_tree( graph , std::back_inserter(spanning_tree) ); Somehow the concept checking became too restrictive, and type E is now expected to be Assignable, which currently doesn't seem to be the case. Any fixes or workarounds? Cromwell D. Enage __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com

2 2

[BGL] problem with adjacency_list and bidirectional_tag
by Tiago de Paula Peixoto 02 Jun '06

02 Jun '06

Hi. I've noted an apparent discrepancy between the documentation of the adjacency_list class and the actual behavior: It's stated in the documentation that graph_traits<adjacency_list>::directed_category should be a typedef to "directed_tag" or "undirected_tag", depending on whether the graph is directed or not. Those are also the only possible types defined by the Graph concept. But when bidirectionalS is used as the Directed template parameter for adjacency_list, the type of the directed_category is set to "bidirectional_tag". Shouldn't this simply be set to "directed_tag", and the bidirectional information be confined to graph_traits<adjacency_list>::transversal_category? At least that's what the documentation says... I didn't provide a test program, since this can easily be checked in the source code. I'm using boost 1.33.1 and GCC 4.1.0. Thanks. -- Tiago de Paula Peixoto <tiago(a)forked.de>

3 2

When Boost::signals will be thread-safe?
by Tanguy Krotoff 02 Jun '06

02 Jun '06

Hi Boost::signals are not thread-safe http://boost.org/doc/html/signals/s04.html#id2738867 "Boost.Signals will support thread safety in the future." I would like to know when this feature will be implemented, if somebody is working on it. I like very much the fact that Qt slot/signals are thread safe, it helps really a *lot* : "the slot is invoked when control returns to the event loop of the thread to which the object belongs. The slot is executed in the thread where the receiver object lives." http://doc.trolltech.com/4.1/threads.html Do you plan also to make Boost::signals executed in the thread where the receive object lives? Thanks in advance for your answers Regards, -- Tanguy Krotoff <tanguy.krotoff(a)wengo.fr> http://openwengo.org

2 1

[Important] Mailing list outage on Saturday, June 3rd
by Douglas Gregor 02 Jun '06

02 Jun '06

All mailing lists and mailing list archives on lists.boost.org will be unavailable from 6am-9pm EST on Saturday, June 3rd due to a scheduled power outage. Doug

1 0

Poll: Do you use Spirit 1.6?
by Nicola Musatti 02 Jun '06

02 Jun '06

Hallo, This is a poll I started on the Spirit General mailing list which I'm reposting here as suggested by Joel de Guzman. I apologize to those who already read this message there. I'd be interested to know how many of you are currently using Spirit 1.6.x and with which compilers, to gauge the effort that might be required to take on maintenance of the 1.6 branch. Currently I'm using Spirit 1.6 with Borland's BCB6 and BCB2006, in conjunction with Boost.Serialization. Thank you, Nicola Musatti

6 6

Antwort: Re: [boost.multi_array] Question about iterators on multi_array_views
by bernhard.maeder＠zkb.ch 02 Jun '06

02 Jun '06

Hi Ron Thanks for your fast reply, I'll going to check out your fix as soon as possible. Thanks Bernhard |---------+-----------------------------------> | | Ronald Garcia | | | <garcia(a)cs.indiana.edu> | | | Gesendet von: | | | boost-users-bounces@list| | | s.boost.org | | | | | | | | | 24.05.2006 19:42 | | | Bitte antworten an | | | boost-users | | | | |---------+-----------------------------------> >-----------------------------------------------------------------------------------------------| | | | An: boost-users(a)lists.boost.org | | Kopie: | | Thema: Re: [Boost-users] [boost.multi_array] Question about iterators on | | multi_array_views | >-----------------------------------------------------------------------------------------------| Greetings Bernard, Indeed you have caught a bug in multi_array's view iterators. I have crafted a fix that I will check into CVS whenever Sourceforge is amenable. Thanks for the report. ron On May 24, 2006, at 8:07 AM, bernhard.maeder(a)zkb.ch wrote: > Hello everyone > > I'm trying to use the boost.multi_array's iterators and came across a > problem: calling end() on a certain multi_array_view only yields the > same iterator, if it is called on exactly the same instance of the > view. When, however, the view is created twice (using the same > expression, of course), end() produces an iterator that doesn't > compare to the first one. > > In other words: > m_a[indices[index_range()][1]].end() > != m_a[indices[index_range()][1]].end() > > I'm not sure whether this is just me, trying to use something in a way > it shouldn't be used, or if this is a limitation of the multi_array > library. I'd appreciate your opinion! > _______________________________________________ Boost-users mailing list Boost-users(a)lists.boost.org http://lists.boost.org/mailman/listinfo.cgi/boost-users ___________________________________________________________________ Disclaimer: Diese Mitteilung ist nur fuer die Empfaengerin / den Empfaenger bestimmt. Fuer den Fall, dass sie von nichtberechtigten Personen empfangen wird, bitten wir diese hoeflich, die Mitteilung an die ZKB zurueckzusenden und anschliessend die Mitteilung mit allen Anhaengen sowie allfaellige Kopien zu vernichten bzw. zu loeschen. Der Gebrauch der Information ist verboten. This message is intended only for the named recipient and may contain confidential or privileged information. If you have received it in error, please advise the sender by return e-mail and delete this message and any attachments. Any unauthorised use or dissemination of this information is strictly prohibited.

2 1