- Boost - lists.preview.boost.org

Why do many examples lack #includes and namespaces?
by TyRoXx 16 Dec '13

16 Dec '13

The first example for Lambda 1.55 is: list<int> v(10); for_each(v.begin(), v.end(), _1 = 1); This is not useful at all. The first two question that every new user would ask are: - which header files do I have to include? - which namespaces are the things from? Is _1 the global that I already know from bind or not? A useful example would be: #include <list> #include <algorithm> #include <boost/lambda/core.hpp> int main() { std::list<int> v(10); std::for_each(v.begin(), v.end(), boost::lambda::_1 = 1); } I cannot even find the usual .cpp example files anywhere on http://www.boost.org/doc/libs/1_55_0/doc/html/lambda.html . I have been using Boost for a long time so I knew that Lambda headers would probably be in a directory called boost/lambda/. I also knew that list and for_each would most probably come from the standard library. But not everyone does! Material for beginners, especially "examples", should be compresensible. I do not understand why that is not the case for so many Boost libraries. Some libraries are telling you the namespace but not the headers: - DateTime - Function - Signals2 in the tutorials - Array (the only example is in the "Design Rationale") ... http://www.boost.org/doc/libs/1_55_0/doc/html/date_time/examples/general_us… http://www.boost.org/doc/libs/1_55_0/doc/html/function/tutorial.html http://www.boost.org/doc/libs/1_55_0/doc/html/signals2/examples.html http://www.boost.org/doc/libs/1_55_0/doc/html/array/rationale.html Some libraries have better examples: - Atomic has useful examples with #includes and namespaces. - Random has very useful examples, too. - Unordered has minimal, but useful examples on the main page. ... http://www.boost.org/doc/libs/1_55_0/doc/html/atomic/usage_examples.html http://www.boost.org/doc/libs/1_55_0/doc/html/boost_random/tutorial.html#bo… http://www.boost.org/doc/libs/1_55_0/doc/html/unordered.html I am talking about the HTML documentation only. That's all for now.

8 7

[git] Error 403 trying to send first push request
by Joachim Faulhaber 16 Dec '13

16 Dec '13

Hi all, This is my first maintenance trial using git and modular-boost. I followed the wiki instructions, got an account at GitHub, cloned the modular-boost repository pulled the master branch of my library (icl), made a bug fix and committed it locally. Finally pushing the change failed with this message: Joachims-MacBook-Air:icl jofa$ git push Username for 'https://github.com': jofaber Password for 'https://jofaber@github.com': error: The requested URL returned error: 403 while accessing https://github.com/boostorg/icl.git/info/refs fatal: HTTP request failed (I'm working under OSX 10.8.5 with clang version 4.1, git version 1.7.10.2 (Apple Git-33)). Any help appreciated, Joachim -- Interval Container Library [Boost.Icl] http://www.joachim-faulhaber.de

3 4

[chrono] mac thread_clock support
by Chris Brown 15 Dec '13

15 Dec '13

I noticed that chrono doesn't seem to support thread_clock on mac (the posix implementation doesn't work for it) so I wrote up an implementation. https://github.com/boostorg/chrono/pull/1 I was pretty surprised this didn't already exist, but haven't been able to find anything in the mailing list or trac regarding it. Curious to hear if there was a reason to avoid this implementation or if it's just something no one got around to. Posting here for visibility since the git / Github workflow is relatively new to a lot of people here.

3 7

[fusion][mpl] transform_view and as_vector compilation error
by Karsten Ahnert 15 Dec '13

15 Dec '13

Hi, we have a problem in our code base. I extracted a minimal example, see below. The code works with g++ version up to 4.8.1 and C++11 but fails to compile with gcc-4.8.1 and clang3.2 and newer. Any ideas what goes wrong here? Should I further dig into the problem? I am using boost 1.54. #include <boost/fusion/include/vector.hpp> #include <boost/fusion/include/as_vector.hpp> #include <boost/fusion/include/transform_view.hpp> struct t1 { struct result_type1 {}; }; struct t2 { struct result_type1 {}; }; struct my_transform { template< class > struct result; template< class F , class T > struct result< F( T ) > { typedef typename T::result_type1 type; }; }; int main( int argc , char *argv[] ) { namespace fusion = boost::fusion; typedef fusion::vector< t1 , t2 > vector_type; typedef fusion::transform_view< vector_type , my_transform > transform_type; typedef fusion::result_of::as_vector< transform_type >::type transformed_vector_type; return 0; }

2 2

Project
by Utsav Tiwary 15 Dec '13

15 Dec '13

I am a third year engineering under graduate from India in interested in Open MPI Programming. Could you please furnish some basic details about the project involving Open MPI done under GSoC 2013,like pre requisites ,basic idea behind the project etc. ?

1 0

Re: [boost] [serialization] new portable binary archive based, on HDF5
by Daniel Koester 15 Dec '13

15 Dec '13

On Fri, 13 Dec 2013 22:39, Mateusz Loskot <mateusz(a)loskot.net> wrote: >> My suggestion would be for Robert Ramey and >> other interested developers to consider the inclusion of my work in the >> official serialization code. I am not sure about the exact procedure in >> this case. > > I *guess*, it may qualify for the Fast Track Review: > http://www.boost.org/community/reviews.html I will check out the procedure. >> I have forked from the Git serialization archive. All supplied tests >> have passed for gcc on Linux and MSVC 10 on Windows. My work can be >> pulled from >> >> https://github.com/dk1978/serialization/tree/development > > Hmm, have you updated the development branch with the upstream lately, > as well as the master in your fork? > > AFAICT it is a few months behind the current upstream, the master too, > it seems impossible to generate a diff usable enough to review your additions. You are right, I have neglected to synchronize with the upstream master, and mostly ignored my own master branch. Please try again now. Best regards, Daniel Koester

2 1

[git] Getting set up to test?
by Beman Dawes 15 Dec '13

15 Dec '13

Before making changes to, say, the develop branch of a library, I want to set the super-project to the latest master so tests run in the most realistic environment, as Robert Ramey has advocated for years. Presumably that means doing something like this: cd modular-boost git checkout master git pull --recurse-submodules cd modular-boost/libs/mylib git checkout develop Is "git pull --recurse-submodules" the correct command to update the submodules? Do I also need to do "git submodule update" or possibly "git submodule update --remote"? Thanks, --Beman

2 1

[conversion] try_lexical_cast and 200$
by Antony Polukhin 15 Dec '13

15 Dec '13

There is a bounty for implementing a `try_lexical_cast` from Facebook: https://www.bountysource.com/issues/1352296-exceptionless-lexical_cast Because of that three questions arise: * Is there some official policy for "feature implementation sponsorship"? * `try_lexical_cast` must be added or not? And if yes, what signature is preffered: bool try_lexical_cast<Out, In>(Out&, const In&) bool try_lexical_cast<Out, In>(const In&, Out&) boost::optional<Out> try_lexical_cast<Out, In>(const In&) * Is there a bank account "for Boost's needs" that is used to do hosting payments, conference room rental payments and other support stuff (to were those 200$ may be transfered)? -- Best regards, Antony Polukhin

13 25

Splitting/rearranging Boost.Utility
by Peter Dimov 14 Dec '13

14 Dec '13

Currently, Boost.Utility is something of a mixed bag and consequently, a dependency knot. Nearly everything in Boost depends on Utility via (say) boost/assert.hpp, while at the same time Utility itself depends on much of Boost via Iterator (and others). Something Must Be Done. At minimum, assert.hpp, current_function.hpp, ref.hpp, checked_delete.hpp and addressof.hpp need to be moved out. I suspect that call_traits.hpp, compressed_pair.hpp, enable_if.hpp also don't belong. assert.hpp and current_function.hpp should probably go into their own library, Boost.Assert. ref.hpp might go to Boost.Bind. checked_delete.hpp might perhaps go to SmartPtr. No idea where addressof.hpp needs to go. These are just suggestions. I'm open to any resolution, as long as the dependency (and, to a lesser extent, maintenance) problem is resolved to everyone's satisfaction.

8 23

Any interest in a framework for memoization?
by Giacomo Drago 14 Dec '13

14 Dec '13

Hi, I would like to submit a small framework for memoization I am working on (http://projects.giacomodrago.com/c++memo/) At the moment, it is still in a beta stage and does not comply with Boost guidelines. I would like to know if you might be interested. The idea is to have a tool for rapid prototyping of software components implementing dynamic programming algorithms or, anyhow, requiring memoization. I paste here a simple "fibonacci" example: ---------------------- int fibonacci(int i, std::function<int(int)> prereqs) { if (i == 0) return 0; if (i <= 2) return 1; return prereqs(i-1) + prereqs(i-2); } cppmemo::CppMemo<int, int> cppMemo; std::cout << cppMemo.getValue(30, fibonacci) << std::endl; ---------------------- The code will print 832040, without recursive calls to the "fibonacci" function. A stack is managed "under the hood" and appropriate calls are made to the "fibonacci" callback to gather pre-requirements and then call the function again when they are available. This is just a toy example. There is more to know about, including important caveats (if you are interested, please read the documentation carefully): some defaults for template and method parameters are useful for conciseness but are EVIL if you don't know about them. The framework also provides automatic parallelization (when possible). It is important to remark that at this moment the framework relies on fcmm (fast concurrent memoization map), a custom hashmap that is optimized for this very purpose. This can be changed quite easily, as fcmm is not exposed to the client. A C++11 compliant compiler is needed. Thanks for your attention. Giacomo

3 6

Re: [boost] [serialization] new portable binary archive based on HDF5
by Daniel Koester 14 Dec '13

14 Dec '13

In reply to Hartmut Kaiser: >> I would like to share a contribution to the boost::serialization module. >> It provides an answer to the challenge of supporting portable binary >> archives by using the well-known and stable HDF5 library >> (http://www.hdfgroup.org/HDF5) After first signals of interest by other >> users I have decided to fully integrate my work with the current >> boost::serialization module. My suggestion would be for Robert Ramey and >> other interested developers to consider the inclusion of my work in the >> official serialization code. I am not sure about the exact procedure in >> this case. > Two questions: > > 1) What license do you use for your contribution? The source files don't > mention it. I plan to use the Boost Software License, and document this and the archive interface later when the future of this project is decided. > 2) Do you have performance data comparing the new archive with the existing > ones? I'd be very interested to use this archive for HPX > (https://github.com/STEllAR-GROUP/hpx) as an alternative for our homegrown > derivative of the original portable binary archive available as an example > in Boost.Serialization. But for this I'd like to known what performance > advantages/disadvantages this might cause. I do not have performance data. Do you have suggestions for an interesting benchmark, and a link to your own binary archive implementation? I will be happy to get back to you on this point. Best regards, Daniel Koester

2 1

[build] duplicate files in converted history
by Vladimir Prus 14 Dec '13

14 Dec '13

I am getting this: $ git ls-tree 49ba2cf8dd415c100fef66f35e6686b362abe6ee:jam_src <snip> 100644 blob 92965f3a17e1439dc322b555a2c17c5cf03ffa2e jam.c $ git ls-tree 49ba2cf8dd415c100fef66f35e6686b362abe6ee:historic/jam/src <snip> 100644 blob 92965f3a17e1439dc322b555a2c17c5cf03ffa2e jam.c So, we appear to have this file at two locations in the converted history. The tree I am examined is the one where jam sources were imported to svn trunk. Is this the way things are supposed to be. If not, are there other implications of this bug? Now, I already have my toolbox ready, and will probably change rewrite script to fix this, no problem, but still. - Volodya

1 1

[git] Write permission to branch
by Edward Diener 13 Dec '13

13 Dec '13

For a Boost submodule repository is write permission for that repository needed to create a new branch ?

6 20

[serialization] new portable binary archive based on HDF5
by Daniel Koester 13 Dec '13

13 Dec '13

Dear Boost developers, I would like to share a contribution to the boost::serialization module. It provides an answer to the challenge of supporting portable binary archives by using the well-known and stable HDF5 library (http://www.hdfgroup.org/HDF5) After first signals of interest by other users I have decided to fully integrate my work with the current boost::serialization module. My suggestion would be for Robert Ramey and other interested developers to consider the inclusion of my work in the official serialization code. I am not sure about the exact procedure in this case. I have forked from the Git serialization archive. All supplied tests have passed for gcc on Linux and MSVC 10 on Windows. My work can be pulled from https://github.com/dk1978/serialization/tree/development To compile the new archive code, define the paths to the HDF5 installation using the variables HDF5_INCLUDE_PATH, HDF5_LIB_PATH, and HDF5_BIN_PATH either as environment variables or in "project-config.jam". An example using the new archive is also provided in the "example" subdirectory. Motivation for this work: The hdf5_archive project provides a new serialization archive format based on HDF5 to complement the ones already included in the boost::serialization library: plain text, XML, and native binary. By building on the established and well-known boost::serialization framework, application code can use HDF5 or switch to any of the other established archive formats with only minimal changes to the code involved. HDF5 has become a popular format to store scientific data. It is open and well-documented. Further advantages of HDF5 are the following: - the format is self-describing and portable across computing platforms - efficient storage of large arrays, parallel IO using MPI is possible (though not yet implemented) - hierarchical description of stored data - several low-level storage drivers, including single file or multiple-directory layouts - APIs to C, C++, and Fortran Best regards, Daniel Koester

3 2

[git] Git id for automated commits
by Daniel James 13 Dec '13

13 Dec '13

Hi, I'm currently using a script to update the super project, so I'd rather not use my email address for the commits. Is there an email that I should use for things like this? Daniel

1 0

Equivalent of "svn update"?
by Peter Dimov 13 Dec '13

13 Dec '13

I've followed the instructions at https://svn.boost.org/trac/boost/wiki/TryModBoost and everything worked fine. I then switched to branch develop in libs/bind and libs/smart_ptr, did developmental things there. Switched to master, did a merge, switched back to develop, did more things, git pushed the changes. This also worked fine. Now, what do I need to do to achieve the equivalent of "svn update", that is, to update all other libraries to reflect changes that others might have done in the meantime? When I do "git status" at the "modular-boost" directory (which I've named "boost"), I get: C:\Projects\boost-git\boost>git status # On branch master # Changes not staged for commit: # (use "git add <file>..." to update what will be committed) # (use "git checkout -- <file>..." to discard changes in working directory) # (commit or discard the untracked or modified content in submodules) # # modified: libs/bind (new commits) # modified: libs/smart_ptr (new commits) # modified: tools/build (untracked content) # # Untracked files: # (use "git add <file>..." to include in what will be committed) # # b2.exe # bin.v2/ # bjam.exe # boost/ # bootstrap.log # project-config.jam no changes added to commit (use "git add" and/or "git commit -a") That looks correct. Now, what do I need to do to update the master and all submodules to their latest state? Is it "git submodule update --init", "git pull", or something else? And would I then need to again checkout develop (or master, if I happen to be on that branch) in libs/bind and libs/smart_ptr?

4 8

A proposal for superproject structure and testing
by Peter Dimov 12 Dec '13

12 Dec '13

To expand on what I said in a previous message, I suggest the following structure for the superproject: - branch "latest": automatically tracks "develop" of submodules (scripted). - branch "develop": automatically tracks "master" of submodules (scripted). - branch "master": contains either the last release or the current release candidate. (This naming reflects a gitflow model. An alternative would be to use the more traditional unstable/stable for latest/develop.) Test runners test all three, in sequence. Release preparation starts with a gitflow release branch from boostorg:develop. The release manager then runs local tests and, if necessary, applies fixes on the branch by downgrading submodules to an earlier version or asking the submodule maintainer to resolve a problem and then upgrading the submodule to the later version. Ideally, this release branch should be tested non-locally. However, in the past switching branches for testing has proved a problem. Therefore, the procedure continues instead with: When the release manager is satisfied with the local test results for the release branch, he merges it to master. This becomes a release candidate. If tests prove satisfactory, the release candidate is tagged as an official release and the release branch is deleted. If not, work continues on the release branch.

11 36

Branch names?
by Beman Dawes 12 Dec '13

12 Dec '13

I've been working on Modular Boost Library Workflow Overview. See https://svn.boost.org/trac/boost/wiki/StartModWorkflow In addition to the requirement for "master" and "develop", there is strong encouragement for naming conventions if several other branches are present: * feature/descriptive-name for feature branches. For example, feature/add-roman-numeral-math. * bugfix/descriptive-name for bug-fix branches, including hotfixes. For example, bugfix/ticket-1234-crash-if-result-negative * release.n.n.n for release staging branches. For example, release.1.56.2. Should "bugfix/..." be changed to "hotfix/...", since that's the name given in the Git Flow blog posting? While I see lots of advantages to following Git Flow naming, I've never been sure what "hotfix" means. It seems weird to call a low-priority cosmetic fix a "hotfix". What is the usual practice for description names? Can they include spaces? Is there any reason to recommend spaces vs underbars vs hypthens? Are there any other branches we should recommend names for? --Beman

5 7

[git] The any library does not pull cleanly because of a forced update on develop and master.
by Cox, Michael 12 Dec '13

12 Dec '13

I tried to update my local repositories to the latest changes in each submodule: git pull; git submodule foreach --recursive git pull When it got to the libs/any submodule, I got the following message: remote: Counting objects: 27, done. remote: Compressing objects: 100% (2/2), done. remote: Total 6 (delta 2), reused 6 (delta 2) Unpacking objects: 100% (6/6), done. >From https://github.com/boostorg/any + 3f0b981...e3e5da6 develop -> origin/develop (forced update) + 38039b3...d35b2d6 master -> origin/master (forced update) * [new tag] 20131209-develop -> 20131209-develop * [new tag] 20131209-master -> 20131209-master and my editor popped up with a merge commit message to edit. I think that's because the develop and master branches were force updated, re-writing history in develop and master. Michael

10 40

[githelp] Merging two repositories
by Edward Diener 11 Dec '13

11 Dec '13

I have two separate local repositories cloned from the same other repository and I have made different changes to both. How do I merge those changes together using git ? Or do I have to do it manually ?

4 6

Heads up: I've taken care of Boost.Utility's history
by Peter Dimov 11 Dec '13

11 Dec '13

I'm still awaiting push privileges for Boost.Utility, but locally, I've already taken care of the divergencies between the develop and master branches and created the necessary merge points, so I'm announcing this here to prevent duplication of work, if someone else also decides to do the same in the meantime.

3 2

BOOST_STATIC_ASSERT_MSG bug?
by Peter Dimov 11 Dec '13

11 Dec '13

#ifndef BOOST_NO_CXX11_STATIC_ASSERT # ifndef BOOST_NO_CXX11_VARIADIC_MACROS # define BOOST_STATIC_ASSERT_MSG( ... ) static_assert(__VA_ARGS__) # else # define BOOST_STATIC_ASSERT_MSG( B, Msg ) BOOST_STATIC_ASSERT( B ) # endif #else # define BOOST_STATIC_ASSERT_MSG( B, Msg ) BOOST_STATIC_ASSERT( B ) #endif The second case looks wrong, we have static_assert in this case; should it not be static_assert( B, Msg ) instead?

2 1

RFC: Community maintained libraries
by Beman Dawes 11 Dec '13

11 Dec '13

Boost libraries have always been maintained by an individual maintainer, or perhaps a small number of individuals. That works very well most of the time, and there is no need to change that approach for libraries that continue to have active maintainers. Where we have a problem is libraries that don't have active maintainers. Someone else has to step in from time-to-time to apply patches and perform other routine maintenance. That was easy to do for our svn repo because write permissions were global. GitHub gives us some additional tools; write permissions are given at the Team level, and a team can have permissions across multiple specific repositories. Strawman Proposal ------------------------- For Boost libraries where there is no library maintainer available, turn maintenance over to a "Community" team. Initially the team members would be volunteers who are already known as experienced maintainers or patch submitters. New volunteers for team membership would establish themselves by submitting patches and pull requests. At least to get things started, the release managers could OK requests for team membership. We might seed the list of libraries being community maintained by contacting some current maintainers who have not been active for years. If we can't even contact the maintainer, that's an indication the library is a candidate for community maintenance. Patch submitters who haven't gotten any action can request a library be added to community maintenance. At least to get things started, the release managers could OK requests for community maintenance. Comments? --Beman

14 40

Re: [boost] Branch names?
by Michael H. Cox 11 Dec '13

11 Dec '13

1 0

[variant] Warning: variadic templates in boost::variant
by Antony Polukhin 11 Dec '13

11 Dec '13

Development version of Boost.Variant library now uses C++11 variadic templates when possible. This leads to more compact binaries, more readable compiler error messages and faster compilation times. However this: * breaks ABI (now variant is declared as `template <typename T0, typename... TN> class variant`) * breaks user code that was explicitly using BOOST_VARIANT_LIMIT_TYPES Code that was using BOOST_VARIANT_ENUM_PARAMS and BOOST_VARIANT_ENUM_SHIFTED_PARAMS macro *must* work correctly without any modifications. To disable variadic templates usage in variant define BOOST_VARIANT_DO_NOT_USE_VARIADIC_TEMPLATES. When all the tests will pass successfully changes will be merged to master branch, so there is a high chance that variadic templates will be supported by Boost 1.56. In case of appearance of hard to fix variadic templates errors in Boost libraries dependent on Boost.Variant - report them in this letter and I'll help in resolving issues. -- Best regards, Antony Polukhin

6 21