July 2007 - Boost-users - lists.preview.boost.org

[asio] when deadline_timer::cancel() returns
by Yuval Ronen 29 Jul '07

29 Jul '07

The Asio docs for deadline_timer::cancel say that "This function forces the completion of any pending asynchronous wait operations against the timer". Let's say I have a thread executing io_service::run, and is now invoking a completion handler of a deadline_timer. While this completion handler is underway, a second thread is calling cancel() on that deadline_timer. Is the cancel() call supposed to block until the completion handler is finished? To my best understanding, both according to the docs, and according to what seems to be logical to me, the answer is "yes, it should block". However, it doesn't. Is it a bug, or by design? Yuval

1 1

[mpl] what is not a placeholder expression
by "JOAQUIN LOPEZ MU?Z" 29 Jul '07

29 Jul '07

According to the MPL reference, a placeholder expression X<a1,..,an> must satisfy (among others) the following condition: "All of X's template parameters, including the default ones, are types." According to that the following should then compile without errors: #include <boost/mpl/apply.hpp> #include <boost/mpl/assert.hpp> #include <boost/type_traits/is_same.hpp> using namespace boost; using namespace boost::mpl; template<typename T,int=0> struct foo { template<typename Q> struct apply{ typedef Q type; }; }; typedef apply1<foo<_1>,int>::type t1; BOOST_MPL_ASSERT((is_same<t1,int>)); int main(){} However, I've tried with GCC 3.4.4 and the static assertion fails! I'd like to know: 1. Whether I'm right in assuming that the code above should compile without assertion failures. 2. What the behavior is for other versions of GCC and compilers: reports most welcome!! 3. Whether this is a known problem specific to GCC or, more generally, what the status of this issue is. Thank you, Joaquín M López Muñoz Telefónica, Investigación y Desarrollo

5 5

Trac / Register User
by Olaf van der Spek 28 Jul '07

28 Jul '07

On 7/27/07, Chris Weed <chrisweed(a)gmail.com> wrote: > > Where are the links to Trac from http://boost.org/? > > I've been trying to find Trac before but didn't succeed. > > Report Bugs -> Bug tracking facility > > http://svn.boost.org/trac/boost/report Thanks. But where is the register user link? I only see login.

2 3

[BGL] Parameter: CompareFunction - dijkstra_shortest_paths
by Leandro T. C. Melo 28 Jul '07

28 Jul '07

Hello all. It is clear for me the reason for template parameter CombineFunction in dijkstra_shortest_paths. For instance, prim_minimum_spanning_tree makes use of it to correct compute the MST. Basically, in Dijkstra's it is something like this: distance = distance of vertex v + weight of edge e; And in Prim's only the weight is considered: distance = weight of edge e; This is reflected on the argument used for the CombineFunction in Prim's function (from file prim_minimum_spanning_tree.hpp): prim_minimum_spanning_tree(...) { typedef typename property_traits<WeightMap>::value_type W; std::less<W> compare; detail::_project2nd<W,W> combine; dijkstra_shortest_paths(...); } Where _project2nd is: template <class U, class V> struct _project2nd { V operator()(U, V v) const { return v; } }; However, it is not clear the reason of the CompareFunction template parameter (which is a predicate). First, because, I don't see a situation where not using std::less<Some_type> would make any sence, because of the way the Dijkstra's algorithm is constructed. Second, because the value_type of the DistanceMap must be the same of the value_type of WeightMap (according to the documentation). Therefore, why the following two lines of code in Prim's implementation (from the function I pointed at beginning of the email)? typedef typename property_traits<WeightMap>::value_type W; std::less<W> compare; Since, in Dijkstra's algorithm the only difference is that DistanceMap is used as an argument for the definition of the value_type from property_traits. Am I loosing something here? If yes, would anyone explain it to me? Thank you, Leandro.

1 0

[parameter] Parameter-Enabled Class Templates: combining keyword and deduction for the same parameter
by Joaquín Mª López Muñoz 28 Jul '07

28 Jul '07

The docs explain that a parameter can be named: typedef parameter::parameters< required<tag::threading_policy> ... > class_signature; ... class_<threading_policy<single_threaded>,...>; or else be deduced: typedef parameter::parameters< required< deduced<tag::threading_policy>, is_base_and_derived<threading_policy_marker,_> > ... > class_signature; ... class multithreaded: threading_policy_marker{...}; class_<multithreaded,...>; Now, is it possible to combine these two options into one, so that the the two syntaxes class_<threading_policy<single_threaded>,...>; class_<multithreaded,...>; are accepted? If so, how? Thank you, Joaquín M López Muñoz Telefónica, Investigación y Desarrollo

5 7

Program Options: parse_config_file: How to ignore unknown options?
by Olaf van der Spek 28 Jul '07

28 Jul '07

Hi, How do I tell this function to ignore unknown options?

2 1

Boost Threads and Cancel Method
by Jeffery Cavallaro 27 Jul '07

27 Jul '07

Per the C++0x discussions earlier in the week, and the calls for the Bjarne, et al to make some definitive decisions regarding a threading API, I would ask the Boost thread developers not to wait and please add a cancel method now. A lot of the scheduling priority stuff can wait, but it is going to be real hard to turn people from pthreads until some sort of a cancel method, however basic, is available. I believe that this is the state of things; however, it not, someone please feel free to correct me. Jeffery Cavallaro Postini Engineering 650-486-8342 ----------------------------------------------------------- This message may contain confidential and/or privileged information. This information is intended to be read only by the individual or entity to whom it is addressed. If you are not the intended recipient, you are on notice that any review, disclosure, copying, distribution or use of the contents of this message is strictly prohibited. If you have received this message in error, please notify the sender immediately and delete or destroy any copy of this message.

1 0

program_options: documentation dead end
by Steven Mackenzie 27 Jul '07

27 Jul '07

I am trying to understand composing, multitoken and zero_tokens methods of the typed_value class. This link http://www.boost.org/doc/html/boost/program_options/typed_value.html#id6986… is to the composing method from the reference page for the typed_value class: it suggests looking at the is_composing documentation, which is unfortunately empty. Can this be fixed for Boost 1.34.1? Also, can someone explain to me what the multitokens method does. The doc states "Specifies that the value can span multiple tokens." What is the definition of a token here? Finally, what does zero_tokens() do? Thanks in advance Steven

8 25

[Important] Boost moving to Subversion on Tuesday, July 31st
by Doug Gregor 27 Jul '07

27 Jul '07

============================================= Boost will be moving to Subversion on Tuesday, July 31st. ============================================= === What's happening === We are converting the existing Boost CVS repository, hosted by SourceForge, over to Subversion. The contents of the CVS repository will join the Boost sandbox and (beta) web site in the Boost Subversion repository, hosted by the Open Systems Laboratory at Indiana University. === Timeline === At 8:00 AM EDT on Tuesday, July 31, we will be cutting off access to the current Boost CVS repository. The Boost Subversion repository, mailing list archives, Boost Trac, beta.boost.org, and anything else hosted at OSL will also be unavailable during this time. We expect to have everything back online by midnight Tuesday. I will send another note to the mailing lists when the conversion is complete. The mailing lists and the main Boost web site should still work during this time. === What you should do === If you have any outstanding changes to Boost in your local CVS tree, commit them before 8:00am EDT on Tuesday. Any changes to your local CVS tree that have not been committed to CVS will be lost and will have to manually be moved over Subversion. The Boost CVS repository will not be reopened. If you have not done so yet, familiarize yourself with Subversion, read the Boost Trac page about the Boost Subversion repository, and, if you have write access to CVS, request a Subversion account: http://svn.boost.org/trac/boost/wiki/BoostSubversion === Who's in charge? === I will be the primary contact person for questions about and problems with the Boost Subversion repository. Most questions should be directed to the Boost mailing list. Account requests should go to all of the moderators at boost-owner(a)lists.boost.org. === How you can help === In several Boost tools and many places in Boost's documentation, we refer to the Boost CVS repository. We need help upgrading these tools and fixing the documentation to refer to the Boost Subversion repository. Specifically: * regression.py, used by regression testers, must be updated to use Subversion * Various developer and release-manager documentation talks about Boost's CVS, but should now refer to Subversion. Better yet: migrate this documentation over to the Boost Trac, where it is easier to maintain and update, and won't be a part of the Boost distribution * Write Subversion-specific documentation on the Trac. Quick-start guides, tips and tricks, etc. would be helpful. One particularly important bit: using svnmerge.py to simplify the handling of release branches. Doug Gregor Open Systems Lab @ Indiana University

1 0

portable lock-free reference counting
by Bryan Green 27 Jul '07

27 Jul '07

Hello, I have a class that contains a reference count. I would like to maintain the count without using a mutex. Can anyone recommend a way to do this portably, using Boost? I see the needed mechanics in 'detail/sp_counted_base.hpp', but since it's in 'detail', I'm not sure what the most portable (and maintainable) way to use it is. Thanks, -Bryan

1 0