- Boost - lists.boost.org

[gsoc] heaps
by Tim Blechmann 11 Jun '10

11 Jun '10

hi all, since the heaps/queues project is likely to be used by multiple people, i think it may be a good idea to share some ideas on the list ... for now, i have been implementing a (non-intrusive) d-ary heap, which was very useful to do some prototyping of the actual interface. all data structures support the following template arguments: typename T // value type bool stable // stability class Cmp // compare object class Alloc // allocator for internal data structures the interface matches the forward container requirement, except that it does not support iterators (only const_iterators) and container comparison. at the moment, i distinguish between immutable and mutable heaps. the interface for immutable heaps would look like this: class immutable_heap { public: // priority queue support value_type const & top(void) const; void push(value_type const & v); void pop(void); // merge support void merge (immutable_heap const & rhs); void merge_and_clear (immutable_heap & rhs); // moves all elements from rhs to *this }; for mutable heaps, i have been using a `handle' to refer to specific elements: class mutable_heap { public: typedef handle_type; // priority queue support value_type const & top(void) const; handle_type push(value_type const & v); void pop(void); // update via function void update(handle_type const & node, value_type const & v); void increase(handle_type const & node, value_type const & v); void decrease(handle_type const & node, value_type const & v); // force update after element has been changed via the handle void update(handle_type const & node); void increase(handle_type const & node); void decrease(handle_type const & node); // merge support void merge (mutable_heap const & rhs); void merge_and_clear (mutable_heap & rhs); }; the forced update api is rather dangerous, since the heap would enter an undefined state, of the node is changed, but the heap is not updated. otoh, it would help to update values using a different api than the assignment operator. also, i am unsure, whether by default the heaps should be min-heaps (like in most textbooks on data structures) or max-heaps (like std::priority_queue). so i'd be curious, what people think about it ... the code is available from my public git repository. i would be curious to hear some thoughts! http://tim.klingt.org/git?p=boost_heap.git cheers, tim -- tim(a)klingt.org http://tim.klingt.org A year ago, six months ago, I thought that I was an artist. I no longer think about it, I am. Henry Miller

3 5

[Filesystem] Dual V2 + V3 support progress report
by Beman Dawes 11 Jun '10

11 Jun '10

I've created a branch to develop dual V2 + V3 support via namespace hoisting, as suggested by Scott McMurray, Andrey Semashev, and others. See https://svn.boost.org/svn/boost/branches/filesystem3 Seems to work well, and is lots easier to use that the switching approach currently in trunk. I've tested only on Windows, but will give Linux and Mac OS X a try later today or tomorrow. Namespace hoisting will break some existing V2 code: * User specializations in namespace filesystem. This is presumably very unusual, breaks noisily, and the fix is trivial, so I don't see it as a show stopper. * Possible ADL problems. I don't fully understand when this might be a problem, but neither V2 or V3 hit it in any of the tests, so I doubt it is very serious. --Beman

1 0

[msm] Problems with no_message_queue and pseudo exit states
by Michael.Herchel＠rohde-schwarz.com 11 Jun '10

11 Jun '10

Hi Christophe, I know your are quiet bussy releasing msm but we found something and we do not know whether it is the intended behaviour or a bug. We implemented a little state machine with a submachine. The submachine has an pseudo exit state. If we use "typedef no_message_queue" in the state machine as well as in the submachine, then the state machines get somehow mixed up and do not act normally once you leave the submachine via the pseudo exit state. In that case msm does not exit the pseudo exit state. We followed almost your example that you provide in the documentation. Is that known to you? Does it make sense to fix that or are there serious reasons why msm behaves like that? Regards, Michael

1 0

Re: [boost] boost::asio 1.43 regression (canceling read operation doens't trigger operation_aborted)
by admin＠thefireflyproject.us 11 Jun '10

11 Jun '10

Philippe Vaucher <[1]philippe.vaucher(a)gmail.com> wrote: > The problem is that when BOOST_ASIO_ENABLE_CANCELIO is not defined, > asio\detail\win_iocp_handle_service.hpp makes it so > impl.safe_cancellation_thread_id_ is equal to 0 and then when cancelling the > following code is executed: This problem seems to be windows specific. I compiled and ran Philippes testcase ([2]http://codepad.org/yYrQEqSo) on Debian GNU/Linux with Boost 1.43. I got successful results with both and without BOOST_ASIO_ENABLE_CANCELIO defined. I compiled with g++-4.1, 4.2, 4.3 and 4.4 - system specs were: wash@Dortmunder:~/sandbox/concept_tests$ uname -a Linux Dortmunder 2.6.26-2-amd64 #1 SMP Tue Mar 9 22:29:32 UTC 2010 x86_64 GNU/Linux Bryce References 1. mailto://philippe.vaucher@gmail.com/ 2. http://codepad.org/yYrQEqSo

1 0

boost log: attributes and different loggers
by Dirk Griffioen 11 Jun '10

11 Jun '10

Hi, I am trying to use 2 loggers - 1 severity logger - 1 'normal' logger However, when I add sink->set_filter(flt::attr< severity_level >("Severity") >= level); to the severity logger sink and try to use the second logger I get: terminate called after throwing an instance of 'boost::exception_detail::clone_impl<boost::exception_detail::error_info_injector<boost::log_mt_posix::missing_value> >' what(): Requested attribute value not found The second logger has no formatter: backend->set_formatter(fmt::stream << fmt::message()); Nor has it any attribute. I tried different solutions, but nothing works - what am I missing? Best, Dirk

2 2

boost::asio 1.43 regression (canceling read operation doens't trigger operation_aborted)
by Philippe Vaucher 11 Jun '10

11 Jun '10

Hello, I had working 1.42 code that suddenly stopped working with 1.43. With a bit of investigation I found out that canceling an async_read operation on a serial port when a timer was fired didn't trigger the expected boost::asio::error::operation_aborted like it used to do. Here is a testcase: http://codepad.org/XnLwybDl The idea is to implement a "read with timeout" functionality on a serial port. Btw, is this the correct way to implement this? Is there a simpler way to read from a serial port with a timeout? Thanks, Philippe

2 5

Re: [boost] MSM - 1.44 release
by Christophe Henry 10 Jun '10

10 Jun '10

>> Before the deadline, you'll need to merge your code to the release branch at >> <https://svn.boost.org/svn/boost/branches/release/>. > > Can you wait until we give you the go ahead. If we're slow then that's > our fault, so we're not going to stop you releasing. We need to check > the documentation, inspect report, test results etc. Yes, of course. The library is ready anyway so I can merge fast. If it could happen during the WE, it would be really great. Christophe

1 0

Re: [boost] MSM - 1.44 release
by Christophe Henry 10 Jun '10

10 Jun '10

>>Christophe Henry wrote: >> *Component has been added to Trac >> Not done. How to do? > > > I've done this. > > -- Marshall Thanks a lot Marshall! > Rene Riveira wrote: > You could do it, or you could post what you want in the release > announcement and Daniel will likely do it (or I will eventually). The > file to edit is > <https://svn.boost.org/svn/boost/website/public_html/beta/feed/history/boost…>. Done. It would be nice if you could check I did it ok. > Usually Daniel, or me eventually :-) Again, just post what you want. Or > if you are brave edit the database file yourself > <https://svn.boost.org/svn/boost/website/public_html/beta/doc/libraries.xml>. Done. >> *Component has been added to Trac >> Not done. How to do? > One of the admins will do it.. Which is often me.. Just tell me what > your trac user ID is. I suppose the same as svn: chenry Thanks, Christophe

1 0

MSM - 1.44 release
by Christophe Henry 10 Jun '10

10 Jun '10

Dear release managers, MSM is now almost ready for the 1.44 branch (hear hear, I didn't lie at the BoostCon ;-) ). I need a little help for some of the steps. The current status is: *root/index.html includes the library in its list of new libraries. Not done. How should I proceed? *root/libs/libraries.htm entry is OK, both in the alphabetic list and in the category lists. Done. *root/libs/maintainers.txt entry is OK. Done *website/public_html/beta/feed/history/ entry is OK Not done. Who does this? *The library has been added to the website documentation list Not done. Who does this? *The library's root/libs/xxx directory contains an index.html file with either the main docs or a redirection to the main docs Done *The primary docs pages meet Boost requirements and guidelines. Don't leave this until too late; it has turned up lots of issues in the past I suppose. Any comment on what I could have done wrong? *Component has been added to Trac Not done. How to do? *Trunk inspection report is clean Done (only irrelevant warnings about license in html files left) *Trunk regression tests are reasonably clean or marked up Done How do we proceed further? What do I also need to do to get MSM into the 1.44 branch? Thanks, Christophe

6 6

Bug sprint daily update for Thursday, June 10th
by Marshall Clow 10 Jun '10

10 Jun '10

As of 0800 Pacific time today, we had 854 open trac tickets. That is down by 40 from yesterday, and 135 since the start of the bug sprint! Of the 135 tickets closed, 100 were marked as "bugs". I am very impressed; keep up the great work! Total ticket counts: Saturday 989 Sunday 988 Monday 958 Tuesday 938 Wednesday 894 Thursday 854 Activity summary for yesterday (I'm counting tickets by hand here, here so blame my crummy math if it doesn't add up) New tickets: 6 Modified: 29 Closed: 49 Reopened: 3 If you're looking to help, and don't know what you can do, there are 13 outstanding "Support Requests" You can see them at <https://svn.boost.org/trac/boost/query?status=assigned&status=new&status=re…> Today, rather than the summary totals for yesterday, I have deltas; changes in the totals. If you'ld like to see the totals, they are available online at: Open ticket count by owner: https://svn.boost.org/trac/boost/report/18 Open ticket count by component: https://svn.boost.org/trac/boost/report/19 Open ticket count by type https://svn.boost.org/trac/boost/report/20 (Overwhelmingly, these are 'bugs') Open ticket count by milestone: https://svn.boost.org/trac/boost/report/21 (All changes are since the start of the bug sprint; 0800 PDT Saturday June 5th) Changed totals by library: thread -35 None -17 website -10 string_algo -9 function -6 iterator -6 concept_check-5 date_time -5 multi_array -5 Getting Started Guide -4 asio -4 random -4 tuple -4 variant -4 lambda -3 units -3 Building Boost -2 foreach -2 fusion -2 mpl -2 tokenizer -2 GIL -1 Regression Testing -1 assign -1 graph -1 io -1 lexical_cast -1 mpi -1 pool -1 regex -1 signals2 -1 spirit -1 trac / subversion -1 circular_buffer 1 program_options 1 serialization 1 utility 1 Documentation 2 property_tree 2 build 4 Changed totals by author: anthonyw -34 <none> -20 dave -13 grafik -9 pavol_droba -9 dgregor -8 jsiek -6 no-maintainer -6 az_sw_dude -5 djowel -5 garcia -5 chris_kohlhoff -4 ebf -4 agurtovoy -2 dlwalker -2 eric_niebler -2 matthiasschabel -2 bschaeling -1 cnewbold -1 danieljames -1 fmhess -1 hkaiser -1 hljin -1 jefffaust -1 johnmaddock -1 nasonov -1 nesotto -1 speedsnail -1 steven_watanabe -1 cornedbee 1 danielw 1 jano_gaspar 1 jewillco 1 matias 1 ramey 1 witt 1 vladimir_prus 7 Changed totals by category: Bugs -100 Patches -16 Feature Requests -11 Tasks -4 Support Requests -2 Library Submissions -1 -- Marshall

1 0