- Boost - lists.boost.org

[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

New option for regression.py
by Markus Schöpflin 31 Aug '04

31 Aug '04

Hello. I have done a patch for regression.py which adds a new option '--test-options'. I needed this to pass options to the actual test run. For example: python regression.py --toolset=tru64cxx65 --test-options='-sPYTHON_VERSION="2.3" -sPYTHON_ROOT="/opt" -sPYTHON_INCLUDES="/opt/include/python2.3 /opt/exec/OSF1-V5.1-alpha/include/python2.3"' Any interest in this? Markus

6 11

[admin] Mail archive looks 'stuck'
by Jeff Garland 31 Aug '04

31 Aug '04

Not sure who's in charge of this, but the archive page looks as though it's recording any longer. Last recorded email was from this morning: http://lists.boost.org/MailArchives/boost/bydate.php Jeff

1 0

Re: Re: Re: Review of Daryle Walker's"MoreIO"librarybeginstoday, August 21, 2004.
by JOAQUIN LOPEZ MU?Z 30 Aug '04

30 Aug '04

----- Mensaje original ----- De: Jonathan Turkanis <technews(a)kangaroologic.com> Fecha: Sábado, Agosto 28, 2004 9:27 pm Asunto: [boost] Re: Re: Re: Review of Daryle Walker's"MoreIO"librarybeginstoday, August 21, 2004. > > "Pavel Vozenilek" <pavel_vozenilek(a)hotmail.com> wrote in message > news:cgpgmb$gla$1@sea.gmane.org... > > > > "Jonathan Turkanis" wrote: > > > > > Yes, I think skip_line/skip_lines is a good combination. > > > > > > I like newline and newlines. > > > > > It is often recomended not to distinguish names only > > on plural suffix. Easy to overlook bug here. > > I know that's a consideration. But then there's tuple and tuples... > Boost.Tuple's authors would have used "tuple" alone everywhere but they found difficulties, as explained at: http://boost.org/libs/tuple/doc/design_decisions_rationale.html section "For those who are really interested in namespaces". Joaquín M López Muñoz Telefónica, Investigación y Desarrollo

2 1

RE: [boost] Copyright and License in XML documentation
by Reece Dunn 30 Aug '04

30 Aug '04

Jeff Garland wrote: >I found a few moments to work on cleaning up licensing issues and decided >to >go make sure all the date_time xml docs have licensing info. The problem >is, >how should the copyright be added to all the files? I hunted around in >other >libraries using boostbook and it looks like they all lack copyright and >license info in all but the top level xml file. I notice that in the >library >main file we correctly notice the following construct in the inspection: > ><libraryinfo> > <author> > <firstname>Jeff</firstname> > <surname>Garland</surname> > </author> > > <copyright> > <year>2001</year> > <year>2002</year> > <year>2003</year> > <year>2004</year> > <holder>CrystalClear Software, Inc</holder> > </copyright> > > <legalnotice> > <para>Subject to the Boost Software License, Version 1.0. (See >accompanying file > <filename>LICENSE_1_0.txt</filename> or copy at <ulink > >url="http://www.boost.org/LICENSE_1_0.txt">http://www.boost.org/LICENSE_1_0.txt</ulink>)</para> > </legalnotice> > > <librarypurpose>A set of facilities to ease programming with dates and >times. > </librarypurpose> > <librarycategory name="category:misc"/> ></libraryinfo> This is so that the legal notice can be rendered in the introduction page of the library documentation. >So it seems one option would be to change the boostbook schema and allow >copyright and legal notice elements in each file. The other possibility >would >be to add this as a comment: > > > >Thoughts on how best to approach this? I think the best approach would be in a comment, since adding copyright markup to each file would require adding code in the XSLT files to prevent it from being rendered. Regards, Reece _________________________________________________________________ Use MSN Messenger to send music and pics to your friends http://www.msn.co.uk/messenger

3 2

VC8.0 and algorithm/string/replace failure
by Victor A. Wagner Jr. 30 Aug '04

30 Aug '04

I compiled replace_test with vs.net2005 and debugged it. It clearly throws an exception, and re-throws it later from the test framework, which causes the "ask to terminate abnormally" message. BUT............ the output window has the following in it _before_ the exception is thrown: error: list iterators incompatible, at g:\program files\microsoft visual studio 8\vc\include\list: 283 I'm not sure why it thinks the iterators are incompatible, but.... it might help the author(s) to have the above error mentioned in the failure log (I don't know why it's not there) Victor A. Wagner Jr. http://rudbek.com The five most dangerous words in the English language: "There oughta be a law"

4 7

[inspect] Status...
by Rene Rivera 30 Aug '04

30 Aug '04

Just updating on the current status from the inspection tests... 8105 files scanned 680 directories scanned 184 problems reported problem counts: 2 files with invalid line endings 0 bookmarks with invalid characters 1 invalid urls 72 broken links 46 unlinked files 59 file names too long 4 files with tabs Of interest in those... libs/thread/test/Jamfile: invalid (cr only) line-ending libs/graph/example/fibonacci_heap.cpp: tabs in file libs/range/test/partial_workaround.cpp: tabs in file libs/serialization/test/test_utf8_codecvt.cpp: tabs in file libs/spirit/test/distinct_tests.cpp: tabs in file 8105 files scanned 680 directories scanned 4397 problems reported problem counts: 2970 files missing Boost license info 1427 files missing copyrights -- -- Grafik - Don't Assume Anything -- Redshift Software, Inc. - http://redshift-software.com -- rrivera/acm.org - grafik/redshift-software.com - 102708583/icq

2 1

Re: Your copyright in the Boost C++ Libraries (second try)
by Christian Engström 30 Aug '04

30 Aug '04

I Christian Engström, give permission for all my contributions to Boost to be moved to the Boost Software License Version 1. Christian Engström John Maddock wrote: > Hi, if you are receiving this reminder again it is because there are > one or more > files in the Boost C++ libraries that contain your copyright declaration. > [...]

1 0

Fw: Your copyright in the Boost C++ Libraries (second try)
by John Maddock 30 Aug '04

30 Aug '04

And another permission: ----- Original Message ----- From: "Nicolai Josuttis" <nico(a)josuttis.de> To: "John Maddock" <john(a)johnmaddock.co.uk> Cc: <boost-owner(a)lists.boost.org> Sent: Sunday, August 29, 2004 2:44 PM Subject: Re: Your copyright in the Boost C++ Libraries (second try) > Hi John, > sorry for not answering immediately, but these > mails were considered as spam. > However,m this time I saw it. > > OK, here it is: > > I, Nicolai Josuttis, give > permission for all my contributions to Boost to be moved to the Boost > Software License Version 1. > > Best > Nico > > John Maddock wrote: > >> >> Hi, if you are receiving this reminder again it is because there are one >> or more >> files in the Boost C++ libraries that contain your copyright declaration. >> >> PLEASE, do take a little time to read through this and respond - it would >> help us a great deal! If you already have responded, then I missed it, >> and I apologise, but if you could take a little time to put me right, >> then that would be appreciated as well. >> >> As some of you will know already, in order to simplify Boost licensing, >> we >> are in the process of moving to a single common license >> (http://www.boost.org/LICENSE_1_0.txt) The aim of the new license is to >> provide a single wording, that encapsulates our previous license >> requirements, and which will simplify the evaluation of Boost for >> commercial >> organisations. >> >> So here's the big question: may we have permission to change code that >> has >> your copyright to use the new license? Your copyright declaration would >> obviously remain, it's only the license text that would change. >> >> If "yes", and you have Boost-cvs write access, please add your name to >> more/blanket_permission.txt. >> >> If "yes" and you have access to the Boost mailing list >> (boost(a)lists.boost.org) then please post a message there, subject "Boost >> License Permission", and content something like: "I Jim Hacker, give >> permission for all my contributions to Boost to be moved to the Boost >> Software License Version 1". (This gives us a public record that your >> permission has been given). >> >> If "yes", but you are no longer subscribed to the Boost list then please >> reply to boost-owner(a)lists.boost.org with content as above. >> >> If "no", please give us some idea what objections you may have. >> >> Finally, if this message was sent in error, please let us know, it's just >> about possible that we have you mixed up with someone of the same name, >> and >> if so apologies. >> >> Many thanks for your time, >> >> John Maddock. >> >> > > -- > Nicolai M. Josuttis http://www.josuttis.de/ > Solutions in Time mailto:nico@josuttis.de > phone: 0700 / 5678 8888 (international: +49 700 5678 8888) > 0700 / JOSUTTIS (international: +49 700 JOSUTTIS) > >

1 0

Nightly CVS docs broken?
by Vladimir Prus 30 Aug '04

30 Aug '04

The link to nightly CVS documentation on boost.org appears to be broken: it leads to http://www.boost.org/regression-logs/cs-win32_metacomm/doc/html/libraries.h… which generates an error. Is it possible to fix somehow? Thanks, Volodya

2 2