- Boost-users - lists.boost.org

[optional] new assignmet semantic and references
by Fernando Cacciola 01 Mar '05

01 Mar '05

{This is copy of a message on the developers list that needs attention from Boost.Optional Users} Hello people, A few days ago, a problem with Boot.Optional's assignment not properly handling aliasing scenarios (like self assignment) was pointed out. We concluded that the solution was to use T's operator=() directly instead of the Destroy+Construct pattern that I was using. I changed the code, tested it, and I was updating the documentation when I recalled the following long-standing issue: Currently, with the Destroy+Construct pattern, references are re-bound in case of assignment: int a = 123 ; int b = 456 ; int& ra = a ; int& rb = rb ; optional<int&> ora(ra); optional<int&> orb(rb); ora = orb ; // now 'ora' contains a reference to 'b' instead of 'a' b = 789 ; cout << *ora ; Prints 789 cout << a ; Prints 123 This is not the way assignment of reference types work in C++: ra = rb ; // The value of 'a' referenced by 'ra' is changed, not the reference itself. cout << a ; Prints 789 This error in the way optional<T&> treated assignment of reference types was pointed out by Joel de Guzman long ago when he started using Optional in Spirit. However, at the time, Joel explicitely asked me not to change this and keep the current rebinding semantics in order to minimize any impact in Spitit itself (which certainly has, o had, a much larger user base). If I change the code now to use T's operator=(), assignment of optional<T&> will have a _radically_ different semantics. A correct one, that's for sure, but end user code might be fundamentally impacted. Note: if you look at Optional documentation you will discover that assignment is only documented for T not being of reference type. This was intentional.. I left this undocumented becasue I knew back then that the current semantics were wrong but I couldn't fix it at the time because Spirit was (or could be) depending on it; yet I didn't want other users to depend on it as well... Of course, users could have "discovered" the (incorrect) semantics themselves and now rely on it, but I hope that the intentionally missing documentation served to keep the amount of users depending on this incorrect semantics very low. [At least it served _me_ to remember about this since I had forgotten] I really want to 'fix' this, i.e, give it correct semantics, so unless Joel or anyone else speaks now I'll move on, finish the documentation updates and commit the new version. Best, Fernando Cacciola P.S: Joel already agreed in the developers list that this needs to be fixed.

1 0

at<AssocSeq,Key,Default>
by Lemings, Eric B. (Contr) 01 Mar '05

01 Mar '05

Where is at<AssocSeq,Key,Default>' defined in MPL? I can't find the header file that defines it. Thanks, Eric.

3 2

date_time: UTC to local time conversion
by Cory Nelson 01 Mar '05

01 Mar '05

Is there a function in date_time to do UTC to local time conversions? -- Cory Nelson http://www.int64.org

2 2

re: [Boost-users] bjam seg faulting on mac os x 10.3.8
by junyi.sun＠supportlinux.net 28 Feb '05

28 Feb '05

I saw bjam got segmentation fault with bjam 3.1.10 + boost 1.32 + gcc 3.3.3 + fedora 2. After playing it around on different environment, I found that I can always reproduce the problem when I set environment variable $TOP for my own development CVS. Once I removed the environment variable $TOP ( I was using bash shell: run "unset TOP" ), the problem is gone. I hope this might help. - Junyi Sun From: boost-users-bounces(a)lists.boost.org [mailto:boost-users-bounces@lists.boost.org] On Behalf Of Andy Espenscheid Sent: Sunday, February 27, 2005 8:46 PM To: boost-users(a)lists.boost.org Subject: [Boost-users] bjam seg faulting on mac os x 10.3.8 I am trying to compile boost 1.32.0 on MacOSX 10.3.8 and getting a Segmentation Fault immediately from bjam immediately. I'm using gcc 3.3. bjam is crashing with a EXC_BAD_ADDRESS exception. I compiled bjam from the boost distribution but it has the same behavior. Any pointers would be greatly appreciated. Seems like I'm doing somehthing dumb. Also, are binaries available via fink? I seached with fink 0.71 but couldn't find anything. Thx.

1 0

bjam seg faulting on mac os x 10.3.8
by Andy Espenscheid 28 Feb '05

28 Feb '05

I am trying to compile boost 1.32.0 on MacOSX 10.3.8 and getting a Segmentation Fault immediately from bjam immediately. I'm using gcc 3.3. bjam is crashing with a EXC_BAD_ADDRESS exception. I compiled bjam from the boost distribution but it has the same behavior. Any pointers would be greatly appreciated. Seems like I'm doing somehthing dumb. Also, are binaries available via fink? I seached with fink 0.71 but couldn't find anything. Thx.

3 2

program-options: multi-token for a single value
by Austin Bingham 28 Feb '05

28 Feb '05

I'm wondering if program_options supports the notion of two command-line tokens forming a single value, and, if so, how. For example, one of the values I'd like to pass on via the command-line is `a position in lat-lon. For consistency with other programs my users know, I'd like to use something like this: --position 10:11:12:N 13:14:15:W During parsing, I'd like program_options to recognize that --position requires two associated tokens. I tried to use the multitoken() method, but this didn't seem to do what I wanted. The operator>> I wrote for parsing the tokens was still only being passed a single token. Either multitoken doesn't do what I expect, or I'm missing something. Any ideas? Thanks. Austin Bingham

2 6

RE: [Boost-users] [serialization] problem with Borland C++ Builder 6
by Richard Jennings 28 Feb '05

28 Feb '05

XML serialization is working fine for me with BCB6. With some of my programs I do get error E2015 reported when I do a make. If I do a build (i.e. rebuild all modules from scratch) then it compiles just fine. I haven't looked in to why this is yet. -----Original Message----- From: boost-users-bounces(a)lists.boost.org [mailto:boost-users-bounces@lists.boost.org] On Behalf Of Pablo Aguilar Sent: 26 February 2005 04:50 To: boost-users(a)lists.boost.org Subject: [Boost-users] [serialization] problem with Borland C++ Builder 6 Hi, I've just started using the serialization library, and am trying to do so with BC++6. If I'm right, it should work just as long as I remove support for xml archives (because Spirit doesn't work on BC++) I've a class Cliente which has an appropriate serialize member fxn. And a manager class called Clientes which stores a std::vector<boost::shared_ptr<Cliente> > When doing: void Clientes::serialize(Archive& ar, unsigned const version) { ar & m_imp->m_clientes; // where m_imp is a pimpl-style struct // m_clientes is of the type specified above } (this is trimmed down code, I can post the original if required) I get the following error: [C++ Error] serialization.hpp(121): E2015 Ambiguity between 'boost::serialization::void serialize< boost::archive::text_iarchive , boost::shared_ptr<cc::servidor::Cliente> > ( boost::archive::text_iarchive & , _STL::vector<boost::shared_ptr<cc::servidor::Cliente> > & , const unsigned int )' and 'boost::serialization::void serialize< boost::archive::text_iarchive , _STL::vector<boost::shared_ptr<cc::servidor::Cliente> > >( boost::archive::text_iarchive & , _STL::vector<boost::shared_ptr<cc::servidor::Cliente> > & , const unsigned int )' (I removed the allocator parts to make the message clearer) So the question is, how do I remove the ambiguity? Thanks for your help... Pablo Aguilar _______________________________________________ Boost-users mailing list Boost-users(a)lists.boost.org http://lists.boost.org/mailman/listinfo.cgi/boost-users

1 0

Handling "actions" with program_options
by Julio M. Merino Vidal 28 Feb '05

28 Feb '05

Hi everybody, I'm working on a program whose its actual syntax is: vcsme [-c conf_file] [-q] action [action_args] (very similar to the CVS one). I'm trying to convert it to use boost::program_options (to get rid of the getopt(3) call, and to have long options), but I don't know how to do it. So far, I've added the code to handle -c and -q options. This was very easy to do, given the examples on the web. The only annoyance is that the options are accepted everywhere, but I'd prefer to only accept them if they come before the action name (because the action itself might allow options). Unfortunately, I'm stuck with the latest part, the one to handle actions. There must always be an action passed to the program, which may optionally accept arguments. Some example invocations: vcsme -q update vcsme update gtk+ glib NetBSD But I can't see any way to do this with program_options. Using getopt, I do: argc -= optind; argv += optind; after processing the options, so that argv[0] becomes the name of the action, and any remaining entries are its arguments. The same can be achieved with getopt_long. So... is it there any way to do what I want with the actual implementation? Can't see anything relevant in the docs... Thanks, -- Julio M. Merino Vidal <jmmv84(a)gmail.com> http://www.livejournal.com/users/jmmv/ The NetBSD Project - http://www.NetBSD.org/

3 4

Re: [Boost-users] File-based container library?
by Beman Dawes 28 Feb '05

28 Feb '05

At 11:23 AM 2/23/2005, John.Wismar(a)autozone.com wrote: >Does Boost have a library that defines a file-based container, perhaps >based on a B-tree or equivalent, for very large collections? Or does >anyone know of such a thing, or has there been talk about developing one? > >We have a few file-based collections with millions of entries. We don't >especially want to install a database. The file is too large to load into >memory. The file's format could be changed to make it compatible with the >type of collection I'm thinking of, but for the moment, we're doing binary >searches by reading the file on disc. As you might guess, this can be >quite slow, especially when the file is on a DVD and not the HD. > >Anyway, I'd like to be able to write something like: > >//file.data contains 100,000,000 entries >file_map<int, int> some_data("file.data"); >file_map<int, int>::iterator it = some_data.find(56789012); > >and have the container (for example via a B-tree or whatever) be able to >locate the item with very few disc accesses. > >Any libraries that can do this type of thing? In 1999 I did some work on a B-tree implementation for Boost, but became discouraged. For maximum portability and very high performance, it would have had to have a lot of limitations; memcopy'able POD types only, an interface that differed in several ways from the standard library containers, etc. Even with those limitations, the prototype took twice as long for typical operations as the old C code B-tree I've used for 20 years. Perhaps someone else will give it a try; a B-tree is an incredibly valuable tool and it would be great to have a good one in Boost. --Beman

2 1

Problems with file_iterator and skip parsers
by chrono232＠fastmail.fm 28 Feb '05

28 Feb '05

Can anyone explain to me what is wrong with the following: ---------------- my_grammer g; // a grammar, the actual grammar I'm using is too long for me to post file_iterator first<char>( "somefile.txt") file_iterator last<char> = first.make_end(); rule_t<> skp = space_p | comment_p("//") | comment_p("/*","*/"); parse_info< file_iterator<char> > = parse( first, last, *g, skp ); ---------------- I keep getting these errors (from MinGW 3.4.2): 189 C:\Dev-Cpp\include\boost\spirit\core\non_terminal\impl\rule.ipp no matching function for call to `boost::spirit::impl::abstract_parser<do_main_thing()::scanner_t, boost::spirit::nil_t>::do_parse_virtual(const boost::spirit::scanner<boost::spirit::file_iterator<char, boost::spirit::fileiter_impl::mmap_file_iterator<char> >, boost::spirit::scanner_policies<boost::spirit::no_skipper_iteration_policy<boost::spirit::skip_parser_iteration_policy<boost::spirit::rule<do_main_thing()::scanner_t, boost::spirit::nil_t, boost::spirit::nil_t>, boost::spirit::iteration_policy> >, boost::spirit::match_policy, boost::spirit::action_policy> >&)' 61 C:\Dev-Cpp\include\boost\spirit\core\composite\intersection.hpp no matching function for call to `boost::spirit::skip_parser_iteration_policy<boost::spirit::rule<do_main_thing()::scanner_t, boost::spirit::nil_t, boost::spirit::nil_t>, boost::spirit::iteration_policy>::skip_parser_iteration_policy()' 61 C:\Dev-Cpp\include\boost\spirit\core\composite\intersection.hpp no matching function for call to `boost::spirit::skip_parser_iteration_policy<boost::spirit::rule<do_main_thing()::scanner_t, boost::spirit::nil_t, boost::spirit::nil_t>, boost::spirit::iteration_policy>::skip_parser_iteration_policy()' According to the FAQ, this shouldn't happen with grammars. I am a beginner, so please forgive my ignorance. Thanks for any and all help. Also, I've tried making the skip rule a grammar, again, with errors.

2 1