August 2004 - Boost - lists.preview.boost.org

link in python docs
by John Maddock 18 Aug '04

18 Aug '04

Dave, In libs/python/doc/tutorial/index.html there is: <link rel="next" href="quickstart.html"> But the file quickstart.html does not exist - and this is fooling bcp, I assume that it should point to doc/quickstart.html ? Whatever it should be can you fix? Thanks, John.

3 2

Boost Software License Progress As Of 20040817
by Stephan T. Lavavej 18 Aug '04

18 Aug '04

http://stl.nuwen.net/bsl3.html (now with serialization, though not python) * 5037 BSLed files. * 39 files converted which have to be manually verified. * 1 file that could be converted, but hasn't yet been. That's boost/variant/detail/move.hpp. It's dual-copyrighted and dual-licensed, but both Eric Friedman and Andrei Alexandrescu are now in blanket-permission.txt. The file should be converted. * 619 files that cannot be converted yet. * 83 files without a recognizable license (including files from format, lambda, pending, any, graph, iterator, mpl, property_map, and utility). * 248 files without a recognizable copyright holder (all mpl, except for one file from serialization, where bcp is getting confused - that's got a copyright and a variant of the BSL). Still need many authors for the BSL. [David Abrahams] > What do you mean? What was broken? python and serialization used to cause bcp to fail. serialization has since been fixed, but something is still wrong with python: **** exception(205): std::runtime_error: Bad file name: libs\python\doc\tutorial\quickstart.html ******** errors detected; see standard output for details ******** [John Maddock] > Done, I've added the Aliases to blanket-permission.txt, and > fixed up the regex scanning to pick up on Robert Ramey's web > address. Cool! > I use "." as the module name, of course then it flags up a > whole load of new problems :-( I tried what you suggest, but it didn't work: C:\Temp\boost\boost>..\bcp --report . bsl4.html **** exception(205): std::exception: boost::filesystem::is_directory: ":sys:stat.h": The parameter is incorrect. ******** errors detected; see standard output for details ******** If you know how to scan everything, perhaps you could post the results? The recent flurry of additions to blanket-permission.txt is /awesome/. Thanks, Stephan T. Lavavej [I don't speak for Microsoft]

4 5

Fw: Boost
by John Maddock 18 Aug '04

18 Aug '04

Another permission: ----- Original Message ----- From: "Maarten" <mkeijzer(a)xs4all.nl> To: <boost-owner(a)lists.boost.org> Sent: Wednesday, August 18, 2004 8:30 PM Subject: Boost > License Permission > Date: Wed, 18 Aug 2004 21:30:43 +0200 > User-Agent: KMail/1.6.2 > MIME-Version: 1.0 > Content-Disposition: inline > Content-Type: text/plain; > charset="us-ascii" > Content-Transfer-Encoding: 7bit > Message-Id: <200408182130.43251.mkeijzer(a)xs4all.nl> > Status: R > X-Status: NQ > X-KMail-EncryptionState: > X-KMail-SignatureState: > X-KMail-MDN-Sent: > > I Maarten Keijzer, give > permission for all my contributions to Boost to be moved to the > Boost Software License Version 1 > > I think this involves a single file in the detail section of the > interval library. FYI. > > -Maarten- >

1 0

Re: lightweight replacement for boost::function
by Max Khesin 18 Aug '04

18 Aug '04

John, just in case you missed this there is a similar effort happening here (from a competing GUI library!): http://tinytl.sourceforge.net/ This email and any files transmitted with it are confidential and intended solely for the use of the individual or entity to whom they are addressed. If you have received this email in error please notify the system manager. This message contains confidential information and is intended only for the individual named. If you are not the named addressee you should not disseminate, distribute or copy this e-mail.

2 1

python/embedding fails under Intel 8
by Stefan Slapeta 18 Aug '04

18 Aug '04

Maybe somebody who is more experienced with boost.python than me could have a look at this: http://tinyurl.com/6p8zf. Thanks, Stefan

3 7

RE: [boost] Re: ETI bug and iterators
by Yitzhak Sapir 18 Aug '04

18 Aug '04

> -----Original Message----- > From: David Abrahams > > "Yitzhak Sapir" writes: > > > I can't send my own code, but the following code causes the > > previously described error with an unpatched minimum_category.hpp > > and an ICE when compiled with patched and unpatched > > minimum_category.hpp: > > Not for me. Of course I had to remove the include of stdafx.h since > I don't have that, but the first error is: > > Dave Abrahams The problem is with debug STLPort iterators, not the pointer iterators that you get in non-debug iterators mode. Yes, release compiles fine for me with the additional BOOST_TT_BROKEN_COMPILER_SPEC. But I also want to be able to compile in debug, with the debug STLPort iterators. :-) Also, when I speak of ETI workarounds removed, I speak only of the #ifdefed lines in the following block: typedef minimum_category_impl< # if BOOST_WORKAROUND(BOOST_MSVC, == 1200) // ETI workaround is_same<T2,int>::value || # endif ::boost::is_convertible<T1,T2>::value , ::boost::is_convertible<T2,T1>::value # if BOOST_WORKAROUND(BOOST_MSVC, == 1200) // ETI workaround || is_same<T1,int>::value # endif > outer; I didn't notice there was another ETI workaround before.

2 1

algorithm/minmax_element regression with Comeau
by Bronek Kozicki 18 Aug '04

18 Aug '04

Following error http://tinyurl.com/5t7cj is due to too strict requirements of libcomo (at least I believe so). I received fix from Comeau Computing (stl_algo.h, too large to send as attachment with this message) and deployed it, tommorow we shall see this error gone. Again, I deployed it without stopping regression tests, thus next results in como-win32-4.3.3-vc7.1 may be screwed. B.

2 1

Suggestion for weak_ptr's
by Jeff Williams 18 Aug '04

18 Aug '04

Simply small change, curious as to what others think: I'll illustrate with a trivial example: struct foo { virtual ~foo() { } }; struct bar : foo { }; typedef shared_ptr<foo> fooptr; typedef shared_ptr<bar> barptr; void some_func(weak_ptr<foo> wf) { barptr pb = wf.lock<bar>(); } // currently the way to do this is (as I am sure everyone knows) void some_func(weak_ptr<foo> wf) { barptr pb = dynamic_pointer_cast<bar>(wf.lock()); } My suggestion seems easier to read. Would require no additional overhead and would preserve existing behavior. I suppose you could provide a static and dynamic version: lock_static<bar> and lock_dynamic<bar> Any thoughts? Or is the general concensus that ideas like this are something indivudual people should do in their own source? Jeff

3 2

Intel: different macros for Windows and Linux
by Stefan Slapeta 18 Aug '04

18 Aug '04

I would like to introduce different config macros for the windows and linux version of ICL which should be defined like BOOST_INTEL. The names should be BOOST_INTEL_WIN and BOOST_INTEL_LINUX. It's just to improve the control over the completely different front ends of these compilers.... (actually they are two different compilers. In many cases it doesn't make sense to make behaviour dependent on "the" Intel compiler!) OK to commit? Stefan --- intel.hpp 29 Jun 2004 10:19:27 -0000 1.36 +++ intel.hpp 18 Aug 2004 10:00:24 -0000 @@ -29,6 +29,12 @@ #define BOOST_COMPILER "Intel C++ version " BOOST_STRINGIZE(BOOST_INTEL_CXX_VERSION) #define BOOST_INTEL BOOST_INTEL_CXX_VERSION +#ifdef _WIN32 +# define BOOST_INTEL_WIN BOOST_INTEL +#else +# define BOOST_INTEL_LINUX BOOST_INTEL +#endif + #if (BOOST_INTEL_CXX_VERSION <= 500) && defined(_MSC_VER) # define BOOST_NO_EXPLICIT_FUNCTION_TEMPLATE_ARGUMENTS # define BOOST_NO_TEMPLATE_TEMPLATES

2 3

[boost-sandbox] deleting old stuff
by Thorsten Ottosen 18 Aug '04

18 Aug '04

Hi all, Does anybody know a quick way fo deleting old libaries in the sandbox? Thanks Thorsten

2 2