- Boost - lists.boost.org

[multi_index] Question about 'modify'
by Vladimir Prus 31 Aug '04

31 Aug '04

I have a question about the 'modify' method of ordered indexes in the multi_index library. I have an iterator obtained from an ordered index and would like to modify the pointed element. I'm pretty sure that the modification won't change the order, won't create collisions and so on. The documentation suggests to use either 'replace', or more efficient 'modify' method. Why can't I do the following: typedef variables_t::nth_index<1>::type index_t; index_t& index = variable_declarations.get<1>(); index_t::iterator existing = index.find(vd); const_cast<Variable_declaration&>(*existing) .make_external(); ? TIA, Volodya

2 3

user defined facets in date_time library
by Markus Schöpflin 31 Aug '04

31 Aug '04

While trying to make date_time tests compilable on tru64cxx65, I have come accross an issue with user defined facets (namely date_names_put and all_date_names_put) that leaves me scratching my head. Is a user defined facet required to be default constructible? Or does it need a constructor of the form "foo(size_t refs = 0) : locale::facet(refs) {}"? [lib.facets.examples] doesn't say so but most examples I could find either have a default constructor or a constructor taking a reference count. Markus

2 3

copyright question (bjam examples)
by John Maddock 31 Aug '04

31 Aug '04

Some of the examples in bjam v2 are actually taken verbatim from the QT libraries examples, and currently have no copyright or license declaration on them at all. Vladimir asks: "Will /* This file is part of Qt and is available under the Q Public License, available from http://doc.trolltech.com/3.3/license.html */ be enough to make inspect tool happy. If we're really going to be extra-picky, I don't think I can add any "Copyright" string to the file I don't own." Is adding such a string acceptable? Clearly we need to do something, but what? Thanks, John.

4 4

Something odd with the meta-comm tests?
by David Abrahams 31 Aug '04

31 Aug '04

It looks like some of them are being run against the mplbook branch of boost/mpl. How else to explain the lack of mpl/aux_/has_xxx.hpp reported in http://tinyurl.com/524ur? -- Dave Abrahams Boost Consulting http://www.boost-consulting.com

2 1

RE: [boost] Re: Review Wizard Status Report
by Glen Knowles 31 Aug '04

31 Aug '04

> From: David B. Held > David Abrahams wrote: > > > [...] > > What is the point of these IDs? It seems like it adds busywork for > > reviewers and no especially useful information. Now we're talking > > about making them more complicated... > > Maybe they have to do with a nagging feeling that Boost > Libraries don't all have a canonical name, and even when it > seems like they do, people tend to call them whatever they > feel like. Perhaps we should either go with a numerical > index (which the DBA in me prefers), or insist that all > libraries have a canonical name and that everyone use that > name to refer to it (perhaps to make it easier to filter > messages containing the name). There is also the case of a library having multiple reviews, the serialization library comes to mind. Glen Knowles Centor Software Corporation

2 1

Is boost::fsm suitable for embedded systems?
by Thomas Stevens 31 Aug '04

31 Aug '04

Hi, forgive me for my ignorance but I'm a boost newbie... The rationale document for boost::fsm states that the library should meet a number of requirements, numbered from 1 to 9. Especially, the sixth requirement says "support the development of arbitrarily large and complex state machines.". Ok, this may be good - I guess it is in fact good. But I'm a bit anxious that this implies that the library is therefore not suitable for enviroments where no unnecessary overhead is acceptable, like in many embedded applications. However, the seventh requirement indicates that my worries are unjustified since it reads: "allow the user to customize all resource management so that the library could be used for applications with hard real-time requirements". And further down the Rationale document one can read a section on the Resource usage of Memory which looks good. So, just to be sure, is boost::fsm suitable for the development of embedded applications where constraints such as memory footprint is an issue? (Hopefully, someone can answer this based on the experience of using the library or by knowing the code - not by reading the documentation (like me ;-)). Thanks, Thomas S.

4 7

VC8.0 and std::list
by Pavol Droba 31 Aug '04

31 Aug '04

Hi, I have encountered a problem with vc8.0 std library. I was able to fix it, but I'm not sure what is correct according to standard. The problem can be narrowed to one simple case. In vc8.0 method iterator std::list<>::erase(iterator First, iterator Last) returns a 'singular' iterator when the list is empty after erase operation. Because I'm dependant on the returned iterator to make insertion at this point later on, I had to make an explicit workaroud. I would expect it to return .end() iterator in this case. Is this behavior correct according to standard or not? If not, is it reasonable to submit bug report? Thanks, Pavol

3 3

Workflow, process flow, mapping of an activity diagram to code.
by Batov, Vladimir 31 Aug '04

31 Aug '04

In my spare time I've been fooling around with an implementation of an infrastructure that would allow workflow to be configured at run-time. A well-defined process flow is clearly easier to configure/implement at compile-time using the conventional means (coding). Say, bool my_process_flow() { if (call_a() == success/true) { return call_b(); } else { return call_c(); } } However, to be able to quickly adjust to changing customer requirements and/or procedures we might decide to provide a set of building blocks (like call_a, call_b, call_c) that could be assembled into a certain process flow at run-time. Maybe even give that ability to the customer themselves through, say, GUI interface. So, the infrastructure boils down to an Activity class that has an operation associated with it and can have two sub-activities to be called when the operation succeeds or fails. For example, // Create an activity with call_a() as an operation. Activity a(call_a); // Register call_b to be called when call_a succeeds. a.success() = call_b; // Regiter call_c to be called when call_a fails. a.failure() = call_b; call_a, call_b, call_c can be Activity instances themselves. So, we are building hierarchical tree-like structure to represent a certain process flow. We can easily support concurrent activities like a.success() = Concurrent(call_b)(call_c); Loops can be supported in a similar fashion (via pre-defined functiors). All mostly based on boost::function (a very handy gadget I must say. Kudos to the authors). Now after such a long and boring introduction to my actual question. :-) Is there any interest in having/developing such an infrastructure under the boost umbrella? Any input is welcomed even if you think it is a crappy idea. Best, Vladimir.

2 1

Boost License Permission
by Fernando Cacciola 31 Aug '04

31 Aug '04

I, Fernando Luis Cacciola Carballal (Fernando Cacciola for short) give permission for all my contributions to Boost to be moved to the Boost Software License Version 1 P.S.: Odd that I need this since I have my name added to blanket_permission.txt, but I recieved the email so just in case. Fernando Cacciola SciSoft

2 1

[format] patch for tru64cxx65
by Markus Schöpflin 31 Aug '04

31 Aug '04

Hello, attached is a patch which makes the format tests compile on tru64cxx65. It turns out that the compiler has troubles with the following code snippet: ---%<--- template <class X> class foo; namespace detail { template<class X> void bar (foo<X>& self); } template<class X> class foo { public: template<class Y> friend void detail::bar(foo<Y>&); }; --->%--- This gives the following error message: cxx: Error: /usr/users/schoepf/net/tmp/test_namelookup.cpp, line 13: declaration is incompatible with function template "void detail::bar(foo<X> &)" (declared at line 5) friend void detail::bar(foo<Y>&); ----------------------^ There are several possible workarounds I found, namely moving bar() out of the detail namespace, or making bar() a simple member of foo (not a member template), or just removing the friend declaration alltogether. I decided to use the latter workaround as it is already present in the format library for other compilers. Markus Index: format_class.hpp =================================================================== RCS file: /cvsroot/boost/boost/boost/format/format_class.hpp,v retrieving revision 1.21 diff -u -r1.21 format_class.hpp --- format_class.hpp 1 Feb 2004 14:57:09 -0000 1.21 +++ format_class.hpp 31 Aug 2004 09:11:02 -0000 @@ -83,7 +83,8 @@ #if !defined( BOOST_NO_MEMBER_TEMPLATE_FRIENDS ) \ && !BOOST_WORKAROUND(__BORLANDC__, <= 0x570) \ - && !BOOST_WORKAROUND( _CRAYC, != 0) + && !BOOST_WORKAROUND( _CRAYC, != 0) \ + && !BOOST_WORKAROUND(__DECCXX_VER, <= 60590041) // use friend templates and private members only if supported #ifndef BOOST_NO_TEMPLATE_STD_STREAM

1 0