February 2010 - Boost - lists.boost.org

Re: [boost] [Review] ITL
by Wuttke Peter 27 Feb '10

27 Feb '10

To Begin with the most important point: YES, boost::ITL should be accepted as a boost library. (Please be don't be to disappointed of my english, for i'm a developer from Germany, and this would be my first review on Boost). The ITL (interval template library) is a very valueable part of our (the Cortex Software GmbH, incidently the company i work for) daily development routine. In our industry (which is health care) sampling of Status-Informations for collections of evaluations (in time) are an essential part of our work, so we are using the ITL-Library for some years with great success. (We have to aggregate Events based on equivalent properties, eg. same admission - which diagnosis, same discharge - which diagnosis, and so on. ) . As far as I know, there is really no alternative for using the ITL do work with Intervals - I've tried is with plain SQL, and, belive me: It's possible - but really no fun - no,no,no. (I've also seen Experiments trying to implement the ITL-Solution for specific problem-domains - they aren't usefull or elegant, either. Forget them.) Until now, we only used the library for intervals of datetimes (coded as integers), but i Im sure, that its usefullness goes far beyond. Concerning the overall architecture, I think, the actual version is more than complete, I'm especially impressed withn the idea of big, sparse bitsets implmenented by the ITL - it's cool. The documentation (using samples) is clear and completly comprehensive - couldn't hope for more. So, being not a specialist in template/generics programming (or in algebraic specification), I LIKE this Library - and use it.

2 1

[random] new documentation
by Steven Watanabe 26 Feb '10

26 Feb '10

AMDG This is hopefully the final version of the work I did back in November to convert the Random docs to Quickbook/Doxygen/Boostbook. I've uploaded the generated documentation to http://www.cs.hmc.edu/~swatanabe/random_doc/libs/random/doc/html/ The patch itself is rather large, even compressed, so I uploaded it to http://www.cs.hmc.edu/~swatanabe/random.patch.zip Okay to commit? In Christ, Steven Watanabe

9 19

[exception] virtual inheritance troubles
by Eric Niebler 26 Feb '10

26 Feb '10

In xpressive, I'm trying to follow the recommended use of Boost.Exception and use virtual inheritance in my exception class, regex_error. I'm also trying to hew as closely as possible to the standard, which says that regex_error should inherit from std::runtime_error. However, this simple use of Boost.Exception does not compile: #include <string> #include <exception> #include <boost/throw_exception.hpp> #include <boost/exception/exception.hpp> struct regex_error : virtual std::runtime_error , virtual boost::exception { explicit regex_error(std::string const& what) : std::runtime_error(what) {} }; int main() { boost::throw_exception(regex_error("what")); } On VC9, I get the following: > C:\boost\org\trunk\boost/exception/exception.hpp(400) : error C2512: 'std::runtime_error::runtime_error' : no appropriate default constructor available The trouble is that std::runtime_error does not have a default constructor, and that boost::throw_exception doesn't simply throw the exception passed to it; it throws an object of type of clone_impl<regex_error>. Because of the virtual inheritance, clone_impl needs to call std::runtime_error's directly, but it doesn't. It can't know to do that in general. What is the recommended practice here? <aside> It actually doesn't seem to be particularly useful in this case to use virtual inheritance because std::runtime_error inherits from std::exception non-virtually. In order to avoid the pitfall described here <http://tinyurl.com/yb2jpns>, it seems *ever* exception type in the hierarchy needs to use virtual inheritance, and that unfortunately is not the case for the standard exception hierarchy. </aside> -- Eric Niebler BoostPro Computing http://www.boostpro.com

4 6

[forward] looking for help to keep MSVC7.1 from ICEing
by Tobias Schwinger 26 Feb '10

26 Feb '10

Hello, I'm trying to make Forward release-ready and I'd really like to support MSVC 7.1 which is currently ICEing (in msc1.cpp, line 2701). According to online research that ICE typically occurs when the ODR is violated and the compiler fails to diagnose it. Although the code should be valid, it's likely to be about name management. The code uses an inheritance chain to add call operators of different arities. IOW the class template for arity N inherits from the class template for arity N-1. A 'using base::operator();' is used to pull in the base class' overloads. A stripped-down version of this pattern compiles. When I take out the 'using' declaration the ICE disappears (and so do the base class' overloads). Ideas anyone? Anyone keen enough to look into it? In case of the latter, I recommend the simpler of the two headers: trunk/boost/functional/lightweight_forward_adapter.hpp [ link to svn: http://tinyurl.com/ydoyoa4 ] Thanks, Tobias

1 0

Boost.Polygon : How to deal with per point attributes
by philippe.lamontagne.2＠ens.etsmtl.ca 26 Feb '10

26 Feb '10

Hi Boost community, I am currently experimenting with the BOOST polygon library and I have a few questions. Question 1: For my specific application the points inside the polygons need to contain attributes in addition to their x and y coordinates. This extra information is required in a per-point basis. I need to compute intersections and differences of polygons. The operations can be performed regardless on the attributes, but it's important that the attributes be present in the polygons resulting from the operations. Does BOOST provide a mechanism for keeping such per point attributes? I followed the example to make my own point and polygon classes work with BOOST, but for what I can see, the polygons operations are performed using GTL's own data types and transferred back to my types, loosing any attributes in the process. I am missing something here? Question 2: I noticed that if I have multiple points located on the same straight line only the initial and final points will be kept in the resulting polygon and that the intermediate points will be discarded. Whereas this behavior is probably preferable in the general case, it causes me a problem if I want to maintain point attributes because the attributes are not guaranteed to be the same all over the straight line. Is there a way to change this behavior? I'm aware that polygon operations may create new points. I can deal with a few empty attributes as long as original points keep their initial attributes. We hope there are some solutions or workarounds because I strongly appreciate the library. Thanks for the excellent work.

2 2

[Boost][Container] Non mandatory template parameter not defaulted
by Mathieu - 26 Feb '10

26 Feb '10

Hi, I was reading the boost.container vector sources, and I can't figure how this vector instantiations can work : boost::container::vector<int> foo; considering that the vector definition looks like : template <typename T, typename A> vector; As the second template parameter A has no default type, I believe it's actually mandatory but it's not the case as this : boost::container::vector<Type> compiles fine. I've been reading through the whole source file but I can't find where the trick is. if anyone can enlighten me? ps : sorry for my bad english..

2 2

[filesystem] 1.40 documentation refers to wrong API version?
by Johan Nilsson 26 Feb '10

26 Feb '10

Hi, the Boost.Filesystem documentation at e.g.: http://www.boost.org/doc/libs/1_40_0/libs/filesystem/doc/reference.html#Ope… seems to refer to more recent version. As a specific example, there are no overloads of remove in 1.40 that accepts an instance of boost::system::error_code (as opposed to what the documentation indicates) - easily verifiable in e.g.: http://svn.boost.org/svn/boost/tags/release/Boost_1_40_0/boost/filesystem/o… Is the actual 1.40 documentation available online somewhere? / Johan

1 0

Re: [boost] review system in place is extremely slow? (was Re: [rfc] rcpp)
by Brian Wood 26 Feb '10

26 Feb '10

Mathias Gaunard writes: >> Currently there are a lot of libraries to review, but no review managers. >> That means that the user community don't want to spend a little bit of >> their time to manage a review. >> >> In addition, the last review didn't had too much of reviewers >> (I'm also concerned by this point) > > So you can see, indeed, that this is fairly concerning. I think this may be due to economic changes. When Boost was getting started the economy was better than today. I think Boost should work on turning traffic on it's pages into advertising revenue. That money could be used to help those who want to be involved in the project. Giving a review manager $200 for 40+ hours of work only makes sense. -- Brian Wood Ebenezer Enterprises http://webEbenezer.net (651) 251-9384

5 4

Re: [boost] review system in place is extremely slow? (was Re: [rfc] rcpp)
by Brian Wood 25 Feb '10

25 Feb '10

On 02/25/2010 11:56 AM, Nevin Liber wrote: > Then comes the "which jobs do we pay for?". If a reviewer > spends 80 hours doing a detailed review, should we pay > them too? What about all the people who tirelessly work > on getting a release out the door every quarter? Why > should they work "for free" if other people are getting paid? I thought about some of these things as well.... I think release managers should be paid as well and more than review managers. For reviewers, I would set up a system that rewarded them also. The review manager might be expected to judge who has given the best and second best reviews. It would add a little excitement to the whole process... seeing who wins the gold and silver medals. > If we don't make enough money from web advertising, you > can't just not pay the review managers. Who will make up > the deficit? Etc., etc. > The way I'd do it is to make offers based on money already collected and not based on anticipated earnings. That way you commit some income to an individual and should be able to follow through and pay them. It might take a few months before you had enough saved to begin doing this. Advertisers love to reach niche audiences, especially audiences with above average incomes. This approach would be mixed with the current approach. If there was money to pay for a review manager it would be offered. If not you would just keep going hoping some generous soul volunteers. -- Brian Wood Ebenezer Enterprises http://webEbenezer.net (651) 251-9384

1 0

Fwd: [Boost-users] [Review] ITL review starts today, February 18th
by Joachim Faulhaber 25 Feb '10

25 Feb '10

Hi, this question of Mathias was posted in both [Boost-users] and [boost] but the answer was accidentally send to [Boost-users] only. ---------- Forwarded message ---------- From: Joachim Faulhaber <afojgo(a)googlemail.com> Date: 2010/2/23 Subject: Re: [Boost-users] [Review] ITL review starts today, February 18th To: boost-users(a)lists.boost.org 2010/2/23 Mathias Gaunard <mathias.gaunard(a)ens-lyon.org>: > Hartmut Kaiser wrote: > >> Interval containers allow to access and manipulate collections of >> intervals >> on an abstract level as sets and maps (of elements), but also to exploit >> segmental information that is given by the interval borders. Interval >> containers are fairly generic. They can be instantiated with discrete >> domain >> (or key) types like integral numeric types or boost::date or time types. >> But >> they also work with continuous domain types like double, rational or >> string. >> Using the latter allows to represent infinite sets and maps. >> >> The interface of the interval containers strives to be very intuitive and >> in >> line with the stl and other boost libraries. >> >> * The set theoretic operations union, difference, intersection >> and symmetric difference are available as operators for all >> interval containers and for many useful overloads between them. > > I have no knowledge of the problem domain (nor did I actually look at the > library yet, sorry about that), but I simply have a question: how does this > compare to geometry libraries or spatial indexes that provide that kind of > facilities for arbitrary dimensions? > > I certainly wouldn't want to force over-generalization to the authors, but I > am wondering how much is gained by limiting things to one dimension. So can this library survive in a field of power called boost of an ever abstracting ever generalizing and ever optimizing bunch of thinkers? A few points: * Geometry usually is based on a numeric coordinate type that comes with a distance measure. Many algorithms in geometry libraries rely on that. Intervals and interval containers of the ITL, similar to SortedAssociativeContainers, only require a strict weak ordering on their key type. So interval containers are less restricted and allow for a larger class of instantiations. You may for example want to work with interval sets of strings. * Interval containers implement interval combining styles that can be useful in providing and manipulating certain time grids. E.g. http://www.herold-faulhaber.de/boost_itl/doc/libs/itl/doc/html/boost_itl/ex… I'm not aware of something like that in geometry. * There are not only interval sets but also interval maps. itl::interval_map is the real workhorse in those real world applications that already exist. interval_maps provide a general idiom of aggregation of associated values. This idiom is very useful for computing statistical data through aggregation that occur in time. * Sometimes a more simple and basic class of data structures is advantageous. It may be simpler to implement optimizations. Like Lukes Manhattan integral polygons compared to more general GGL polygons. * Transfer of geometry knowledge to the more simple models of interval sets and maps may be a step users may not come up with. * The basic set and map concepts of the ITL allow for a variety of applications that may go beyond the scope of geometry. One such example is the self compressing large bitset, which not really has the look and feel of geometry ... http://www.herold-faulhaber.de/boost_itl/doc/libs/itl/doc/html/boost_itl/pr… ... or the example of a system of authorisations. See the example user_groups http://www.herold-faulhaber.de/boost_itl/doc/libs/itl/doc/html/boost_itl/ex… * Well known data structures from geometry like interval_tree or spatial indices do have different focuses and different applications. An interval_tree is more space efficient than an interval_map that can computes all overlaps at once. This may be desired if you need all of them. If you don't and you have reoccurring arbitrary stabbing queries interval_trees are the better choice. * itl::Maps are defining a generalized addition, which I think is a quite powerful feature that accounts for most of the goodies in the applications that we have done. This addition function is a generalisation of the common insert function on SortedAssociativeContainers and it introduces the aggregating properties of itl::Maps. The aggregate on overlap idiom is independent of the question, if the domain_type of the map or interval_map is a scalar or a tuple (has dimensions) or something else. There are already interesting applications of aggregations using itl::maps of tuples, but this exceeds the scope of the core library. Cheers, Joachim

1 0

[date_time] int64 overflow bug with subsecond duration
by Zachary Turner 25 Feb '10

25 Feb '10

Consider the following code, very simple in nature: static const boost::gregorian::date windows_epoch_base(1601, 1, 1); date_time d(windows_epoch_base, microseconds(12908391903991608)); This should resolve to sometime in early 2010 Now, the problem is that this number is already really big, not that many factors of 10 away from the int64 overflow boundary. microseconds() is really just the subsecond_duration<> template, and on my machine it's being parameterized as subsecond_duration<boost::posix_time::time_duration, 1000000>. The code for the constructor is as follows: explicit subsecond_duration(boost::int64_t ss) : base_duration(0,0,0, ss*traits_type::res_adjust() / frac_of_second) {} res_adjust is also resolving to 1000000. So compiler is doing 12908391903991608*res_adjust first, which causes an overflow. I think it should be doing ss * (traits_type::res_adjust() / frac_of_second) instead. I don't know a whole lot about date_time internals, so I thought I would ask here before just submitting a bug. Is the proposed fix correct? I made the change locally and it does indeed appear to resolve the problem. Zach

2 3

[move] will boost::rv work with types with virtual destructor?
by DE 25 Feb '10

25 Feb '10

sorry for a dumb question i noticed boost move lib mentioned in a recent message and decided to look at it i noticed that destructor for boost::rv class is declared but not defined so it can no longer be generated automatically then if an arbitrary type has a virtual destructor and we want to parameterize the boost::rv template with it a vtable for the boost::rv<type> must be generated among other things it must hold a pointer to rv<type>::~rv which is not defined is that really a problem? -- Pavel

2 2

Spirit V1.8 to V2.1 Porting Errors
by Alireza Bemanian 25 Feb '10

25 Feb '10

Hi, Trying to port the code to V2.1, a number of redefinition errors get generated by the compiler. The first instance is the extern InterLockedIncrement variable at line 73 of the interlocked.hpp file. The code is segmented for different platforms using #if #elif and so on directives. The project is at the same time ported from Visual Studio 2008 to 2010 RC1, and the redefinition error is caused by compiler not excluding parts of that file. The define by the Visual Studio is the _WIN32_WINNT base on the use of the 32 bit development PC. This definition is not listed by the #elif chain within the interlocked.hpp code and it seems that BOOST_USE_WINDOWS_H is used instead. Knowing the segments are coded to be mutually exclusive, still the code generates errors? Regards, Reza Bemanian

2 4

Re: [boost] is review system in place is extremely slow? (wasRe:[rfc] rcpp)
by Pete Bartlett 25 Feb '10

25 Feb '10

>From Robert Ramey > > to move those 3 libraries to the 'detail' namespace > > of Boost.Task and have review as it is, as opposed to waiting. What > > do you think? >I think I caught hell for doing something similar in the serialization >library. I had to make a number of components such as >BOOST_STRONGTYPEDEF, state_saver, smart_cast, etc. >which I put into boost - (not detail) and year afterwards this >was raised as a huge problem. And this was even though the >components had been their through two reviews. So I would >be careful about doing this. This seems a little more negative than I remember. If we are thinking of the same thread (see e.g. http://lists.boost.org/Archives/boost/2007/11/130567.php ) then a lot of the discussion was about headers directly in the boost root directory rather than a subdirectory, rather the issue at hand. My opinion is that there is no overarching programming design deity that is going to come down and smite us if we be a bit pragmatic. So what if Move [*], Fiber and Atomic get "smuggled" in if Task gets reviewed and accepted first? They don't have to be documented as accepted - on the contrary the documentation could have appropriate warnings as *not* formally accepted apart for indirect use via Task. >Another issue is: if Boost.Task depends upon Boost.Fiber >and Boost.Atomic, what happens if the Boost.Fiber or >Boost .Atomic are not approved? Sometimes the laudable goal of perfection turns open source projects into games of Nomic! Again, pragmatism should play a role here else Boost will never be the rich and full set of libraries that it fantastically could be. Pete [*] Move seems so core these days - another bit of pragmatism would see that fast-tracked through.

5 5

typeof or similar
by Steffen Roeber 25 Feb '10

25 Feb '10

Hi, I have 3 classes: 1. class Base { public: virtual Base *createDerived() { return 0; } }; 2. template <class T> class BaseT : public Base { public: virtual Base *createDerived() { return new DerivedT<T>(); //nicht smart, because DerivedT must be known } }; 3. template <class T> class DerivedT : public BaseT<T> { }; So far so good. Or not. But now the problems are starting. Base *base = new BaseT<int>; This object it created at a place that is not known where I want to create: Base *derived = base->createDerived(); //base is a Base * The result is an object that has the same template type as 'base'. Finally I got my desired result, but the problem is the relation downward relation between BaseT and DerivedT. My preferred way would be something like a factory: class Factory { public: template <class T> static Base *createDerived<T>() { return new DerivedT<T>(); } }; But how can I tell the factory which type to be used. Base *derived = Factory::createDerived<sameAsBase> (base); //base is a Base * I think there is no "normal" C++ language support. But Boost has some methods that by-pass some language constraints. I took a loot at something like 'typeof'. But I don't get the things together. Has anybody any idea? Or is it impossible to do this? Kind regards Steffen

4 3

phoenix docs hidden
by Neal Becker 25 Feb '10

25 Feb '10

In 1.42, there is no link at top level docs to phoenix. It's fairly hidden under spirit. Isn't phoenix intended to be used independently?

2 1

[phoenix] typo in docs
by Neal Becker 25 Feb '10

25 Feb '10

In algorithms doc: for_each(r, c) << Shouldn't that say 'foreach(r, f)' to match this: for_each(begin(r), end(r), f)

1 0

[fusion] make_unfused_generic broken
by Rutger ter Borg 25 Feb '10

25 Feb '10

Hello, apparently the header #include <boost/fusion/functional/generation/make_unfused_generic.hpp> got broken somewhere in the upgrade process, including it yields lots of errors. Boost version 1.42. (replaced with make_unfused?) Cheers, Rutger

2 1

Re: [boost] [msm] scalability
by Christophe Henry 25 Feb '10

25 Feb '10

>Sorry for the delay. That is awesome Christoffe! I've turned off /Gm for >all my projects and my state machine now compiles in the 6-sec range, >and Incredibuild is able to keep the entire compile farm running at >capacity. You shaved >30% off my full build times. Great :) In case your state machines grow bigger, you can also give a try with the "faster compile" policy which I will further improve after inclusion. I think we should make /Gm a forbidden option. I will of course write about it in MSM's doc but it would be interesting to see which libraries also profit from disabling this option and we could add the tip to their doc too. Christophe

1 0

Re: [boost] [Review] ITL review starts today, February 18th
by Jeff Flinn 25 Feb '10

25 Feb '10

Hartmut Kaiser wrote: > Hi all, > > The formal review of the Interval Template Library (ITL) starts today, > February 18th, 2010 and will end February 27th, 2010. ITL is being developed by Joachim Faulhaber. > > ------------------------------------------------ > > About the library: > > The Interval Template Library (ITL) is a library of interval containers that > facilitates computations on collections of intervals and associated values. ... Hmm, not much discussion yet. While this is not my review I'll make a few comments. I was able to port at least my necessary portions of split interval map to CodeWarrior 9.4 and was successful in using it to generate a column coverage graph of what is effectively the columns of possibly large sparse matrices. I would have expected the containers to be templated on an interval type, negating the need for run-time determination of boundtype. With the current design, a rather heavy interval is required with it's boundtype data member. My domain type is a simple int, an index into a row of a sparse matrix. Our in-house Interval class is always right_open, so there is only the need for a begin and end value leading to minimal memory usage. The current library design requires me to convert between ITL intervals and my own application's intervals. I'd have like to be able to provide a model of an interval concept that the ITL container would use. I wasn't able to find any description of the class invariants for interval. For example in our own Interval class the upper/end value must be greater than/equal to the lower/begin value which is a precondition to the constructor taking a pair of domain values. We do provide a static makeNormalized method to produce a proper interval when the order of the domain arguments is unknown, which is not very often in our usage so you don't pay that cost. In my case I hadn't the need for 'reversed' intervals, but I could see that being required for some use cases. Does ITL handle that? Along the same lines I've found little need for the mutators. An interval really only needs default and copy constructors, and one taking the upper and lower values. All mutations merely construct a new interval using calculated values and begin/end values from source intervals. Jeff

6 15

Review of Interval Template Library (ITL) for inclusion
by Eric Jonas 25 Feb '10

25 Feb '10

At the outset, I want to say that I enthusiastically support the inclusion of this library in Boost. I also encourage others without a specific application in mind to check out the library, as there are a lot of fun examples that you might later find useful in other coding projects. My Background -------------- My background: I develop combined hardware and software systems for recording from high-throughput neural systems. One of the primary challenges we face with such systems is the aggressive demands that 10 GB/hour of time-series data places on a typical workstation when you want to visualize it -- especially in real time, with appropriate historical context, at multiple zoom levels. I wrestled with this problem for a great deal of time before settling on various rendering and caching mechanisms. But the number one challenge was efficiently representing the temporal regions that had been viewed, those that had been rendered, and those that had been cached. Then I needed to propagate this state information through a complex dataflow pipeline. ITL solved this problem for me, in a way that honestly freed me from having to think about a lot of the unsavory low-level details. Specifics ------------ All of my work has been on Linux with a modern (4.3+) gcc. I have had collaborators build and use the resulting artifacts on OS X. I originally only anticipated needing interval_sets, but interval_maps became extremely useful as I started to keep track of metadata in my timing epochs. The segregation of interval_* and split_interval_*, while not necessary for my application, provides historical context that I can imagine being useful in some scenarios. What is your evaluation of the documentation? --------------------------------------------- I'd say the documentation was "Boost-quality", but some might see that as pejorative :) The tutorial had me up and running quickly, and the extensive examples helped me map my ideas to the library's primitives. How much effort did you put into your evaluation? A glance? A quick reading? In-depth study? ------------------------------------------------------------------- I have used this library in production code that I have shipped to users, and have been working with it for several months. What is your evaluation of the potential usefulness of the library? ------------------------------------------------------------------- For my applications, it was crucial -- they would have been far less optimal (both from a software-engineering and asymptotic perspective) without ITL. The examples showcase a wider range of potential applications, especially the use of an example "Project" as a more complex example. Are you knowledgeable about the problem domain? ------------------------------------------------ As a consumer of many boost libraries for many years, from the simple (shared_tr) to more esoteric (variant) to completely crazy (fusion), I feel like I have a good sense of modern C++ API aesthetic. As someone who spends a great deal (too much) of his life working with time-series data structures, I am familiar with many of the underlying algorithms and data-structures. ...Eric Jonas Eric Jonas (jonas(a)mit.edu) PhD Candidate, MIT Brain and Cognitive Sciences

2 1

[math] long double autoconfiguration
by Vladimir Prus 24 Feb '10

24 Feb '10

Hi, at present, Boost.Math library always tries to compile long long versions of the function, unless explicit --disable-long-long is passed by the user. I have just checked in a patch that attempt to automatically determine if long long is available. Attached is a sample output of "bjam --with-math" build. The detection is done by compiling new file named libs/math/src/long_double_check.cpp -- if it compiles, it is assumed long double works. Unfortunately, I don't really know what kind of long double support Boost.Math needs. John, Paul, could either of you add some more content to that file? This change an SVN version of bjam as of today. If you use an older version, nothing should break, but no auto-detection will be performed. Also, if you need to look at compiler messages produced when compiling the test file, look at bin.v2/config.log Thanks, -- Vladimir Prus http://vladimir_prus.blogspot.com Boost.Build: http://boost.org/boost-build2 Building the Boost C++ Libraries. Configuration checks summary - long double support : yes Component configuration: - date_time : not building - filesystem : not building - graph : not building - graph_parallel : not building - iostreams : not building - math : building - mpi : not building - program_options : not building - python : not building - regex : not building - serialization : not building - signals : not building - system : not building - test : not building - thread : not building - wave : not building ...patience... ...found 1584 targets... The Boost C++ Libraries were successfully built! The following directory should be added to compiler include paths: /home/ghost/Work/Boost/boost-svn The following directory should be added to linker library paths: /home/ghost/Work/Boost/boost-svn/stage/lib

4 5

virtual function templates
by strasser＠uni-bremen.de 24 Feb '10

24 Feb '10

Hi, is there interest in a boost utility to implement (pseudo-) virtual function templates? real virtual function templates are obviously not possible without a JIT-compiler, but if all derived types are known to the base type, something like the following is. I've used a non-generic form of this a couple of times: struct base : enable_vtemplate<derived1,derived2>{ template<class OutputIterator> void f(OutputIterator it){ BOOST_VTEMPLATE(f,(OutputIterator),(it)); it << "not overridden\n"; } BOOST_SUPPORT_VTEMPLATE(); }; struct derived1 : base{ template<class OutputIterator> void f(OutputIterator it){ it << "derived1\n"; } BOOST_SUPPORT_VTEMPLATE(); }; struct derived2 : base{ template<class OutputIterator> void f(OutputIterator it){ it << "derived2\n"; } BOOST_SUPPORT_VTEMPLATE(); }; the runtime overhead besides the virtual call is one switch().

6 9

shortest path between vertex A and B
by Alexandre de Morais Amory 24 Feb '10

24 Feb '10

Hi, I am starting to use BGL. I manage to read/write in dot format, add few attribs, etc. my next task is to get the set of vertex in the shortest path between vertex A and B. In a graph like this: a - b b - c c - d assuming 'a' and 'c', the result would be {a,b,c}. I know it sounds silly, but most similar examples I could find are much more complex than this. I guess breadth_first_search would solve the problem, but I am having trouble with the visitor concept... how to implement a visitor to get the set of vertex in the path between two points. Other features of the problem: I am assuming an undirected graph. all edges have unitary distances. If there are multiple shortest paths between A and B, any of the valid solutions can be returned. thanks in advance for any clue ! Regards, Alexandre Amory

2 1

[thread] compile errors on Sandia-intel-10.1/11.0/11.1
by Emil Dotchevski 23 Feb '10

23 Feb '10

The error below is detected building a couple of Boost Exception tests that use Boost Thread. Thanks Noel and Steven for helping diagnose the problem. "/opt/intel/Compiler/11.1/067/bin/intel64/icpc" -xc++ -w1 -O0 -g -inline-level=0 -vec-report0 -DBOOST_ALL_NO_LIB=1 -DBOOST_THREAD_BUILD_LIB=1 -DBOOST_THREAD_POSIX -I".." -c -o "/Volumes/Scratch/kbelco/boost/results/boost/bin.v2/libs/thread/build/intel-darwin-11.1/debug/link-static/threading-multi/pthread/thread.o" "../libs/thread/src/pthread/thread.cpp" ../libs/thread/src/pthread/thread.cpp(385): error: identifier "get_nprocs" is undefined return get_nprocs(); ^ Emil Dotchevski Reverge Studios, Inc. http://www.revergestudios.com/reblog/index.php?n=ReCode

1 0