March 2006 - Boost - lists.preview.boost.org

Re: [boost] asio review results -- asio accepted
by Bulygin, Sergey 29 Mar '06

29 Mar '06

"Christopher Kohlhoff" <chris(a)kohlhoff.com> wrote: > The current plan is for the next version to include breaking > interface changes arising from the review, plus some features on > which I would like to get more feedback (such as custom memory > allocation and dual IPv4/6 support). At that time, in addition > to the changelog, I will publish a list of review changes that > are yet to be made. Congrats, Chris and all relevant community! Is the Asio task list available somewhere to see what is likely to be changed, compared to the current 0.3.6 version? What is the suggested roadmap for those, who wish to adopt Asio right now? Should we use 0.3.6 as is, or should we consider those "breaking interface changes" since they are going to change conceptual library design? Best Regards, Sergey Bulygin PS Just as a minor hint for review wizard: it's time to change Asio status in formal_review_schedule.html from "Result Pending" to "Accepted".

2 1

problem when using boost::noncopyable_::noncopyable
by bwlee 29 Mar '06

29 Mar '06

I defined a class inherited from noncoyable and singleton, but when compiled by vc8.0(VC++ express 2005), it report a error like this: template<typename T> class thread_pool : boost::noncopyable, public boost::singleton { //... }; center\code\project.net\include\thread_pool.hpp(227) : error C2248: 'boost::noncopyable_::noncopyable::operator =' : cannot access private member declared in class 'boost::noncopyable_::noncopyable' c:\boost\include\boost-1_33_1\boost\noncopyable.hpp(28) : see declaration of 'boost::noncopyable_::noncopyable::operator =' c:\boost\include\boost-1_33_1\boost\noncopyable.hpp(22) : see declaration of 'boost::noncopyable_::noncopyable' This diagnostic occurred in the compiler generated function 'core::thread_pool<worker> &core::thread_pool<worker>::operator =(robert::core::thread_pool<worker> &)' with [ worker=core::worker_thread ] I wonder why it need to generate function overriding operator '='. Is there any simple solution?

2 1

reference counted object?
by Neal Becker 29 Mar '06

29 Mar '06

I wonder if we boost has anything useful for implementing reference counting (or other gc) for objects? I know we have shared_ptr, but I was interested in objects rather than pointers.

4 3

[iostreams] status report (and apology)
by Jonathan Turkanis 29 Mar '06

29 Mar '06

Dear Boosters, Thomas Witt wrote me this morning to alert me of some iostreams issues which need to be resolved for 1.34. While I knew that I had been neglecting my boost duties recently, I'm embarrassed to see how long it's been since I responded to post on this list. I'm going to try to resolve all the outstanding iostreams issues this week. I'll read through all the old post on this list and respond to them; if I miss some, please let me know. I hope to see some of you next week in Berlin! Best Regards, -- Jonathan Turkanis www.kangaroologic.com

1 0

[lambda] documentation error
by Edward Diener 29 Mar '06

29 Mar '06

In the lambda documentation, at the end of the "Construction and destruction" section, the wrapped expression for "delete_array()(p)" is given as "delete p[]". It should be "delete[] p".

1 0

[lambda] documentation spelling error
by Edward Diener 29 Mar '06

29 Mar '06

In the lambda documentation, the last line of the section "Lambda expressions for control structures" reads: "As more experience is gained, we may end up deprecating one or the other of these syntaces." The last word should be spelled "syntaxes" according to my dictionary.

1 0

Responsibility for orphaned libraries
by Keith MacDonald 28 Mar '06

28 Mar '06

The mysterious disappearance of Jonathan Turkanis (not heard from here since last October; home page says he's at UCB, but a directory search at UCB does not find him) set me wondering about the risk of using Boost libraries. Like iostreams, they are generally one-man works, which may be beyond an end user's ability or remit to maintain. Even if the author does not disappear, he may eventually lose interest or simply be too busy to maintain it forever, so do Boost have any procedures in place for such eventualities?

7 8

Re: [boost] [optional] swapping initialized with uninitialized
by Václav Veselý 28 Mar '06

28 Mar '06

Ulrich Eckhardt wrote: > On Friday 10 March 2006 17:14, Vaclav Vesely wrote: > > When two optionals (one initialized and other not) are swaped, then the one > > value in uninitialized optional is constructed and the other one in > > initialized optional is destructed. > > > > IMHO it's not necessary. It should be more effective to swap optional's > > m_storages without any values constructiona and destruction. ... > It could only be done if the object was moveable, i.e. depending on a > type_trait. Looking at the Boost type_traits library, I wonder why this trait > doesn't exist - many things can be sped up using memcpy when is_pod is true, > but even things like vector/string/deque (list/map/set depending on whether > their root node is stored by pointer) can be moved in memory without negative > effects (yes, talking about typical implementations here, not guarantees). I see now. But IMHO is_pod is not satisfactory condition for is_movable. For example the class struct X { X* me; // Can points to the instance }; seems to be POD, but it's not movable. The best implicit implamentation is probably the has_trivial_copy trait. For each other moveble type must be is_movable explicitly specialized. (BTW for compiler writers: compiler support for test a trait on all class members would be very useful for many such traits. And existing compilers with support of has_trivial_copy and similar traits must use this technique internally.) I think it worth it. For example I use optional<string> frequently. If you think that the is_movable trait is good enogh for Boost I would like write it including test and documentation. Regards, Vaclav

6 11

[hash] typo in tutorial
by Joaquín Mª López Muñoz 28 Mar '06

28 Mar '06

the first code snippet of Boost.Hash tutorial reads: std::unordered_multiset<std::vector<int>, boost::hash<int> > set_of_ints; ... I guess it should read: std::unordered_multiset<int, boost::hash<int> > set_of_ints; ... Best regards, Joaquín M López Muñoz Telefónica, Investigación y Desarrollo

1 0

[release 1.34] merging index.htm and CVS $Date$ feature
by Joaquín Mª López Muñoz 28 Mar '06

28 Mar '06

When merging changes in boost/index.htm from HEAD to RC_1_34_0, seems like there's no way to have both branch and trunk to show *exactly* the same contents, as the CVS $Date$ facility kicks in on every commit thus automatically desynching files when a merge is tried. This $Date$ stuff can be seen at about the end of index.htm: <p>Revised $Date: 2006/03/28 07:01:13 $</p> So, the result (if I'm not wrong) is that you'll always have merge conflicts when synching this file between trunk and branch, or the other way around, no matter how carefully you play with merge ops and merged_to_RC_xx tags. * Is this really so, and it's a (minor) problem we have to live with? * Am I wrong and there's a way to handle the issue? Thank you, Joaquín M López Muñoz Telefónica, Investigación y Desarrollo

2 2