- Boost - lists.boost.org

[repost] regressions in como-win32-4.3.3-vc7.1
by JOAQUIN LOPEZ MU?Z 14 Aug '04

14 Aug '04

This post caught nobody's attention two days ago, but the error is still showing... -- Original message -- Please see http://tinyurl.com/5f7ba "C:\Comeau\libcomo\cnames\cstring", line 31: error #247: function "std::strlen" has already been defined extern "C++" inline size_t strlen(const char *arg) { return ::strlen(arg); } ^ "C:\Comeau\libcomo\cnames\cstring", line 31: error #247: function "std::strlen" has already been defined extern "C++" inline size_t strlen(const char *arg) { return ::strlen(arg); } ^ 2 errors detected in the compilation of "..\libs\multi_index\test\test_basic_main.cpp". The problem has nothing to do with Boost.MultiIndex, actually, as test_basic_main.cpp does not do anything more than including <boost/test/included/test_exec_monitor.hpp> These tests used to pass one or two days ago. I've inspected Boost.Test for recent changes and found none. Other libraries are affected by this problem: algorithm/minmax algorithm/string program_options ... Thanks Joaquín M López Muñoz Telefónica, Investigación y Desarrollo

3 5

regression - failed tests for date_time and regex in vc7.1 toolkit
by Bronek Kozicki 14 Aug '04

14 Aug '04

I've noticed that regression test I'm running are failing on almost all tests in date_time and many tests in regex. Here's what I've found: bjam is building static library: libboost_date_time-71-mt-gd-1_32.lib Please note that there's no "vc71", but "71" instead. File: libboost_date_time-vc71-mt-gd-1_32.lib is not build, thus all tests linking to this file are failing. When I copy libboost_date_time-71-mt-gd-1_32.lib into libboost_date_time-vc71-mt-gd-1_32.lib , date_time tests are no longer failing. The same applies to dynamic library: what is actually being build is named: boost_date_time-71-mt-gd-1_32.lib boost_date_time-71-mt-gd-1_32.dll but for tests to succeed there should be: boost_date_time-vc71-mt-gd-1_32.lib boost_date_time-vc71-mt-gd-1_32.dll What's even more interesting, library files "71" are actually being used, too (besides "vc71" files). I've found that "cmd" files are referring to "71" files. I do not know what is referring to "vc71" files (maybe autolink feature?) The same problem plagued regression tests for regex library - libraries build by bjam are named : libboost_regex-71-mt-gd-1_32.lib boost_regex-71-mt-gd-1_32.lib boost_regex-71-mt-gd-1_32.dll while tests require also files named : libboost_regex-vc71-mt-gd-1_32.lib boost_regex-vc71-mt-gd-1_32.lib boost_regex-vc71-mt-gd-1_32.dll Any ideas what might be wrong? B.

3 6

"Making a Boost Library" - article for C/C++ user's journal - request for critique.
by Robert Ramey 14 Aug '04

14 Aug '04

Fellow boosters, Editors of the C/C++ users journal accepted my proposal for an article titled "Making a Boost Library" (subject to final review of course). Rather than being "code centric" as other articles for this journal are, this will focus on the Boost library development and review process from the vantage point of a library submitter. I've got the article done. Its 8 pages (3500 words) in Microsoft word format. If anyone is interested in reading this and critiquing it, please advise me by private email and I'll send you a copy for your review. Robert Ramey

1 0

[info] Copyright permission.
by John Maddock 14 Aug '04

14 Aug '04

Forwarded here, so we have a permanent record only: ----- Original Message ----- From: "Gregory Colvin" <gregory.colvin(a)oracle.com> To: "John Maddock" <john(a)johnmaddock.co.uk> Cc: <boost-owner(a)lists.boost.org> Sent: Friday, August 13, 2004 3:29 PM Subject: Re: Your Copyright in the Boost C++ Libraries. > > On Aug 13, 2004, at 5:29 AM, John Maddock wrote: > > > Hi, if you are receiving this message it is because there are one or > > more > > files in the Boost C++ libraries that contain your copyright > > declaration. > > > > 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 the 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). > > I, Greg Colvin, give permission for all my contributions to Boost to > be moved to the Boost Software License Version 1". > > > > > 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. > > (Boost Moderator). > > > >

1 0

Regression test results (metacomm broken?)
by Victor A. Wagner Jr. 14 Aug '04

14 Aug '04

Is it just me or has the metacomm site quit processing results. I've done _two_ runs, and my logs show the ftp suceeding, since the one they credit me with. (16 hours old at this point) the results from me (RudbekAssociates) is also the one showing with the latest timestamp Victor A. Wagner Jr. http://rudbek.com The five most dangerous words in the English language: "There oughta be a law"

2 1

Re:RE: [boost] [serialization] BOOST_CLASS_EXPORT_GUID: expected ...
by Robert Ramey 13 Aug '04

13 Aug '04

>The load tests need the output of the save tests, but they are not run in >that >order. If you run the tests again, then the output from the previous run >should still be there, and then you can see if they actually pass or not If any Jamfile guru wants to explain a convenient way to require certain tests to be performed in a certain order I would be very grateful. Also, I would caution against using the development tree in production code. We're constantly fiddling with things like Jamfiles, options, etc for compilers we don't always have direct access to. Sometimes stuff is temporarily broken. I do role in bug fixes so any improvements you add to your production package should automagically show up in the next release. Robert Ramey

2 1

Re: Boost License changes
by Juergen Hunold 13 Aug '04

13 Aug '04

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi John ! On Friday 13 August 2004 14:07, John Maddock wrote: > Juergen, > > I hope I'm addressing this to the correct person, Yes. > As 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 the our previous > license requirements, and which will simplify the evaluation of Boost > for commercial organisations. I've seen the various threads about that change. > So here's the big question: may we have permission to change code > that has copyright "Institute of Transport, Railway Construction and > Operation, University of Hanover, Germany" to use the new license? > The copyright declaration would obviously remain, it's only the > license text that would change. > If "yes", please post a message to the main boost list > (boost(a)lists.boost.org) indicating that this is the case. Yes, you may change all code with the above copyright to use the new boost license. > John Maddock. > [Boost Moderator] Yours, Jürgen - -- * Dipl.-Math. Jürgen Hunold ! Institut für Verkehrswesen, Eisenbahnbau * voice: ++49 511 762-2529 ! und -betrieb, Universität Hannover * fax : ++49 511 762-3001 ! Appelstrasse 9a, D-30167 Hannover * hunold(a)ive.uni-hannover.de ! www.ive.uni-hannover.de -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.4 (GNU/Linux) iD8DBQFBHPz5ljbJ/LLrxrYRAm3aAKDLADBjQ0OVixaObAy+gxLFXbc5ZgCeN59Z eSnTmoEBwoFjUWMXrcXDU3Q= =v4gh -----END PGP SIGNATURE-----

1 0

Re:[boost] [serialization] export macros
by Robert Ramey 13 Aug '04

13 Aug '04

> Is there an alternative to the macros for registering types? > (A sample program exhibiting what I am trying is attached.) Have you considered using "forward registration" rather than EXPORT? int main() { Status< Algo< int, double > >* p_s = new DerivedStatus< Algo< int, double > >; ====> boost::archive::text_oarchive oa(std::cout); oa.boost::archive::register_type( static_cast<DerivedStatus_Algo_int_double *>(NULL) ); oa << p_s; } Robert Ramey

1 0

Re:[boost] [serialization] export macros
by Robert Ramey 13 Aug '04

13 Aug '04

> Does anyone know a work around that does not need the typedefs? I had > a look at the preprocessor library but I couldnt come up with a > solution. Is there an alternative to the macros for registering types? The BOOST_CLASS_EXPORT facility has been a struggle to get implemented so that it can be used in a natural way. Besides the problem you mention here (which is a new one for me) There were some new ones that have become apparent with compilers that use two-phase lookup. Another problem is over-zealous unreachable code optimization of some compilers eliminates the code that implements the serialization. The fix for this is inherently non-portable. Subject to the above it does work - but is kind of fragile and really not quite right - it will need to be re-implemented (yet) again. So I'm interested in your cases. Is it possible you could make a few boostified test programs that we would like to see work but currently fail at compile time? The really funny part about BOOST_CLASS_EXPORT is that is started out soooo simple and ends up touching on all sorts of unexpected difficulties. The good part is that its not a fundamental building block of the system so fixing it up won't impact the rest of the system. So there is hope. Robert Ramey

1 0

date_time broken on intel-win32-8.0
by Stefan Slapeta 13 Aug '04

13 Aug '04

http://tinyurl.com/4y6tx LINK : fatal error LNK1104: cannot open file 'libboost_date_time-iw-mt-gd-1_32.lib' Has anybody an idea where to search? Stefan

3 5