- Boost - lists.boost.org

[unordered] 1.41.0 unordered_map doesn't allow declaration with incomplete type.
by Jon Spencer 27 Nov '09

27 Nov '09

The new version of the unordered library doesn't allow you to declare an unordered map with an incomplete type. The following code would compile with the old version but not the new: struct a; struct b { boost::unordered_map<int, a> c; }; Or the more useful: struct d { boost::unordered_map<int, d> e; }; As far as I know these constructs are allowed with the other standard containers. The problem show up in "boost/unordered/detail/fwd.hpp" with the the declaration of value_base, which requires that the value_type of the map not be incomplete at declaration time. Jon.

2 1

Re: [boost] Core libraries should separated from experimental libraries
by Tom Brinkman 27 Nov '09

27 Nov '09

Thomas Klimpel wrote >> Some of the accepted boost libraries seem to >> receive only minimal maintenance by their original authors, >> and making it easier to contribute >> libraries to boost would probably only increase the number of >> libraries in minimal maintenance mode. So I >> guess the recent efforts to encourage the boost >> community members to spend some maintenance time on >> libraries they haven't authored is exactly the >> right thing to do for the current situation. Agreed, that is a problem, but different from the one I'm trying to emphasize here. >> I'm not sure whether you want to highlight the excessive >> quality criteria for acceptance into boost here, >> or the lack of review managers. Probably more the lack of review managers. I wish more of the core boost library developers would step up and mentor new library submissions. Not sure why they don't. Its actually quite fun. >> But the sandbox is actually quite a nice place to flesh out a >> potential library. And there is quite a number of libraries in the >> sandbox that are currently in active flesh out mode, and will stay in >> that mode for quite some time. The sandbox is just a nice place to dump a library. Most are quickly abandoned and receive very little attention. Why? Exposure, they need exposure. Having a place in for the promising experimental libraries to to get some exposure, and where the authors can toot how great their library is would be fun for all involved. That's why I come to boost -- for an exchange of ideas and to be inspired by others. The sandbox does not do that for me. Having libraries in the experimental branch would focus everyone attention on the upcoming libraries. Nothing can focus a library's authors attention if he knows that his library is getting exposure and being noticed by others. I played a part in developing the the review queue, but it never changes. I don't like how its evolved and its stuck in the status quo. The review queue does not give enough exposure and its now taking about a year to get your library reviewed. Uggh. Actually, I would like to see that the review queue is abandoned and replaced with the experimental branch. The only important review would then be when to decide when the library has matured enough to move it from the non-stable to the stable branch of boost. The review wizards would manage this process. It would be much more fun for all involved. The current system is stale and not inspiring the level of innovation that I believe is possible here.

15 21

Re: [boost] Review Queue Needs Attention
by Tom Brinkman 26 Nov '09

26 Nov '09

>> judging by the review schedule (almost no review managers), that would mean >> boost isn't very interested in new libraries in general. It would seem so. >> I would recommend that authors try find a review manager before they request a review. Review managers are generally recruited from people that already have a library that has been approved in boost, that means you luke (hint). The wizards could open it up to anyone. Its been considered, but rejected in the past. I would like to focus attention on something different. That would be creating a "non-stable" branch of boost, where the proposed libraries would live for a while, get some exposure. Since most libraries are header file only, it should not be a problem. While not all will compile on all platforms, and certainly not have perfect documentation, it would be a place for future boost authors to elicit feedback. The experimental libraries would not affect the core boost libraries in any way. I'm absolutely convinced that it would encourage lots more participation. There is so much more to do, we haven't even scratched the surface of what is possible. The core boost authors have put in place a wonderful place to come and share ideas, but we are letting it flounder. Ugh.

7 6

Failures on SPARC64 [Was [1.41.0] Beta 1 available]
by Alexander Churanov 26 Nov '09

26 Nov '09

Folks, I've tried to re-built 1.41 and 1.39 using the suggested patch. The libraries build successfully, but a single test ( "spinlock_try_test" ) for smart_ptr fails. Please, examine the issue and provide with solution or workaround. I've created a ticket on that - https://svn.boost.org/trac/boost/ticket/3678 Alexander Churanov, maintainer of devel/boost-* for FreeBSD

1 0

[Review] GGL review starts today, November 5th
by Christophe Henry 26 Nov '09

26 Nov '09

I apologize in advance for not being able to bring a complete review due to my lack of knowledge of the problem domain, but I hope to bring a few elements. I also apologize in advance for the unfair treatment to Luke that I was not able to do the same about Polygon due to my lack of time at the point of the Polygon review. *What is your evaluation of the design? The design looks clear enough so that I was able to understand about what was being done. Adding concept checking is a good idea. It also surely is a modern generic design. I found only 2 points to criticize: - linestring has std::vector as basis class, and therefore it should be clearly documented that this class is not made to be derived from. - most or all of geometries classes have no virtual destructor and thus this should be clearly documented or changed. Better even would be to give the user choice through a policy. *What is your evaluation of the implementation? The code is clean and nice to read and I found it pretty easy to understand (apart from the deep geometry stuff of course). *What is your evaluation of the documentation? The examples are nice to read, the rest of the doc is a bit light and would probably not let me do more complicated stuff. I'd love an example about every feature. *What is your evaluation of the potential usefulness of the library? For me quite big. I'd probably never need the advanced features but the simple ones would solve my biggest problem, which is how to avoid ever having to implement something like this myself, which I would probably never manage anyway. About the completeness of the library, I am not able to judge it and trust the more experienced reviewers. *Did you try to use the library? With what compiler? Did you have any problems? I compiled several examples with VC9 + VC10 beta2. Both compiled without warnings level 3. The VC10 code analyser found a few points (reported to one the library writers), which I suppose are minor. Interestingly, the analyser did not find a bug in the linestring example, which happens only with VC10. This bug needs fixing but must be pretty simple to correct. *How much effort did you put into your evaluation? A glance? A quick reading? In-depth study? I spent a few hours reading and running the examples and checking the analyser and compiler level 4 warnings. I also spent 2 hours looking at the code. *Are you knowledgeable about the problem domain? Not at all. I'm simply looking for a library in case I ever need to do something in this domain. *Do you think the library should be accepted as a Boost library? Be sure to say this explicitly so that your other comments don't obscure your overall opinion. YES, eventhough I wouldn't count my own opinion as a full vote due to my lack of domain expertise. I like: - some nice features I understood like the linestring and polygon - the fact that the code is clean (meaning lots of care has been spent on it) and that the authors are responsive, which is a sign that the library will be well-maintained. - the examples I like less: - the documentation which gave me the frustrating feeling I missed most of the cool stuff and needs some warning messages for the problems I cited above. Best regards, Christophe Henry

5 7

GGL Review
by Gordon Woodhull 26 Nov '09

26 Nov '09

I believe GGL should be accepted as Boost.Geometry My main concern is the number of things which are incomplete. Certainly the obvious lackings such as missing boolean operations, clockwise versus counterclockwise polys, closed versus open polys, should be fixed or rationalized before this is released. It sounds like Barend and friends have promised to complete all of these things. I hope Hartmut is willing to manage a mini-review when the authors believe this is ready for release, to make sure that all concerns raised in the first review have been addressed in some way. I asked the same for Boost.Polygon; I think it is even more important here because the scope is much broader, whereas Polygon seems to have covered integer polygons pretty well. Then there are a lot of things which the authors claim are possible but which will probably need to be contributed by others: triangles, 3D, ellipses, splines, homogeneous coordinates ... I am confident that the authors will help these things to happen, and I have no reason to think that they won't fit in. OTOH it may be that some concepts do not fit in; for example, it's not clear to me whether or not the 45/90 degree concepts of Boost.Polygon are orthogonal or skew. It is going to take some time for everyone to figure this out, and I think that Boost is the right forum for this discussion. I hope that future domain libraries are developed as extensions to GGL, but it may be that they demand their own frameworks. I think it's naive to assume outright that all can fit into one framework, but I am optimistic. Both libraries allow external data structures and concepts to be adapted to them, and so the first step in determining their commonality will be to write those adapters. I'm pleased to see that Luke has already started this effort from his side. Oh, for concept maps! My two cents on floating point robustness: in many applications, the use of geometrical computations is "one-way": there is a model, some operations are applied to produce coordinates, which are then output. This is mostly true in the domain I wish to use this library, graph drawing (network visualization). If the results are not fed back into the system, floating point robustness is not going to be an issue. While some work could probably be done here, I think it is OK that users must understand the issues with floating point for best results. I'd like to see GGL run with Boost.Polygon's robustness tests and randomly generated efficiency tests. It sounds like this has already begun. A couple of people asked for GIL examples. This would require a line and curve bitmap rendering library, which would be a really cool addition to Boost. ===== How much effort did you put into your evaluation? A glance? A quick reading? In-depth study? ===== Sorry I have not had time to conduct a more thorough review. I have spent a couple of hours looking at the documentation and examples, and have read the review discussion thoroughly (another five or six hours). I've also followed most of the discussion in the past. ===== What is your evaluation of the design? ===== It seems very well thought out, for the domain which has been explored so far. The only concern I have (which I haven't thought out thoroughly) is the absence of hierarchical concepts. IIUC a new class must subscribe to all concepts, not just the most derived. This seems like it might get onerous when there are a lot more concepts / if the granularity is reduced. ===== What is your evaluation of the implementation? ===== From the discussion and from the presentation at BoostCon, I understand that not everything which is described is implemented, and everything which is implemented does not completely live up to the standards which the library sets. Again, I think these discrepancies should be resolved either by implementation or by narrowing the scope and documenting. As is, it's rather confusing. ===== What is your evaluation of the documentation? ===== I'd like to see an overview of the concepts - there are the detailed descriptions in doxygen but nothing written in English. :-) Doc seems mostly to consist of very technical rationales, which are good, and then examples, which are also good. But not much overview. I think doxygen is convenient for developers but not very useful for users, because of the high structure/content ratio. ===== What is your evaluation of the potential usefulness of the library? ===== Very useful. ===== Did you try to use the library? With what compiler? Did you have any problems ? ===== Not yet. ===== Are you knowledgeable about the problem domain? ===== As others have pointed out, it's not clear the domain is well- defined. I'm not familiar with GIS. I have developed visualization and graphics applications which need 2D geometry for more than a decade, and this looks like it will serve many of my needs. I wish both this review and the Polygon review had been delayed until the libraries were more mature. As it is, we're going on trust that the promise will be fulfilled. But I understand that there are social and real-world constraints as well as technical and I think it's best that both libraries are accepted a little early so that the community has a chance to cooperate in lifting the concepts. I am very happy to see the authors resolving some of their differences. There's a long road ahead... Cheers, Gordon

2 2

Re: [boost] [Serialization] warnings and maybe a 'bug'
by bernard cupic 26 Nov '09

26 Nov '09

Is it possible tha the problem is in the shared_ptr_helper.hpp file in the line 114: Â Â Â template<class T> Â Â Â struct non_polymorphic { Â Â Â Â Â Â Â static const boost::serialization::extended_type_info * Â Â Â Â Â Â Â get_object_identifier(T & t){ Â Â Â Â Â Â Â Â Â Â Â return boost::serialization::singleton< Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â BOOST_DEDUCED_TYPENAME Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â boost::serialization::type_info_implementation<T>::type Â ==>Â Â Â Â Â Â Â >::get_const_instance(); Â Â Â Â Â Â Â } Â Â Â }; If you compare it with the following code (same file, just below the upper code): Â Â Â template<class T> Â Â Â struct polymorphic { Â Â Â Â Â Â Â static const boost::serialization::extended_type_info * Â Â Â Â Â Â Â get_object_identifier(T & t){ Â Â Â Â Â Â Â Â Â Â Â return boost::serialization::singleton< Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â BOOST_DEDUCED_TYPENAME Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â boost::serialization::type_info_implementation<T>::type Â ==>Â Â Â Â Â Â Â >::get_const_instance().get_derived_extended_type_info(t); Â Â Â Â Â Â Â } Â Â Â }; it smells like a bug to me. Best regards. -- Bernard Cupic Software Development Manager USCS d.o.o. Flaciusova 1, HR-52100 Pula e-mail: [1]bernard.cupic(a)uscs.hr tel/fax: +385 52 380 687 References 1. mailto:bernard.cupic@uscs.hr

1 0

[xpressive] Significant compiler warning
by Ian Emmons 26 Nov '09

26 Nov '09

I have a program that uses Boost.Xpressive. When compiling this code on the Macintosh with the darwin-4.2.1 toolset, I get the following compiler warning: .../boost_1_41_0/boost/xpressive/traits/cpp_regex_traits.hpp:84: warning: left shift count >= width of type Here is the offending line of code: umaskex_t const highest_bit = 1 << (sizeof(umaskex_t) * CHAR_BIT - 1); The problem (if my thinking cap is screwed on correctly) is that the 1 that is being shifted is of type int, and is not converted to the target type of umaskex_t until after the shift operation. I believe that this should be changed to the following to avoid highest_bit being set to zero: umaskex_t const highest_bit = static_cast<umaskex_t>(1) << (sizeof(umaskex_t) * CHAR_BIT - 1);

2 1

[Exception] SFINAE use breaks Borland in Trunk
by John Maddock 25 Nov '09

25 Nov '09

Boost.Exception is currently failing for Borland compiles with for example: http://tinyurl.com/ycu779q The issue appears to be use of SFINAE like techniques without a BOOST_NO_SFINAE fallback. The problem here is that since Boost.Exception gets pulled in by almost everything, we need it to be lightweight and well supported, otherwise the rest of us are stuck even if we want to continue to support the older compilers :-( Thanks! John.

3 4

Re: [boost] [msm] scalability
by Christophe Henry 25 Nov '09

25 Nov '09

> a library that I plan to submit to boost causes > 50 s compile times even for simple examples, > so do you have any recommendations on where to start? There is a template profiler from Steven Watanabe, you could start there: http://www.boost-consulting.com/vault/index.php?action=downloadfile&filenam… > my library invokes a lot of Boost.Serialization and some MPL/fusion code, > so I'm not even sure if I can do anything about that within my library. As you use some MPL, you should check all your algorithms for complexity (especially if you use many mpl::fold). A quadratic complexity in a metaprogram will hurt you pretty fast. I had the feeling that SFINAE also hurts you quite much on some compilers. HTH, Christophe

3 2