October 2008 - Boost-users - lists.preview.boost.org

MPL: Arguments to transform metafunction
by Missing Rainbow 15 Oct '08

15 Oct '08

Hi All, I have a doubt regarding the arguments to the mpl::transform metafunction. In the MPL book (page 42), the mpl::transform is presented as follows: template <class Sequence1, class Sequence2, class BinaryOperation> struct transform; Here mpl::transform takes 2 sequences and a binary operation and it results in a sequence. But in a later portion of the book (page 124), while discussing about sequence building algorithms, the following is presented: typedef mpl::vector_c<int, 1, 2, 3> v123; typedef mpl::plus<_1, mpl::int_<5> > add5; typedef mpl::transform<v123, add5>::type v678; Here mpl::transform takes 1 sequence and a unary operation and it results in a sequence. How does mpl::transform supports such different arguments? Rgds, anna

2 1

boost::program options - why not allow_unregistered() for environmental variable parser?
by Ivan Godard 15 Oct '08

15 Oct '08

Vladimir Prus wrote: > > The environment seems to be the one place that is guaranteed to have > > unregistered options, because it is full of things (like > > $LD_LIBRARY_PATH) of no interest to the program doing the parsing. A > > call on parse_environment always throws on these. As far as I can see in > > the code, the way to avoid the throw is to pass parse_enviroment a > > function that checks for each of the options that are actually wanted > > and returns the null string for the rest. Of course, this duplicates > > what the parser just did. > > > > Is there a better way? > The recommended way is to use some name prefix for environment variables affecting your program -- say MY_PROG_. Then, pass this prefix to parse_environement and all variables without that prefix will be ignored. Does this help? - Volodya Unfortunately no. My code must interact with a large third-party application with a ton of environmental variables that follow no particular naming convention, and I wanted to integrate the probe for those variables with my own command line so I could accept the environment as default but overwrite them from the command line using the same names (which are already documented by the third party and mostly familiar to the users). Having no accept_unregistered precludes that approach, and I have to use getenv() and do it by hand instead. It's not a hard workaround, just a day or two of work, but lacking accept_unregistered for only one of the three kinds of parser seems odd given the general elegance of the rest of your library. Ivan

1 0

[program_options] maps as arguments
by Christoph Duelli 15 Oct '08

15 Oct '08

Does program_options somehow support something like --base x=hello --base y=world where the parsed "--base"s will be put in a std::map<std::string, std::string> (or std::map<X,Y> in the general case?) (At the moment I parse --base into a vector<string> and process the read strings after program_options has done its job.) Thank you, Christoph

2 1

[threads] How to test for running thread
by Koen Vermeer 15 Oct '08

15 Oct '08

Hi, Suppose I have a thread running, I issue an interrupt(), and the thread starts to perform its cleanup code. As this may take some time, I'd like to know when the thread actually is stopped. Is there some standard approach on how to do this? I was thinking about testing if the thread object equals Not-a-Thread, but I've yet to find out how... Thanks! Koen

2 4

how to expose an stl vector class porperty
by Massimiliano Bellomo 15 Oct '08

15 Oct '08

Dear all, i've a property of a class which is a std::vector<Foo*> that i would like to expose to python. Foo is a class already exposed. How do i implement the setter/getter methods for the vector property and declare them in the python module ? Many thanks, regards, Max -- --------------------------------------- Massimiliano Bellomo Dep. of Nuclear and Theoretical Physics INFN Pavia - +39 0382 987 234 ---------------------------------------

1 0

boost::program options - why no allow_unregistered() for environmental variable parser?
by Ivan Godard 15 Oct '08

15 Oct '08

The environment seems to be the one place that is guaranteed to have unregistered options, because it is full of things (like $LD_LIBRARY_PATH) of no interest to the program doing the parsing. A call on parse_environment always throws on these. As far as I can see in the code, the way to avoid the throw is to pass parse_enviroment a function that checks for each of the options that are actually wanted and returns the null string for the rest. Of course, this duplicates what the parser just did. Is there a better way? Ivan

2 1

[IoStreams]: bzip2_compressor not working?
by Roland Bock 14 Oct '08

14 Oct '08

Hi, does the attached program write something to stdout for anyone? It should print out the bzip2-compressed version of the string "Original". It compiles fine but prints out exactly nothing. My platform is Ubuntu 8.04 64bit, gcc 4.2.3, using boost version 1.36, and bzip2 1.0.4. Any ideas? Thanks in advance, Roland PS: The equivalent code works fine for gzip and zlib compression PPS: bzip2 itself works fine, too.

1 1

[graph]-Minor errors in docs
by Michael Fawcett 14 Oct '08

14 Oct '08

At the bottom of this doc page: http://www.boost.org/doc/libs/1_36_0/libs/graph/doc/graph_concepts.html#sec… There are numerous add_edge() calls that are incorrect, e.g. add_edge(digraph, u, v, Weight(1.2)); >From what I can tell, these should be: add_edge(u, v, Weight(1.2), digraph); Now, I'm only using 1.35, but the 1.36 documentation for add_edge here: http://www.boost.org/doc/libs/1_36_0/libs/graph/doc/adjacency_list.html supports my guess. Thanks, --Michael Fawcett

1 0

How to serialize class derived from standard container?
by Terdale, Shantibhushan 14 Oct '08

14 Oct '08

This could be very stupid question but I am stuck with this situation. I have class A derived from List<B> and B derived from C. How do I write Serialize method for A? See below, both options I tried are not working. Any help is appreciated. Thanks. class B : public C { private: friend class boost::serialization::access; template<class Archive> void serialize(Archive & ar, const unsigned int version) { ar & BOOST_SERIALIZATION_BASE_OBJECT_NVP(C); } }; class A : public list<B> { public: void Append( const list<string>& rBList ); A::const_iterator FindSpecialB(const B& rB); private: friend class boost::serialization::access; template<class Archive> void serialize(Archive & ar, const unsigned int version) { ar & BOOST_SERIALIZATION_BASE_OBJECT_NVP(List<B>); OR ar & boost::serialization::make_nvp("AddressList", boost::serialization::base_object<list<B>>(*this)); OR ?? } };

1 0

[thread_safe_signals][signals2] call for reviewers (review tentatively scheduled Nov 1st - Nov 10th)
by Stjepan Rajko 14 Oct '08

14 Oct '08

Hello all, The review for the Signals2 library (formerly known as thread_safe_signals - description below) submitted by Frank Mori Hess has been tentatively scheduled for Nov 1st - Nov 10th. As per a recent discussion on the review process (http://tinyurl.com/48tdjs) and after consulting with the review wizards, we would like to try something new for this review: have a number of reviewers commit beforehand to submitting a review. The goal of signing up is to ensure that there is both sufficient interest in the library, and that there will be enough reviews submitted for an informed decision on acceptance. If you are interested in this library, I encourage you to sign up now to be a reviewer (by this you commit to submitting a review preferably by the end of the review period). Of course, additional reviews from non-signed-up reviewers will also be welcome, but without enough reviewers signing up beforehand it is not clear that we should proceed with a review at this time. To sign up, please send me a reply (here or privately), and let me know whether the review period is suitable for you (we can make some adjustments if necessary). I strongly encourage participation from reviewers that would examine the library from a purely user standpoint (commenting on the interface and / or the documentation), as well as reviewers that would be willing to look into the details of the implementation (i.e., you don't have to focus on both). I you would like to sign up, please let me know by this Saturday (October 18th). About the library: -------- * The library can be downloaded from: http://www.boostpro.com/vault/index.php?&directory=thread_safe_signals (latest version is signals2-2008-10-08.zip) * Documentation is also available online: http://www.comedi.org/projects/signals2/libs/signals2/doc/html/index.html * A synopsis: The Boost.Signals2 library (formerly known as thread_safe_signals) is an implementation of a managed signals and slots system. Signals represent callbacks with multiple targets, and are also called publishers or events in similar systems. Signals are connected to some set of slots, which are callback receivers (also called event targets or subscribers), which are called when the signal is "emitted." Signals and slots are managed, in that signals and slots (or, more properly, objects that occur as part of the slots) can track connections and are capable of automatically disconnecting signal/slot connections when either is destroyed. This enables the user to make signal/slot connections without expending a great effort to manage the lifetimes of those connections with regard to the lifetimes of all objects involved. When signals are connected to multiple slots, there is a question regarding the relationship between the return values of the slots and the return value of the signals. Boost.Signals2 allows the user to specify the manner in which multiple return values are combined. * Relationship to Boost.Signals: This is a thread-safe variant of the original Boost.Signals library. There have been some changes to the interface to support thread-safety, mostly with respect to automatic connection management. The following thread offers some more details on the differences between the two implementations, as well as a plan of a phased replacement of Boost.Signals should Signals2 be accepted: http://tinyurl.com/4sqau3 [nabble] Again, if you are interested in this library, please consider signing up as a reviewer. Kind regards, Stjepan (review manager)

2 2