August 2007 - Boost - lists.preview.boost.org

SVN commit and inspection checks?
by John Maddock 05 Aug '07

05 Aug '07

Just wondering, but can we set up the SVN commit script to verify the file with our inspect program before a commit is allowed? No doubt folks will be greatly annoyed by this at first, but at least we should gradually get those inspection-failures fixed that way :-) John.

6 5

new proposal: boost-config
by Domenico Andreoli 05 Aug '07

05 Aug '07

Hi, Suppose two systems, a Fedora based on gcc 4.1 and a Debian based on gcc 4.2, both sporting Boost 1.35.0 built with --layout=system, for simplicity. Now Joe User writes his app on the Fedora system and links Boost.DateTime using the familiar -lboost_date_time switch. He builds the app and tries it on both his Fedora system and the friend's Debian one. What is going to happen? BTW, the user is a free software developer and his app is going to be ported also to other Microsoft, UNIX and popular embedded operating systems, where Boost has been built mixing --layout=system and --layout=versioned. Now he faces the problem of linking to the Boost library of interest. How he is supposed to solve the problem? This is a new summary of the binary and source portability problems that current build system is inflicting on the Boost libraries. My proposal is: 1) dump the --layout= thing, versioned/decorated sonames are always used 2) provide symlinks like the following, when appropriate: /usr/lib/libboost_date_time-gcc42-1_34_1.so -> libboost_date_time-gcc42-1_34_1.so.1.34.1 /usr/lib/libboost_date_time-gcc42-d-1_34_1.so -> libboost_date_time-gcc42-d-1_34_1.so.1.34.1 /usr/lib/libboost_date_time-gcc42-mt-1_34_1.so -> libboost_date_time-gcc42-mt-1_34_1.so.1.34.1 /usr/lib/libboost_date_time-gcc42-mt-d-1_34_1.so -> libboost_date_time-gcc42-mt-d-1_34_1.so.1.34.1 3) provide also these symlinks, if desired: /usr/lib/libboost_date_time.so -> libboost_date_time-gcc42-1_34_1.so /usr/lib/libboost_date_time-d.so -> libboost_date_time-gcc42-d-1_34_1.so /usr/lib/libboost_date_time-mt.so -> libboost_date_time-gcc42-mt-1_34_1.so /usr/lib/libboost_date_time-mt-d.so -> libboost_date_time-gcc42-mt-d-1_34_1.so 4) add more decorations as required. e.g. Python version to Boost.Python library name. This way multiple Boost.Python libraries, every built with a different Python version, could coexist on the same system. 5) provide a program, say boost-config, that may be invoked to find the right build parameters for a given library, toolset and variant(s). some examples (suppose default compiler is gcc 4.0, the one used to build Boost): $ boost-config --name=DateTime --ldflags -lboost_date_time-gcc40-1_34_1 $ boost-config --name=DateTime --toolset=gcc-4.1 --variant=mt,d --ldflags -lboost_date_time-gcc41-mt-d-1_34_1 $ boost-config --name=Threads --toolset=gcc-4.2 --ldflags -lboost_threads-gcc42-mt-1_34_1 $ boost-config --name=DateTime --toolset=gcc-4.2 --variant=python-2.5 --ldflags Error: unknown variant python-2.5 for the given library $ boost-config --name=Python --variant=python-2.4 --ldflags -lboost_python-gcc40-python24-1_34_1 $ boost-config --name=Python --toolset=gcc-4.2 --variant=python --cxxflags -I/usr/include/python if toolset is omitted, the default system compiler is used, if any (e.g. cc/c++ under linux), otherwise an error is reported. Point 1) is really optional. In case --layout is kept, the subsequent points would be related only to --layout=versioned and --layout=system should be documented as source of any sort of portability issues. Point 2) is required only on UNIX systems where GNU ld is used, it might have not sense with other linkers. Every linker has its own story. Point 3) is applicable only where 2) is and would be handy only when source portability is not an issue, to the user discretion. Big warning about source portability should be placed nearby its documentation. Point 4) would allow packaging Boost.Python on distributions in which multiple versions of Python may be available. Point 5) is the way to the complete source portability, at least across sane systems (read, probably not Windows/MSVC). Note that if Boost is built using a compiler different from the one asked to boost-config, the linker will complain about the inexistent library. It would be easy to make boost-config extract allowed parameter combinations from an XML file built by Boost.Build at build time. That's all, good night. Domenico -----[ Domenico Andreoli, aka cavok --[ http://www.dandreoli.com/gpgkey.asc ---[ 3A0F 2F80 F79C 678A 8936 4FEE 0677 9033 A20E BC50

5 9

1.34.1 PDF docs
by Vladimir Prus 05 Aug '07

05 Aug '07

Over at https://sourceforge.net/project/showfiles.php?group_id=7586&package_id=1597… I see PDF docs for 1.33.1, but not for 1.34.*? Is this omission intentional? - Volodya

1 0

interest in a stats library?
by Chris Fairles 04 Aug '07

04 Aug '07

I'm wondering if anyone has started a statistical analysis library at all? Something I might be interesting in starting if not. It could include features such as: sample mean, alpha-trimmed mean, weighted-mean, geometric-mean, harmonic-mean, mode, median, midrange. mean deviation, sample std deviation, RMS, sample range, interquartile range higher order sample moments maximum likelihood estimators (and other estimators for various distributions), confidence intervals, hypothesis testing linear regression, ANOVA even Any interested in such a lib? (i think i saw a probability lib somewhere, it could build on that as well). Chris

4 4

Testing tools location.
by Rene Rivera 04 Aug '07

04 Aug '07

NOTE: This is not question about the whole testing & release management stuff on that /other/ thread. Since I'm going to be mucking about with Buildbot this weekend I'm wondering about where to put things in the SVN repo. Instead of continuing to shove stuff into the boost-root/tools tree, I'd rather move some of the testing infrastructure tools onto their own tree as we did with the beta web site. The goal, as with the web site, is to make it easier to develop without getting in the way of the release content. In particular because like the web site, testing doesn't follow the same release schedule as the libraries. The obvious location I have in mind is a top level testing dir, with subdirs as needed: svn/testing svn/testing/buildbot svn/testing/reports ... Thoughts? ...I'll assume this is OK after a short time. And move stuff around/back if people object. -- -- Grafik - Don't Assume Anything -- Redshift Software, Inc. - http://redshift-software.com -- rrivera/acm.org - grafik/redshift-software.com -- 102708583/icq - grafikrobot/aim - grafikrobot/yahoo

2 3

[x-files] the code's out there...
by Tobias Schwinger 03 Aug '07

03 Aug '07

Dear fellow Boosters, There is lots of useful, Boost-worthy utility code out there: Implementation details of existing / upcoming / (maybe even) proposed and rejected libraries that could stand alone, code on private hard disks never brought up for review, additions to existing libraries, etc. Fact is, that for small submissions the administrative overhead of (even) a formal (fast-track) review can be too high for developers' decency. Just imagine someone allocating five fast-track slots at once just for factoring out some unrelated bits of code from an existing library. So, why not group a bunch of unrelated utilities into a set of "Boost X-Files" and have them reviewed as a single library? It's completely non-intrusive to the review process and could seriously increase its efficiency. Like the idea? As a start, I stuffed my (tested and documented) "Boost candidate bits and pieces" into a common directory in the vault: http://tinyurl.com/35vlvb Feedback and more X-files welcome, of course. Regards, Tobias

3 6

Improving --layout=system
by Caleb Epstein 03 Aug '07

03 Aug '07

I sent this to the boost-build list, but I think it should have been CC'ed to this list as well. I'd like to commit this change to the SVN trunk if there are no objections. I think its a big win for anyone who uses --layout=system. This is the way BBv1 (correctly IMHO) handled --layout=system. ---------- Forwarded message ---------- From: Caleb Epstein <caleb.epstein(a)gmail.com> Date: Aug 2, 2007 2:23 PM Subject: Improving --layout=system To: "Boost.Build developer's and user's list" <boost-build(a)lists.boost.org> On systems that support shared library version numbers, I believe that these should be used even when building with --layout=system. At present, when you build Boost with --layout=system, you end up with libraries called libboost_xyz-mt.so that have this same value set at their ELF SONAME (this is the shared library version number embedded into the library itself). The problem with these libraries is that you can never have multiple versions of them installed side-by-side in the same place. Go too far down this road and you end up with Windows-style 'DLL hell'. If instead the BOOST_VERSION is appended to the target name as it is with --layout=versioned, you end up with SONAME'd libraries that can live side-by-side with earlier/later versions of Boost. For example, by adding the version number to the libraries, you end up with a lib directory that looks like: lrwxrwxrwx 1 cepstein 33 Aug 2 13:44 libboost_date_time-mt-d.so -> libboost_date_time-mt-d.so.1.35.0 -rwxrwxr-x 1 cepstein 388K Aug 2 13:44 libboost_date_time-mt-d.so.1.35.0 lrwxrwxrwx 1 cepstein 31 Aug 2 13:44 libboost_date_time-mt.so -> libboost_date_time-mt.so.1.35.0 -rwxrwxr-x 1 cepstein 52K Aug 2 13:44 libboost_date_time-mt.so.1.35.0 ... Now when you link applications using -lboost_date_time-mt, the linker records the dependency on the library's SONAME, which is libboost_date_time_mt.so.1.35.0. No other name will satisfy this runtime dependency. As a result, you can install the (hypothetical) Boost 1.36.0libs into the same directory and old applications will continue to pick up the *.so.1.35.0 libs correctly, while new applications built when the symlink points to the new library versions will pick up *.so.1.36.0. This is how Boost.Build v1 handled --layout=system and I believe this is the most correct and desirable behavior for platforms that support versioned shared libraries. Attached is my simple patch against the latest SVN trunk to make this work. -- Caleb Epstein

1 0

SVN problem with commit
by Paul A Bristow 03 Aug '07

03 Aug '07

I am unable to commit any file or files to Boost-sandbox\math_toolkit with the following message. (I have tried an update and it does it OK). Modified: I:\Boost-sandbox\math_toolkit\boost\math\policy\policy.hpp Sending content: I:\Boost-sandbox\math_toolkit\boost\math\policy\policy.hpp Error: Commit failed (details follow): Error: MERGE request failed on '/svn/boost/sandbox/math_toolkit/boost/math/policy' Error: 'pre-commit' hook failed with error output: (nothing follows error output:) Update to I:\Boost-sandbox\math_toolkit\boost\math\policy\error_handling.hpp Completed: At revision: 38396 Modified: I:\Boost-sandbox\math_toolkit\boost\math\policy\error_handling.hpp Sending content: I:\Boost-sandbox\math_toolkit\boost\math\policy\error_handling.hpp Error: Commit failed (details follow): Error: Cannot write to the prototype revision file of transaction '38396-1' because a previous representation is currently being written by this process Any suggestions? Patience and re-try? Thanks Paul --- Paul A Bristow Prizet Farmhouse, Kendal, Cumbria UK LA8 8AB +44 1539561830 & SMS, Mobile +44 7714 330204 & SMS pbristow(a)hetp.u-net.com

5 9

Missed 1.34.1 announcement site?
by Daryle Walker 03 Aug '07

03 Aug '07

Our SourceForge project page at <http://sourceforge.net/projects/boost/> doesn't (yet) have an announcement about 1.34.1. -- Daryle Walker Mac, Internet, and Video Game Junkie darylew AT hotmail DOT com

1 0

[multi_index][ann][rfc] Preview of Boost.MultiIndex 1.35 release
by Joaquín Mª López Muñoz 03 Aug '07

03 Aug '07

Hello, I've uploaded a preview version of Boost.MultiIndex as planned for shipping with Boost 1.35 --I don't think there'll be time to include more major features. Download at the Vault: http://www.boost-consulting.com/vault/index.php?&direction=0&order=&directo… (file multi_index.zip) You can try the preview with Boost 1.34.x; if you want to test the newly added interoperability with Boost.Interprocess, you'll have to use a snapshot of Boost CVS, where this library currently lies. The new important features: * New global_fun predefined key extractor. * Added iterator_to facility. * Included support for non-standard allocators such as those of Boost.Interprocess, which makes multi_index_containers placeable in shared memory. Don't miss example #12 at the examples section, which fully exercises this new capacity. * New versions of modify and modify_key with rollback, as described in the tutorial. * Indices provide the new cbegin, cend and, when applicable, crbegin and crend member functions, in accordance with the latest drafts of the next revision of the C++ standard. A longer list of changes can be consulted at the release notes page. I'd be extremely grateful if you can try this preview version and report any problems you find. As we're still in time to discuss the new features, do not hesitate to comment on these and bring forward your proposals for improvement. Thank you very much for your interest in Boost.MultiIndex, Joaquín M López Muñoz Telefónica, Investigación y Desarrollo

3 4

[Accumulators]multiple statistic instances
by Hugo Duncan 02 Aug '07

02 Aug '07

I am trying to create an IIR filter (see e.g., http://en.wikipedia.org/wiki/IIR) using the delay buffer implementation recently discussed. To do this I need two delay buffers for my filter, one for the samples and one for the computed outputs. I have tried templating my delay buffer with a discrimating type, but I can't get more than one buffer instantiated. ie the following creates only one buffer instantation. accumulator_set<int, stats< tag::delay<>, tag::delay<my_other_delay> > > acc( tag::delay<>::cache_size = 3, tag::delay<my_other_delay>::cache_size = 4 ); Is this worth perservering with? any hints? or would it be better to wait for the next version of accumulators?

2 3

Re: [boost] [Subversion] Library authors, please read
by dan marsden 02 Aug '07

02 Aug '07

The Fusion files were all resurrections as well. They're removed again now. Thanks for the heads up Dan ___________________________________________________________ Yahoo! Mail is the world's favourite email. Don't settle for less, sign up for your free account today http://uk.rd.yahoo.com/evt=44106/*http://uk.docs.yahoo.com/mail/winter07.ht…

1 0

Cast free NULLs in algorithms
by Michael Marcin 02 Aug '07

02 Aug '07

Hello, I have something like: class A; boost::array<A*,10> the_array; std::fill( the_array.begin(), the_array.end(), NULL ); Which wont compile (at least under Visual Studio 2005) without a cast for NULL... I've had this problem before and decided to try to find an easy cast free expressive solution for it. After a few minutes I came up with: namespace detail { struct null_ptr_impl { template< typename T > operator T* () const { return static_cast<T*>(0); } }; } // end namespace detail const detail::null_ptr_impl null_ptr; std::fill( the_array.begin(), the_array.end(), null_ptr ); Does this exist somewhere already? Is it a good solution? Thanks, Michael Marcin

12 24

[type_traits] missing boost::is_complex<T>
by dherring＠ll.mit.edu 02 Aug '07

02 Aug '07

Whatever happened to boost::is_complex<T>? It looks like the idea was accepted back in 2005, but nobody ever added it to trunk... http://thread.gmane.org/gmane.comp.lib.boost.user/13692/focus=13696 http://svn.boost.org/trac/boost/ticket/492 Thanks, Daniel

1 0

about: tutorial for using boost 'Unit Test Framework'
by Jagsir Singh 02 Aug '07

02 Aug '07

Hi All, I am sure, this might allready be answered here on the list. Is there a simple step by step tutorial for using the Unit Test Framework, and the bjam,(simplest jam file). It will be highly appreciated. Is there a searchable mailing list for boost some place ? in other words, --can you search through the mailing list archives ? -- /*/Jagsir /*/////////////

3 4

[Report] 0 regressions on RC_1_34_0 (2007-08-02)
by Douglas Gregor 02 Aug '07

02 Aug '07

Boost Regression test failures Report time: 2007-07-30T10:21:37Z This report lists all regression test failures on release platforms. Detailed report: http://engineering.meta-comm.com/boost-regression/CVS-RC_1_34_0/developer/i… No failures! Yay!

1 0

Looking for project
by Sergey Razin 02 Aug '07

02 Aug '07

Hello! My name is Sergey Razin, and it was recommended to me by one of the Boost peoplzz to contact the dev. list on that question. I've e-mailed a few days ago but did not see my topic/request being included in the e-mails, therefore I've decided to try again. As I started to have more free time I would like to contribute it to Boost as a huge fan of C++ as well as Boost user. I was wondering if any of the current Boost projects require help? I will be glad to jump in and help or start a new one as well. I am looking forward to contributing to Boost as well as to help open source community with such a great set of libraries. Thank you! Sergey A. Razin, Ph.D. Software Engineer _________________________________________________________________ See what you’re getting into…before you go there. http://newlivehotmail.com

2 1

[Multi-index] questions on max_size, incomplete docs, and common code
by Daryle Walker 02 Aug '07

02 Aug '07

I'm trying out a multi-indexed container that uses a set-like (ordered_unique) interface and a vector-like (random_access) interface. Each interface has a "max_size" member function. Could they ever return different values? Some functions, like (member) swap and clear, have no definitions given. It seems like the author assumes that we'll assume that the undefined functions work like those in STL. But I'll prefer something more precise, to (1) confirm the exact semantics and (2) know the exact complexity these functions may have. (I noticed this problem while I was trying to define the complexity of my wrapper class's member functions.) Some member functions, like the aforementioned "max_size" and "clear," seem to be same for all index styles. Could those be re-factored into the main class, like "get_allocator" is? Or would that mess up the ability to keep each index's interface independent? -- Daryle Walker Mac, Internet, and Video Game Junkie darylew AT hotmail DOT com

2 1

Non-reentrant implementation of data_time/c_time.hpp under MS-Windows VC 8.0
by Alexis Wilke 02 Aug '07

02 Aug '07

Hi guys, In the code below, there are three severe bugs: 1. The addition of the #pragma warning instead of using the localtime_s() and gmtime_s() functions which are similar to the localtime_r() and gmtime_r() functions (the parameters are swapped.) Thus, if _MSC_VER >= 1400, you need a new implementation with the "safe" versions. 2. The documentation says: //! requires a pointer to a user created std::tm struct because the function is expected to WRITE in that buffer. At this time, it does not do so and if you don't use the returned pointer, the pointer to the user defined tm struct is useless. You are missing a couple of asterisks in there! *result = *std::localtime(t); 3. In a multi-thread environment, which you detect with some BOOST_THREAD_ENABLED flag, you must use a scoped lock to protect these calls. Lock, call, return is what I'd expect to see here (is it prohibited for the c_time.hpp file to include the thread lock mechanism?) I hope I will see these three bugs fixed in a future version. Thank you, Alexis Wilke P.S. I tried to post this text to your Trac system and it was rejected by Akismet. According to them, it is spam. #if (defined(_MSC_VER) && (_MSC_VER >= 1400)) #pragma warning(push) // preserve warning settings #pragma warning(disable : 4996) // disable depricated localtime/gmtime warning on vc8 #endif // _MSC_VER >= 1400 //! requires a pointer to a user created std::tm struct inline static std::tm* localtime(const std::time_t* t, std::tm* result) { result = std::localtime(t); return result; } //! requires a pointer to a user created std::tm struct inline static std::tm* gmtime(const std::time_t* t, std::tm* result) { result = std::gmtime(t); return result; } #if (defined(_MSC_VER) && (_MSC_VER >= 1400)) #pragma warning(pop) // restore warnings to previous state #endif // _MSC_VER >= 1400 ------------ Alexis Wilke President Made to Order Software, Corporation e-mail: alexis(a)m2osw.com cell-text: alexis.cell(a)m2osw.com (please, only small messages) Web Pages: http://www.m2osw.com http://www.turnwatcher.com Company e-mail: contact(a)m2osw.com Phone: +(1) 916 220 6482 Fax: +(1) 916 988 1450 Address: 9275 Blue Oak Drive Orangevale, California 95662 United States of America CONFIDENTIAL This document contains non-public proprietary information that is subject to restrictions on use and/or disclosure. If you are not the intended recipient, any dissemination, distribution or copying is strictly prohibited. If you think that you have received this e-mail message in error, please e-mail the sender and delete all copies. Thank you.

1 0

Re: [boost] Boost C++/tcl integration lib
by Maciej Sobczak 01 Aug '07

01 Aug '07

Fanzhe Cui wrote: > On the boost document page it has entry of the library: > C++/Tcl - C++/Tcl is a library that allows the easy integration of C++ and > Tcl. As was already mentioned, C++/Tcl is not a Boost library. It just uses Boost. > But the pkg from ActiveState is not free. It is free: http://www.activestate.com/store/activetcl/ Follow the "Download" link to get the free package. Distributions from ActiveState are probably the most easy way to get Tcl up and running on Windows and I recommend them for use with C++/Tcl. Regards, -- Maciej Sobczak : http://www.msobczak.com/ Programming : http://www.msobczak.com/prog/

2 1

[Accumulators] Passing construction arguments to dependent statistics
by Hugo Duncan 01 Aug '07

01 Aug '07

Given recent discussion on Time Series and Accumulators, I thought I would have a go at implementing a FIR filter for Accumulators, and have run into a problem. I decided to do this in two parts; first create a (reusable) delay statistic (similar to tail) that provides a history buffer for the signal and then create the filter (similar to tail_variate) , that will depend on the delay buffer. I have created the delay buffer with a length determined by a cache_size argument. This works well. I am now creating the filter, which has a "coefficents" argument to set the filter coefficients. I would like to pass the size of the coefficients list as the cache_size argument to the underlying delay buffer. Is this possible?

2 10

[thread_safe_signals] assertion failure (patch)
by Rodolfo Schulz de Lima 01 Aug '07

01 Aug '07

There's a little bug in member function nolock_force_unique_connection_list which causes an assertion inside nolock_cleanup_connections because the _shared_state.unique() is false because the local variable newState also references _shared_state. The attached patch corrects this rather simple bug. Regards, Rodolfo Lima.

1 0

[Subversion] Short down-time this evening
by Douglas Gregor 01 Aug '07

01 Aug '07

The Subversion repository will be down for a few hours (starting... now) while I work to repair some issues caused by a bug in the CVS- >SVN conversion utility. We should be back online this evening. My apologies for the short notice, but it's important to get this done *right*. - Doug

1 0

Re: [boost] [Important] Boost Subversion repository is now online
by David Abrahams 01 Aug '07

01 Aug '07

on Tue Jul 31 2007, Beman Dawes <bdawes-AT-acm.org> wrote: > Douglas Gregor wrote: >> Hello all, >> >> The Boost Subversion repository is now back online. All of the files >> in CVS (including their histories) have been imported into the >> Subversion repository. > > After an apparently successful checkout, diff from log isn't working. > > Edit a file. Say trunk/configure. > Using TortoiseSVN, select it. Right click to get context menu. Select > diff. Works fine. > > Using TortoiseSVN, select it again if needed. Right click to get context > menu. Select show log. Works fine. Select one of the log messages. Right > click for context menu. Select Compare with working copy. > > This results in an error dialog: PROPFIND request failed on > '/svn/boost/svn/bc/36681/trunk/boost/configure' ... 207 Multi-Status > (https://svn.boost.org) Beman, Can you reproduce your problem using the canonical command-line tool? I doubt the repo maintainers are set up to test with tortoise. -- Dave Abrahams Boost Consulting http://www.boost-consulting.com The Astoria Seminar ==> http://www.astoriaseminar.com

4 6

[function_types] [wave] Build problems (trac #1100)
by dherring＠ll.mit.edu 01 Aug '07

01 Aug '07

I hit two difficulties when building Boost from SVN-trunk. They are related to trac ticket #1100, and I tried adding some comments to it, but my changes yielded Internal Error "Submission rejected as potential spam". Here's the issue and workarounds: > svn info Path: . URL: http://svn.boost.org/svn/boost/trunk Revision: 38337 > ./configure > ./tools/jam/src/bin.linuxx86/bjam --layout=system -q ----- MkDir1 bin.v2/libs/function_types MkDir1 bin.v2/libs/function_types/build MkDir1 bin.v2/libs/function_types/build/gcc-4.1.2 MkDir1 bin.v2/libs/function_types/build/gcc-4.1.2/release MkDir1 bin.v2/libs/function_types/build/gcc-4.1.2/release/threading-multi MkDir1 /svn/boost/libs/function_types/build/timestamps Jamfile</svn/boost/libs/function_types/build>.wave /svn/boost/libs/function_types/build/timestamps/arity_loops /bin/sh: line 1: chdir: command not found /bin/sh: line 2: /svn/boost/dist/bin/wave: No such file or directory chdir /svn/boost/libs/function_types/build /svn/boost/dist/bin/wave -S/svn/boost libs/function_types/build/preprocess_arity_loops.cpp -o /svn/boost/libs/function_types/build/timestamps/arity_loops ...failed Jamfile</svn/boost/libs/function_types/build>.wave /svn/boost/libs/function_types/build/timestamps/arity_loops... ----- The file <boost>/libs/function_types/build/Jamfile.v2, contains the offending 'chdir'. When 'chdir' is changed to 'cd', the build still fails on line 2. Things finally build correctly with a > bjam --layout=system --with-wave -q Thanks, Daniel

3 3