June 2010 - Boost - lists.boost.org

[gsoc][review/comments] Request community input on project interface
by Brian Bartman 22 Jun '10

22 Jun '10

Hello, I'm currently working on my GSoC project and I wanted some feedback from the community about the interface of a data structure I'm working on called a "bitfield_tuple". The bitfield_tuple makes it easy and simple to create and use specific bitfields. Currently, this works only over integral types, although there are plans to extend this in later versions of the data structure. bitfield_tuples correctly work with different types of endianness as well as provide the mechanisms to allow users to explicitly define the size of the structure's storage space. The bitfield_tuple is a tuple like sequence of elements but, instead of providing direct access to an instance of a type, a proxy to the element stored within the bitfield_tuple is returned. This is similar to std::vector<bool>'s reference type. Example bitfield_tuple declaration, struct red; struct green; struct blue; typedef bitfield_tuple< storage<unsigned int>, member<int,red,5>, member<int,green,6>, member<ing,blue,5> > bft_example; The internal storage type which all of the bitfields are stored in is specified by the storage template. The members are used to specify the different bitfields which are to be stored within that storage type. The member template is specified the exact same way a normal c++ bitfield would be specified inside of a struct or class. int red:5; member<int,red,5> A member template is specified in the order of: type, name, width of bitfield. The order the member templates are given in is the order in which they are stored within the bitfield_tuple. However, the storage template does not take up an index within the bitfield_tuple. So the template member<int,red,5> will be stored at index 0. Each member template requires a name be specified; this is for ease of access, later described in the get function. Each name must be unique within a single bitfield_tuple. The width of the bitfield is handled internally. The starting point for each bitfield is the bit after the previous bitfield, or 0 if it's the first bitfield. bitfield_tuple interface summary: Default constructor, copy constructor, and a constructor which allows for the construction over an initial value which is of the same as the storage type. bitfield_tuple(); bitfield_tuple (bitfield_tuple const&); bitfield_tuple (storage_type); Assignment operators for assignment from the storage type, similar to the constructor above, and copy assignment. bitfield_tuple const& operator=(storage_type); bitfield_tuple const& operator=(bitfield_tuple const&); The get function operates similar to the boost.tuple with slight modification. Specifically, the get function will return a reference type. There are two ways to use the get function. The first way is by using a unsigned integral constant as an index: bft_example temp; temp.get<0>(); // returns a proxy reference to the bitfield represented by member<int,red,5> The second way to use it is to call get using the name provided to the member parameter: temp.get<red>(); // does the same thing as get<0>(); I provide both const and non const versions of both functions. The bitfield_tuple is also a boost.fusion sequence. It models an associative array type by using the following category: struct bft_category : boost::fusion::random_access_ traversal_tag, boost::fusion::associative_tag { }; The btifield_tuple's fusion extension was created through the fusion sequence extension provided by boost.fusion. bitfield_tuple supports the following fusion functions and all of the intrinsic functions which are supported from the implementation of these functions. begin end is_sequence is_view category_of at size at_key tag_of The btifield_tuple_iterator supports the following fusion functions: deref value_of next prior distance key_of value_at_data deref_data advance If you would like to look at the code for bitfield_tuple, it is located here: http://svn.boost.org/svn/boost/sandbox/SOC/2010/bit_masks/boost/integer/bit… -- thanks, Brian Bartman

3 5

Re: [boost] [msm] Message queue and copy constructed events
by Christophe Henry 22 Jun '10

22 Jun '10

Hi Richard, >Lets separate 2 thinks. > >1: no heap operation this is a nice to have feature we can live with >it. According my understanding the only place where heap is used is in >the std:queue ? Am I correct ? If yes than if the user has a >possibility to provide its own queue implementation than he/she can >make sure about avoiding heap usage. In the queue AND in the boost::function objects contained in the queue. So merely changing the container will not be sufficient to eliminate all heap usage (read further on). >2: event copying. You saying copy is happening when bind is used to >anonymize events to be able to store them in a queue. >According my understanding of event driven systems we can say that as >log an RTC step is not finished -> means that the most outer >process_event does not retuned the event which was used in the call of >the process_event should be valid. This means that events will be >valid even with queuing for queues which are used during pseudo entry >and exit because the processing of those events are always finished >before the more outer process_event is returned. >The only place where this is not true ->when the event should be >available longer than the RTC step is when the event is deferred. >So I think it would be enough for non deferred queues to internally >before the bind store the event in a ref/cref and bind that and put >to the queue -> no copy. This is incorrect. You can have the same behaviour with the event queue. Consider test/SimpleWithFunctors.cpp: You have a transition: Row < Empty , cd_detected , Stopped , store_cd_info , ... > And in the action, we generate another event: struct store_cd_info { template <class EVT,class FSM,class SourceState,class TargetState> void operator()(EVT const&,FSM& fsm ,SourceState& ,TargetState& ) { fsm.process_event(play()); // temporary play will be queued, then the function returns // ooops, temporary is gone } /// end of action, transition now ends and will start examining the queue and will find inside a reference to a destroyed event }; Now, if msm was using ref/cref, we would be processing a destroyed event, which would not end well ;-) Unfortunately, it is not possible for process_event to know if the event object is a temporary, therefore the copy. The only way to change this would be to have the user tell msm about it and take responsibility for ensuring event validity. We could: - have an extra parameter to process_event (run-time cost for all users) - allow some compile-time flag (compile-time cost for all users) None is very appealing to me as I doubt many will require this, especially as I fail to see why you could not pimpl the event data, thus making the event cheap to copy. But everybody would have to pay for it somehow. And the heap issue would still be unsolved. This would likely also end at the very bottom of my feature list. I think a possible solution would be the fact that I intend to offer customization of the queue container, which is a much more popular feature. Instead of calling push on it with the boost::function object, I could call push(event), so that you can provide a container not of function objects but of anything (like of a pointers to some base event type or void* if you prefer to cast a little, which I prefer not knowing of ;-) ). This allows complete removal of any heap usage AND could also be used to get rid of event copying (in this case though, you are taking responsibility of having thought of the event objects' lifetime). The downside is that providing a new container will require slightly more work. >For deferred queues there is 2 ways to go. >First is to copy the event during bind or have a user implement a >queue which interacts whit the underlaying zero copy scheduler. and >only again store a ref/cref in a bind. maybe it is enough to have 2 >callback to the user when an event is deferred and when it is recalled >and not needed any more. Fore the deferred queue, I can provide the same mechanism, so that you have have different containers for both if you wish. Using boost.parameter will make these extra template arguments more bearable. Regards, Christophe

2 1

[GIL] integration between boost libraries (gil / multiarray)
by fabien.castan＠free.fr 22 Jun '10

22 Jun '10

Hi, I'm a big fan of gil. It's one of the best library I know. It has really improved my work, which consists of writing image processing algorithms. Thanks to the authors. I have only one criticism: the integration with other boost libraries and firstly with multiarray. Because manipulating an image consists of manipulating an array of pixels. Many gil's operations are not present in multiarray: flip_view, subsampled_view, subimage_view, n_channel_view, x_iterator, y_iterator, xy_locator, cache_locator, plannar or interleaved arrays, virtual views, static_for_each, ... For example, inside an algorithm I need to maintain a data structure for each pixel. So I would like to create an array of MyCustomStruct. I can use gil to do that but I need to have a notion of colorspace and operators between pixels, which have no sense in this particular case. But subsampled_view, cache_locator, etc. have a sense. Another less important thing, I find unfortunate that a structure like "point" is inside gil. Many operators are not defined. Some parts of my code are becoming dependant to gil just for a "point" class, or I need to work with one structure and make a conversion... Is there really nothing in boost for such thing ? Maybe just a typedef on a bounded_vector<2> could be sufficient ? Best regards, Fabien

3 6

[polygon] 1.44 release?
by Simonson, Lucanus J 22 Jun '10

22 Jun '10

I notice that Monday is the deadline for closing the 1.44 release to new libraries. Boost.Polygon is checked into the trunk and more or less passing tests on all the platforms I expect tests to pass. Is there any possibility of making it into the 1.44 release? Thanks, Luke

6 13

Re: [boost] [filesystem] wfstream file open does not create file in
by Ramakrishna, Kirankumar IN BLR SISL 22 Jun '10

22 Jun '10

Hi, I upgraded to boost version 1.43, but that did not help. Still file creation fails with wfstream and error code set is ERROR_FILE_NOT_FOUND. Below code captures the error codes #include "boost/filesystem/fstream.hpp" int main() { boost::filesystem::path pA("C:\\testA.txt"); boost::filesystem::wpath pW(L"C:\\testW.txt"); boost::filesystem::fstream fileHandleA; boost::filesystem::wfstream fileHandleW; fileHandleA.open(pA, std::ios_base::out); cout << "fileHandleA.fail() return value is: " << fileHandleA.fail() << endl; cout << "Error code is: " << GetLastError() << endl; fileHandleW.open(pW, std::ios_base::out); cout << "fileHandleW.fail() return value is: " << fileHandleW.fail() << endl; cout << "Error code is: " << GetLastError() << endl; return 0; } Am I doing something wrong? Regards, Kiran On 2010-06-19 00:44, Beman Dawes wrote: > On Thu, Jun 17, 2010 at 9:48 AM, Ramakrishna, Kirankumar IN BLR SISL > <Kiran.Ramakrishna@???> wrote: > > Hello, > > > > I am using boost version 1.42 and the below code is tested on Windows Xp SP3 and Suse 11.2 > > > > The below code creates new file when opened using fstream object, but fails to create when opened using wftream object. > > This behavior is consistent across platforms (Windows and Linux). > > Also the error code which I get after fileHandleW.open call is, > > > > > > Â 1. Â ERROR_FILE_NOT_FOUND for Â std::ios_base::out > > Â 2. Â ERROR_NEGATIVE_SEEK for std::ios_base::app > > > > In case of append mode, the file gets created. But still I get the above mentioned error code. > > > > Important notice: This e-mail and any attachment there to contains corporate proprietary information. If you have received it by mistake, please notify us immediately by reply e-mail and delete this e-mail and its attachments from your system. Thank You.

1 0

[website] Smaller favicon.ico file
by Jeremiah Willcock 21 Jun '10

21 Jun '10

Our current favicon.ico file (on www.boost.org) is very large (22k), and includes several icons that are too large for the standard 16x16 size for a favicon. Looking at some directions for compressing them, I have created the attached icon which is only 318 bytes and looks almost as good as the current icon. It might be worthwhile to consider replacing the current favicon with this file. It is also available as https://svn.boost.org/htdocs/site/favicon.ico (the favicon for svn.boost.org, but your browser is unlikely to use the new icon if you've downloaded the old one); Dong Inn and I put that in last week as part of some other performance optimizations in Trac. Purging old favicons from Firefox's cache is difficult, though, so you may just want to download it directly and view it in an image viewer. -- Jeremiah Willcock

2 1

Library design Q : overloads v/s default args
by Roshan 21 Jun '10

21 Jun '10

Hope this is not a silly question. The C++ Standard library does not use default arguments. Instead it provides an additional overload. Anybody know why ? For e.g. It provides the following two overloads for accumulate : template<class Itr, class T> accumulate(Itr first, Itr last, T init); template<class Itr, class T, class BinOp> accumulate(Itr first, Itr last, T init, BinOp op); instead of: template<class Itr, class T, class BinOp> accumulate(Itr first, Itr last, T init, BinOp op = std::plus<T>() ); I am wondering if there is any reason to avoid default arguments in general purpose libraries ? - Roshan

6 13

[uuid] Dependency on Boost.MPL & Boost.TypeTraits
by Kedar Bodas 21 Jun '10

21 Jun '10

Hello, The <boost/uuid/uuid.hpp> header currently has a dependency on <boost/mpl/bool.hpp> & <boost/type_traits/is_pod.hpp>. This is only used to declare the type_trait is_pod<uuids::uuid> AFAICT for the uuid_serialize.hpp header. No other part of the library seems to have any dependency on it. Here's my problem: By company policy I am not allowed to make any changes to a boost library header. The MPL & type_traits libraries are not currently authorized for use within our code. This means I can't use Boost.Uuid either. Is it possible for the above included headers to be enclosed within an inclusion guard like UUID_NO_TYPE_TRAITS_DEPENDENCY so that by defining this macro I can remove this dependency without having to change the <boost/uuid/uuid.hpp> header? Or better still since only uuid_serialize.hpp seems to need it this dependency might be moved there. Since we don't use Boost.Serialization in our code this will solve my problem. This will also serve the purpose of reducing the dependencies of the core part of the library and make it more modular. PS. I am not really sure if this belongs in this list or the boost-users mailing list. Thanks, -- Kedar Bodas, Think3 Designs Pvt. Ltd.

3 2

[trac] spam
by Steven Watanabe 21 Jun '10

21 Jun '10

AMDG ticket https://svn.boost.org/trac/boost/ticket/1556 just got spammed, and I notice that it has a bunch of earlier spam too. In Christ, Steven Watanabe

4 4

[boost[program options] locale support
by Christoph Heindl 21 Jun '10

21 Jun '10

hi, from the documentation [1] : "[...] the library uses the codecvt<wchar_t, char> locale facet from the global locale." The quote is taken from the unicode-documentation part. May i ask why the locale is taken from the global locale, instead of the locale assigned to the stream? What I'd like to do is (currently yields a logic_error) // assume stream contains "a = 1,2" // ... std::locale loc("German_Germany"); stream.imbue(loc); po::parse_config_file(stream, opts, false); instead i have to std::locale loc("German_Germany"); std::locale::global(loc); po::parse_config_file(stream, opts, false); to parse the config file correctly. Best regards, Christoph [1] "http://www.boost.org/doc/libs/1_43_0/doc/html/program_options/howto.html#id…

3 4