- Boost - lists.boost.org

Volunteers needed: clearing the last regressions for 1.34
by John Maddock 20 Oct '06

20 Oct '06

Folks, We really should get 1.34 out sometime this month/year/decade/century (take your pick)! We're now down to a small but persistent hard core of regressions at http://engineering.meta-comm.com/boost-regression/CVS-RC_1_34_0/developer/i… My guess is that the respective library authors are either stuck (no access to the compiler concerned: please speak up if so), or else too busy or just not paying attention right now :-( The largest block is for the serialisation lib, at least one of the errors appears to be due to http://support.microsoft.com/kb/883655 which requires splitting the .cpp file into smaller blocks as a workaround. Robert, any ideas about the others? The mpl failures look like a problem in the test code (can someone with the relevant gcc-4.x installed try removing the typecast from the problem expression and see if that fixes things). The iostream and rational lib failures are the subject of other messages. All the others need volunteers to take a look and either suggest fixes, or at least try and diagnose the issue. Many thanks! John Maddock.

7 15

Web page broken again
by John Maddock 20 Oct '06

20 Oct '06

Our web page appears to be broken again, I'm seeing: An error has been encountered in accessing this page. 1. Server: www.boost.org 2. URL path: / 3. Error notes: File does not exist: /home/groups/w/ww/www/htdocs/ 4. Error type: 404 5. Request method: GET 6. Request query string: 7. Time: 2006-10-19 02:40:00 PDT (1161250800) The website is intact on the server, but how come we're looking in /home/groups/w/ww/www/htdocs/ and not /home/groups/b/bo/boost/htdocs ? John.

6 8

Documentation: Automatic redirection. How?
by Roland Schwarz 20 Oct '06

20 Oct '06

In the doc/html directory there are some stub files for docs. Can anyone please tell me about their intended use? When is this documentation being built? Which branch? Also it happened that: http://www.boost.org/regression-logs/cs-win32_metacomm/doc/html/threads.html does not work. My explorer ends up in some infinite attempts to constantly reload this page because the content is also a stub reloading itself ??? I am lost... Roland

1 1

[ConceptGCC] for Douglas Gregor
by Andy Little 20 Oct '06

20 Oct '06

Hi Doulas, I sent this privately but have a feeling it may have got sopammed out. If you got it, apologies for abusing the boost list. Hi Douglas, Here is a sample test, but IMO the concept in main should fail, else it is not possible to replace a traits class, hence I have had to use one in the following: regards Andy Little //-----------------------------//line1 //defines result types for binary ops and for static types with no runtime functions #include <quan/meta/binary_operation.hpp> namespace quan {namespace meta{ // is a type mpl::void_? // (binary_operation returns mpl::void for invalid ops, hence the need to query ) namespace detail{ auto concept NotMplVoid<typename T> { typename type = T; }; concept_map NotMplVoid<boost::mpl::void_> { typedef void type; }; } auto concept IsNotMplVoid< typename T> { where std::SameType<detail::NotMplVoid<T>::type,T>; }; auto concept OperatorPlus<typename T1, typename T2>{ typename type = binary_operation<T1, plus, T2>::type; where IsNotMplVoid<type>; }; // should I need to do this ?... see below template <typename T1, typename T2> where OperatorPlus<T1,T2> struct operator_plus{ typedef OperatorPlus<T1,T2>::type type; }; }} #include <quan/meta/rational.hpp> int main() { using quan::meta::rational; // succeeds but shouldnt AFAICS // as this is invalid op and type is mpl::void_ here typedef quan::meta::OperatorPlus<rational<1>,int >::type res; // line 45 // fails as expected but why does the Concept version above succeed? typedef quan::meta::operator_plus<rational<1>,int >::type res; // line 49 } ConceptGCC output: concept_test.cpp: In function 'int main()': concept_test.cpp:49: error: template arguments for 'struct quan::meta::operator_plus<quan::meta::rational<1, 1>, int>' do not meet the requirements of the primary template concept_test.cpp:49: note: no concept map for requirement 'quan::meta::OperatorPlus<quan::meta::rational<1, 1>, int>' concept_test.cpp:49: error: 'type' in class 'quan::meta::operator_plus<quan::meta::rational<1, 1>, int>' does not name a type ConceptGCC --version: conceptg++ (GCC) 4.1.1 (Indiana University ConceptGCC alpha 3) Copyright (C) 2006 Free Software Foundation, Inc. Copyright (C) 2005, 2006 Trustees of Indiana University. This is free software; see the source for copying conditions. There is NO warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. regards Andy Little

1 0

[rational][operators] VC6/7 supported? (Trying to clear regressions)
by John Maddock 20 Oct '06

20 Oct '06

What's the state of support for VC6/7 in Boost.Operators? I'm asking because I started investigating the remaining regressions for Boost.Rational and I've come up against a road-block: I can get everything to work (at least I think so), but not the comparison operators, so: my_rational_1 != my_rational_2 doesn't compile with VC6/7, but does with everything else. I'm assuming that's because these compilers don't find the necessary friend-functions in the operators base classes, or is there something else going on here? The error message is: rational_test.cpp(261) : error C2678: binary '!=' : no operator defined which takes a left-hand operand of type 'class boost::rational<short>' (or there is no acceptable conversion) ../../boost/test/test_case_template.hpp(73) : while compiling class-template member function 'void __thiscall basic_rational_suite::rational_comparison_test<short>::test_method(void)' BTW is the use of Boost.Operators by the rational lib new to this version? If so we might just have to mark up VC6/7 as unsupported. Thanks, John.

5 16

[array] Construct a variable-length array with "alloca"
by XHChen＠winbond.com 20 Oct '06

20 Oct '06

Dear boost interesters: "alloca" maybe a suitable tool to construct a variable-length array. The attachment demonstrates the class for variable-length array, most of the codes in which are copied from original "array.hpp" in boost libraries. Hope it is useful. Regards, xhchen =========================================================================================== The privileged confidential information contained in this email is intended for use only by the addressees as indicated by the original sender of this email. If you are not the addressee indicated in this email or are not responsible for delivery of the email to such a person, please kindly reply to the sender indicating this fact and delete all copies of it from your computer and network server immediately. Your cooperation is highly appreciated. It is advised that any unauthorized use of confidential information of Winbond is strictly prohibited; and any information in this email irrelevant to the official business of Winbond shall be deemed as neither given nor endorsed by Winbond.

3 3

Re: [boost] Runtime Dynamic Dispatch (boost-dispatch)
by Sohail Somani 20 Oct '06

20 Oct '06

> From: boost-bounces(a)lists.boost.org > [mailto:boost-bounces@lists.boost.org] On Behalf Of Dean > Michael Berris > Subject: Re: [boost] Runtime Dynamic Dispatch (boost-dispatch) > > Hi Sohail, > > On 10/20/06, Sohail Somani <s.somani(a)fincad.com> wrote: > > > > > > Can you elaborate on what you mean so that I don't > > > misunderstand what you mean? > > > > What I mean is overloading operator() with the types of the > tuple so you > > can call the dispatcher like a function. > > > > Does that clarify? > > A bit, though I'm still not so sure I understand. I'l' try to explain > why the dispatcher doesn't have that feature: [snip] I think you hit it on the head. Thanks!

1 0

Re: [boost] Runtime Dynamic Dispatch (boost-dispatch)
by Sohail Somani 20 Oct '06

20 Oct '06

> -----Original Message----- > From: boost-bounces(a)lists.boost.org > [mailto:boost-bounces@lists.boost.org] On Behalf Of Dean > Michael Berris > Sent: Thursday, October 19, 2006 10:46 AM > To: boost(a)lists.boost.org > Subject: Re: [boost] Runtime Dynamic Dispatch (boost-dispatch) > > Hi Sohail, > > On 10/20/06, Sohail Somani <s.somani(a)fincad.com> wrote: > > > > Is it not possible at all to have a d(1,2,3) interface? > > > > With the tuples, you are specifying the interface already. > > > > I'm not sure what you mean by `d(1, 2, 3)'... If you have > something like: > > void function(int a, int b, int c) { > // do something with a, b, and c > }; > > dispatcher<void(int, int, int)> d; > d[0] = &function; > d[0](1, 2, 3); > > Then definitely that's possible. > > > It would be more natural. > > Can you elaborate on what you mean so that I don't > misunderstand what you mean? What I mean is overloading operator() with the types of the tuple so you can call the dispatcher like a function. Does that clarify?

2 1

Re: [boost] GIL Image Processing Library - Undefined Behavior
by Ullrich Koethe 20 Oct '06

20 Oct '06

Hi again, since there was no response to my earlier questions about undefined behavior, I'd like to bring this up again: GIL makes extensive use of reinterpret_cast and gil_reinterpret_cast. The former is, for example, used in code similar to the following: short * imagedata = new short[3*image_size]; rgb16s_t * rgbdata = (rgb16s_t *)imagedata; gil_reinterpret_cast is defined as template <class OutPtr, class In> OutPtr gil_reinterpret_cast(In * p) { return static_cast<OutPtr>(static_cast<void *>(p)); } I suspect that undefined behavior results in both cases. IIRC, there is no guarantee that the memory layout of an rgb array is equal to that of a corresponding scalar array with three times as many elements (although this will probably work in practice). Likewise, a static_cast from void * is only allowed back to the original pointer type, not to a different type. Can someone more familiar with the fine points of C++ confirm these suspicions? If they are true, I think the issue should be resolved before inclusion into boost. Regards Ulli -- ________________________________________________________________ | | | Ullrich Koethe Universitaet Hamburg / University of Hamburg | | FB Informatik / Dept. of Informatics | | AB Kognitive Systeme / Cognitive Systems Group | | | | Phone: +49 (0)40 42883-2573 Vogt-Koelln-Str. 30 | | Fax: +49 (0)40 42883-2572 D - 22527 Hamburg | | Email: u.koethe(a)computer.org Germany | | koethe(a)informatik.uni-hamburg.de | | WWW: http://kogs-www.informatik.uni-hamburg.de/~koethe/ | |________________________________________________________________|

2 1

GIL Review: example files coming...
by Lubomir Bourdev 20 Oct '06

20 Oct '06

Since several reviewers requested this, we are working on splitting the GIL examples as separate unit example files and consolidating the ones we made due to people's requests. We will provide an "example" directory that compiles the files we have discussed so far: - convolution - affine_transform - packed_pixel (defining 565 RGB pixel) - histogram - x_gradient - mandelbrot (defining a synthetic view) - interleaved_ptr (how to construct your own pixel pointer and reference) (if we get to these:) - vigra_integration - agg_integration (from Joel) We are also going to provide the regression tests (which you can get today if you install ASL) Hopefully these (or most of them) will be available sometime tomorrow, together with a Makefile and jpeg-only dependency. (You can still access all of the above but it is inconvenient) Lubomir

1 0