April 2009 - Boost-users - lists.boost.org

Ideas for something that looks like metaprogramming
by John Dlugosz 23 Apr '09

23 Apr '09

I'm revamping my atomic_counter template (see < http://www.dlugosz.com/Repertoire/refman/Classics/atomic_counter_whitepa per.html#toptop>) and plan on exposing all the available "magic" (e.g. things you can't do in C++) op-codes in the atomic read-modify-write department, as primitives. Then, use those in the template class. But, I have gained a deeper appreciation for some of the LOCKable instructions I never thought were interesting. For example, you can LOCK ADD, which changes the value in memory but doesn't return back what it was or became. What good is that? Directly translating to C++, none at all. But... you do get back some information. In assembly language, you see the flags. So, a thread can add something to the location, and although doesn't know what it became in general, at least know if this thread is the one that made it zero. Or, overflowed, or whatever. Those flags are the basis of if-statements in assembly language. So, given primitive calls that return the whole set of flags, it occurs to me to be able to harness that with a metaprogramming-like interface. For example, I'm hesitant to implement something called operator &= that doesn't return the new value, but "something else" instead. I think the only interesting flag here is whether the result is zero. So bool operator&=(left,right) returning whether the result was non-zero would be very confusing, even if it did work right when used in an 'if' condition! So, just brainstorming, x&=y would return something not directly usable, but does work with an overloaded ==, so if you were to write (x&=y)==0 the result would be what you expect from C++. Maybe it's not worth the effort. But if anyone wants to discuss it... --John (sorry about the footer; it's not my idea) TradeStation Group, Inc. is a publicly-traded holding company (NASDAQ GS: TRAD) of three operating subsidiaries, TradeStation Securities, Inc. (Member NYSE, FINRA, SIPC and NFA), TradeStation Technologies, Inc., a trading software and subscription company, and TradeStation Europe Limited, a United Kingdom, FSA-authorized introducing brokerage firm. None of these companies provides trading or investment advice, recommendations or endorsements of any kind. The information transmitted is intended only for the person or entity to which it is addressed and may contain confidential and/or privileged material. Any review, retransmission, dissemination or other use of, or taking of any action in reliance upon, this information by persons or entities other than the intended recipient is prohibited. If you received this in error, please contact the sender and delete the material from any computer.

2 1

[interprocess] Shared memory compile problem
by Po-Chun 23 Apr '09

23 Apr '09

I am a new user and try to write some codes with shared memory using boost.interprocess. Basically, I followed the examples and the allocation seems fine. But the use of shared-memory container gave me a bunch of error messages related to "instantiated from here." during compile. The error seems related to the template. But I have no idea how to fix it yet. Any suggestion would be helpful. Thanks. Po-Chun My code: struct LongIntLess { bool operator()(const uint64_t a, const uint64_t b) { return a < b; } }; main () { using namespace boost::interprocess; try{ shared_memory_object::remove("SharedMemory"); managed_shared_memory segment (create_only, "SharedMemory", sizeof (uint64_t)*200); typedef allocator< uint64_t, managed_shared_memory::segment_manager> shmAllocator_t; typedef set<uint64_t, shmAllocator_t> shrSet_t; const shmAllocator_t alloc_inst (segment.get_segment_manager()); shrSet_t *WrSet_1; WrSet_1 = segment.construct<shrSet_t> ("shrSet_t") ("LongIntLess", alloc_inst); // <= not sure about this // Is it right way to construct a set ? WrSet_1->insert(2); // <====== This line causes the error } catch (...){ shared_memory_object::remove("SharedMemory"); throw(); } shared_memory_object::remove("SharedMemory"); } The error messages are from these files: boost/interprocess/containers/ detail/tree.hpp: boost/intrusive/detail/tree_algorithms.hpp boost/intrusive/rbtree.hpp boost/interprocess/containers/detail/tree.hpp boost/interprocess/containers/set.hpp One of the error messges is like: tree.hpp:331: instantiated from 'bool boost::interprocess::detail::rbtree<Key, Value, KeyOfValue, KeyCompare, A>::key_node_compare<KeyValueCompare>::operator()(const KeyType&, const typename boost::interprocess::detail::node_alloc_holder<A, typename boost::interprocess::detail::intrusive_rbtree_type<A, boost::interprocess::detail::value_compare_impl<Key, Value, KeyCompare, KeyOfValue> >::type>::Node&) const [with KeyType = long unsigned int, KeyValueCompare = boost::interprocess::detail::value_compare_impl<long unsigned int, long unsigned int, boost::interprocess::allocator<long unsigned int, boost::interprocess::segment_manager<char, boost::interprocess::rbtree_best_fit<boost::interprocess::mutex_family, boost::interprocess::offset_ptr<void>, 0ul>, boost::interprocess::iset_index> >, boost::interprocess::detail::identity<long unsigned int> >, Key = long unsigned int, Value = long unsigned int, KeyOfValue = boost::interprocess::detail::identity<long unsigned int>, KeyCompare = boost::interprocess::allocator<long unsigned int, boost::interprocess::segment_manager<char, boost::interprocess::rbtree_best_fit<boost::interprocess::mutex_family, boost::interprocess::offset_ptr<void>, 0ul>, boost::interprocess::iset_index> >, A = std::allocator<long unsigned int>]'

2 3

using offset_ptr
by Rex Battenberg 23 Apr '09

23 Apr '09

Hi, I am having trouble using offset_ptr. I want to dynamically allocate memory in windows shared memory and create a point to it so it can be used by two different processes. I figures I would create a named object in the shaed memory to contain the pointer to the memmory, but when I try it I get a compile error. Here is the code I have so far. Note the the the shared memory is created elsewhere and it is big enough. bip::managed_windows_shared_memory* pAldisSharedMemory = new bip::managed_windows_shared_memory(bip::open_only, ALDIS_SHARED_MEMORY_NAME.c_str()); if (pAldisSharedMemory) { bip::offset_ptr<char> pImageData; // // This line causes the following error. Error 1 error C2440: 'initializing' : cannot convert from //' //boost::interprocess::offset_ptr<PointedType> *' to 'char *' //C:\stopbar\3rdParty\boost\include\boost\interprocess\offset_ptr.hpp 111 AldisHardware // pImageData = pAldisSharedMemory->find_or_construct<bip::offset_ptr<char>>("ImageData")(); // char * *pImageData = pAldisSharedMemory->find_or_construct<char *>("ImageData")(); if (pImageData == NULL) { // // This value will be unuseable by other processes, need to use an offset_ptr here, but I have not figured // out yet how to make them work. // pImageData = (char *)pAldisSharedMemory->allocate(1280 * 860 * 3); } I have really no idea what is wrong or how to do this. If anybody has any examples, that would be great. Thanks Rex

1 0

Re: [Boost-users] regular expression problems
by John Maddock 23 Apr '09

23 Apr '09

> Sorry, but this strays a little from actual boost. Since there are a > handful of you that are probably using regex, I was hoping someone could > help with syntax. > > I have a regex that validates a latitude format. > > > > These are the restrictions: > > Must have an N or S as first letter. > > It can be all values from 0.0 to 90.0 > > My first shot at it confirmed format, but not data constraints: > "(N|S)\\d{2}\\.\\d{1,6} <file:///\\d%7b2%7d\d%7b1,6%7d> " > > But any attempts to limit the constraints caused an exception in regex. > Is > there something wrong w/ how I formatted this: > "(N|S)(9[0]\\.0{1-6}|[0-8]\\d\\.\\d{1-6} > <file:///\\.0%7b1-6%7d|[0-8]\d\d%7b1-6%7d> )" > > > > To me, it reads: 1st char must be N|S > > If second char is 9, the next chars must be 0.0 through 0.000000 > > Else > > If char is 0-8, then the next just has to be digits [0-9].[0-9]{1-6} > > > > Can anyone tell me what I am doing wrong? There are no numeric constraints available in the regular expression language, and the {} operator are for constrained repeats with 1{1-6} not being a valid syntax. How about something like: (N|S)\\d(\\.\\d+|[1-8](\\.\\d+)?|9(\\.0+)?)? HTH, John.

3 2

[asio] Named pipes on Windows?
by Allen Cronce 23 Apr '09

23 Apr '09

Hi all, I'm working on a cross platform server daemon. Clients communicate with the daemon using boost asio. For the posix implementation, it's convenient and appropriate in our case to use Unix domain sockets. I've experimented with the related asio "local" sample code and verified that it will work for our purposes. For Windows it looks like named pipes are sort of the equivalent of Unix domain sockets. But while asio has a "stream_handle" interface that accepts a Windows handle, all of the sample and test code passes invalid handles. I haven't found any working sample or test code that specifically integrates named pipes with asio. All I've found so far is an email to this list on March 13th from Andrew Solodovnikov who was having problems getting named pipes to work with asio. Has anyone successfully used asio and Windows named pipes? If so, would you be willing to share some sample code that demonstrates the functionality? Thanks in advance. Best, -- Allen Cronce

2 1

[tuple][fusion] Creating tuple type from function type
by Jeremiah Willcock 23 Apr '09

23 Apr '09

I would like to take a given function type and create a (Fusion or Boost) tuple with its parameter types as members. This would be straightforward to do with function_types and Fusion, but I would also like to add a separate tag to each tuple type. For example, I would like to do the following: struct tag1 {}; struct tag2 {}; typedef tagged_function_tuple<tag1, void(int, double)>::type t1; typedef tagged_function_tuple<tag2, void(int, double)>::type t2; and have t1 and t2 be distinct types, each equivalent to fusion::vector<int, double> (or some kind of tuple<int, vector>). In particular, I would like to use the nice constructor syntax to allow things like t1(1, 2.5) to work, with as nice error messages as possible when incorrect argument counts or types are used. I would also like to "hide" the tags in the type, so just creating fusion::vector<tag1, int, double> would not be ideal. Is there a way to do this? Otherwise, I assume I can copy-and-paste the code for fusion::tuple and modify it to add a tag. Is that the recommended way? Thank you for your help. -- Jeremiah Willcock

1 0

Build and Install only a Specified Library from command line using bjam (newbie!)
by Agnel C J Kurian 23 Apr '09

23 Apr '09

Hi, I've been trying to build and install selected Boost libraries (regex, graph, program options) from the command line (MS VC9 Express) using the --with-library option of bjam. However I find that bjam copies headers for all boost libraries to the target location. How do I build and install only the specified library and it's headers? The same problem has been discussed by another user here: http://stackoverflow.com/questions/439402/building-a-subset-of-boost-in-win…

4 7

regular expression problems
by Green, Jason M NSWCDL, W33 23 Apr '09

23 Apr '09

Sorry, but this strays a little from actual boost. Since there are a handful of you that are probably using regex, I was hoping someone could help with syntax. I have a regex that validates a latitude format. These are the restrictions: Must have an N or S as first letter. It can be all values from 0.0 to 90.0 My first shot at it confirmed format, but not data constraints: "(N|S)\\d{2}\\.\\d{1,6} <file:///\\d%7b2%7d\d%7b1,6%7d> " But any attempts to limit the constraints caused an exception in regex. Is there something wrong w/ how I formatted this: "(N|S)(9[0]\\.0{1-6}|[0-8]\\d\\.\\d{1-6} <file:///\\.0%7b1-6%7d|[0-8]\d\d%7b1-6%7d> )" To me, it reads: 1st char must be N|S If second char is 9, the next chars must be 0.0 through 0.000000 Else If char is 0-8, then the next just has to be digits [0-9].[0-9]{1-6} Can anyone tell me what I am doing wrong? Just FYI, I am coding in VS 2005 C++ Thanks

1 0

Boost Thread with own Callback implementation
by programmine 23 Apr '09

23 Apr '09

I am using boost::thread for starting a member function of an object in a seperate thread. This works for an example like this: ProcessData *data = new ProcessData(); ProcessManager *pm = new ProcessManager(); boost::thread t1(&ProcessManager::start, boost::ref(pm), data); Unfortunately I am not using the ProcessManager directly but a Controller which uses a self implemented callback functionality. It looks like this: typedef Callback1 startProcessingEvent; startProcessingEvent::Signal onStartProcessing; The ProcessManager is binded to this event and I could call the method like myController->onStartProcessing(data); so in the end the ProcessManager->start is called through the Controller. If I know want to run the same thing in a separate thread I cannot just do this: boost::thread t1(&Controller::onStartProcessing, boost::ref(myController), data); because boost does not understand the whole callback thing. Does anyone has an idea how to solve the problem? Thanks -- View this message in context: http://www.nabble.com/Boost-Thread-with-own-Callback-implementation-tp23193… Sent from the Boost - Users mailing list archive at Nabble.com.

2 1

identity_property_map
by Andreas Fabri 22 Apr '09

22 Apr '09

Hello, Are the code and the documentation of the identity_property_map in sync. The documentation says that the class identity_property_map has a template member function operator[](T t) that for any type T, just returns t. In the .hpp file though I find //========================================================================= // A property map that applies the identity function to integers struct identity_property_map : public boost::put_get_helper<std::size_t, identity_property_map> { typedef std::size_t key_type; typedef std::size_t value_type; typedef std::size_t reference; typedef boost::readable_property_map_tag category; inline value_type operator[](const key_type& v) const { return v; } }; Best regards, andreas

2 1