- Boost - lists.boost.org

[lexical_cast] Locale used during conversion
by Adam Badura 04 Jun '09

04 Jun '09

lexical_cast does not allow to pass the locale to be used during conversion. This is justified with analogy of lexical_cast to other casts which all take only one parameter. So lexical_cast always uses the current global locale. A reasonable choice. However I wanted to propose some additional features. It would be useful to provide (lets call it so for now) classic_lexical_cast which would use std::locale::classic() instead of the global one. I just finalized globalization/localization of our application by adding setting of the global locale to appropriate value according to the selected language. I noticed that all of sudden serialized data (mainly projects) stopped loading/saving correctly. The reason was easy to find. Where previously we had XML node named "item1000" we now have "item1,000" in English or "item1 000" in Polish for example. The projects are no longer interchangeable between applications running in different language version. I corrected this by adding our own lexical_cast which resets the global locale to the classic one, calls boost::lexical_cast and finally resets the global locale back. This special cast is used now whenever we are dealing with texts which are not presented to the final user. This works fine however it seems that resetting the global locale often (each time we save a number in the project for example...) is causing significant performance overhead. (Exact tests have not been yet performed but this seems the only change which might have affected all serialization slowdown.) I might reset it before saving the project and then save the project as usually but this is not very handy as I would have to remember to do it each time I am serializing any data. Also other problems are implied. Other threads for the short period of time see different global locale as well and thus might produce bad strings. And in the end it is impractical. It would be handy to have classic_lexical_cast as well as it could introduce some optimizations knowing that the requested locale is the classic one. Another thing is that it would be handy to introduce a function textual_convert (another example name) which would act like lexical_cast but would take the requested locale as a parameter (this is no longer a cast so it may). lexical_cast might take that locale as a second parameter with default value being the global locale so it could still be used just as any other casting. But I suppose this would not be well seen. Thus why not introduce a different function for that (or maybe there already is one?). It seems those changes would not be difficult to implement and could be hidden in the details of lexical_cast easily. What do you think of that? Finally lexical_cast's documentation might mention explicitly that current global locale is used as it does not mention that if I am not mistaken (I haven't found "locale" word in that document). Adam Badura

3 6

Bug Sprint Report for Wednesday, June rd
by Marshall Clow 04 Jun '09

04 Jun '09

The boost bug sprint started Friday, May 29th, and runs through Sunday June 7th. Wednesday was not a good day in terms of bug closing; the good news is that we've already (as of 6 AM) closed more bugs on Thursday than all of Wednesday. Status <https://svn.boost.org/trac/boost/report/1> 7:00 AM PDT 5/29: 797 open tickets. 7:45 AM PDT 5/30: 772 open tickets. 6:00 AM PDT 5/31: 761 open tickets. 6:00 AM PDT 6/2: 747 open tickets. 6:00 AM PDT 6/3: 748 open tickets. 6:00 AM PDT 6/4: 739 open tickets. Activity: 5/29: 6 new tickets, 77 modifications, 25 tickets closed 5/30: 2 new tickets, 42 modifications, 17 tickets closed 5/31: 2 new tickets, 27 modifications, 8 tickets closed 6/1: 2 new tickets, 42 modifications, 17 tickets closed 6/2: 5 new tickets, 18 modifications, 5 tickets closed. 6/3 4 new tickets, 19 modifications, 2 tickets closed. [ Numbers may not add up because I am counting via two different methods, using data that is time-stamped in two different time zones. Also, tickets can be closed and reopened. ] What can you do? See <https://svn.boost.org/trac/boost/wiki/BugSprintJune2009> for details. I've added four new reports to the trac: Open ticket count by owner: <https://svn.boost.org/trac/boost/report/18> Open ticket count by component: <https://svn.boost.org/trac/boost/report/19> Open ticket count by ticket type: <https://svn.boost.org/trac/boost/report/20> Open ticket count by milestone: <https://svn.boost.org/trac/boost/report/21> There's been a lot of discussion on the list about tickets; and there are about a dozen patches that have been tested and are ready to be applied. If you're looking to help, you can look at some of the older bugs, say, the ones for Boost 1.36, and determine if they still apply. If they do, note that in the ticket. There are 35 tickets that are still open that are marked as "to be fixed in Boost 1.36" See <https://svn.boost.org/trac/boost/report/3> for a list of open tickets by milestone. --------- Sign-up sheet -------- The following people have committed to participate in the bug sprint. Please add you name here, along with what you can do. 1) Marshall Clow (coordination, Mac OS X characterization) 2) Beman Dawes (filesystem, system, timer) 3) Doug Gregor (function, graph, logic, mpi, numeric, property_map, signals) 4) "joel falcou" <joel.falcou(a)u-psud.fr> ( linux gcc testing/VC 2008 testing, patching) 5) Michael Caisse <boost(a)objectmodelingdesigns.com> (bug fixing) 6) Jeremiah Willcock <jewillco(a)osl.iu.edu> (will test and commit graph patches) 7) David Abrahams (iterator, python) 8) Sohail Somani (patches for g++ 4.2.4) 9) Steven Watanabe <watanabesj(a)gmail.com> (bug fixes, como) 10) vicente.botet" <vicente.botet(a)wanadoo.fr> 11) Stjepan Rajko <stjepan.rajko(a)gmail.com> (fusion) 12) Daniel Wallin (parameter, ...) -- -- Marshall Marshall Clow Idio Software <mailto:marshall@idio.com> It is by caffeine alone I set my mind in motion. It is by the beans of Java that thoughts acquire speed, the hands acquire shaking, the shaking becomes a warning. It is by caffeine alone I set my mind in motion.

1 0

Trac unwell?
by Vladimir Prus 04 Jun '09

04 Jun '09

Hi, it looks like Trac is not feeling very well: 1. Opening an issue takes at least 10 seconds, sometimes a minute. 2. Adding a comment to an issue takes equally weird time. 3. From time to time it gives: "Database locked" errors. The page with the error does not have any additional information. Can somebody look into this? - Volodya

5 6

Ticket #637 - Adjusts mpl::pair concept to be compatible with STL pairs
by Joel Falcou 04 Jun '09

04 Jun '09

https://svn.boost.org/trac/boost/ticket/637 Is this feature request still sensible or does fusion make this useless somehow ? If not, I can provide patch+test -- ___________________________________________ Joel Falcou - Assistant Professor PARALL Team - LRI - Universite Paris Sud XI Tel : (+33)1 69 15 66 35

4 4

trunk vs release
by troy d. straszheim 04 Jun '09

04 Jun '09

From another thread: Steven Watanabe wrote: > AMDG > > troy d. straszheim wrote: >> I've been wondering a bit about this in-syncness recently. >> >> - Does anybody know offhand what the last release was that was >> branched directly from trunk? >> >> - Is there a way to determine, with SVN, how many lines of diff >> there are between trunk and release? I've tried without >> success.... the diff gets to ~80k lines and then the network times >> out. I wonder if I'm doing something wrong. Of course I can get >> some numbers with git but I'm not quite sure I trust them yet. >> >> % git diff --shortstat svn/release svn/master 3505 files changed, >> 209061 insertions(+), 58428 deletions(-) > > This is probably correct as far as it goes. However, it's hard to > tell how much it really means. About half the modified files seem to > be generated html, judging by the output of > > svn diff https://svn.boost.org/svn/boost/branches/release > https://svn.boost.org/svn/boost/trunk --summarize > Hm, that svn command isn't very helpful and runs for ~5 minutes. I'm seeing about 160k lines total changed in actual code (.cpp, .ipp, .hpp). Full report in case anybody is interested: http://sodium.resophonic.com/trunk_vs_release.txt Poking around it looks like the current release branch was created from the fixed 1.34.1 tag in oct 2007 (@ r40341). The 1.34.1 release was in turn branched from... looks like that goes back into the days of CVS, I'll consider it lost. Git maintains that the last common commit between the current trunk and release branches is > 3 years ago: http://gitorious.org/boost/svn/commit/4914d4934fbc7368e2d43e057f6fffdac0f46… So if the trunk is really just a staging area for merges to release, does the notion of having the release branch and trunk 'in sync' at this point make any sense? Why have just one such staging area? -t

13 74

[Serialization]#3118 basic_binary_oarchive.hpp - save_override possible loss of data warnings
by Jeff Flinn 04 Jun '09

04 Jun '09

I've posted the following ticket in light of the current bug sprint. ----------- The following save_override overloads cause possible loss of data warnings on MSVC8 and XCode3.1.2/gcc4.0.1 void save_override(const version_type & t, int) void save_override(const class_id_type & t, int) void save_override(const class_id_reference_type & t, int) with their respective assignments: const unsigned char x = t.t; const int_least16_t x = t.t; While a possible fix would be: const unsigned char x = static_cast<unsigned char>(t.t); const int_least16_t x = static_cast<int_least16_t>(t.t); there is still a possibility of a silent loss of data. We could be safer and use numeric_cast, but that would possibly impact code size and performance, and would introduce a library dependency. Why are the xxx_type strong typedef's using int rather than the smaller types that are being serialized? Any thoughts on the best solution? Thanks, Jeff

4 5

[Bug Sprint][random] overload resolution for constructors/seed
by Steven Watanabe 03 Jun '09

03 Jun '09

AMDG I'm looking at the following tickets: https://svn.boost.org/trac/boost/ticket/351 https://svn.boost.org/trac/boost/ticket/2887 https://svn.boost.org/trac/boost/ticket/2424 #351 and #2887 are effectively the same boost::mt19937 prng1; boost::mt19937 prng2(prng1); // does not call copy constructor #2424 is int seed = 0; boost::mt19937 prng1(seed); // doesn't compile #2424 can be fixed easily without breaking backwards compatibility. Fixing #351 and #2887 will silently change the meaning of the constructor call above. It will also be difficult to emulate the previous (documented) behavior. However, looking at the current standard draft, the constructor that causes these problems is not present, so the standard mersenne twister will have a well behaved copy constructor. Also, only lagged_fibonacci_01, mersenne_twister, and subtract_with_carry provide the offending constructor, so Boost.Random is not consistent. I'd like to have the copy constructor called. What do others think? When testing fixes, I found that rand48 behaves differently with an int64_t seed than with an int32_t seed. According to the current draft standard, they should be equivalent. Should rand48 be changed or should I leave it alone? In Christ, Steven Watanabe

2 2

[iostreams] #2406: implementation in header causes link errors
by Vladimir Prus 03 Jun '09

03 Jun '09

Looking at Trac, I've run into the bug above, which seems both fairly severe and easy to fix, so I took the liberty to commit a patch: https://svn.boost.org/trac/boost/changeset/53427 I've tested it on Linux. Can somebody give it a try on Windows? While I've added necessary BOOST_IOSTREAMS_DECL, I might have missed something. Thanks, Volodya

2 1

[parameter] Testing for existence of a parameter, building ArgumentPacks, more lazy return types
by Jeremiah Willcock 03 Jun '09

03 Jun '09

I have a few questions about the parameter library. Please note that I am using argument packs directly, since I am creating them from another data structure. 1. Is there a way to build Boost.Parameter ArgumentPacks from a program? I am currently using many internal data structures of Boost.Parameter (empty_arg_list and such). Is there another interface to use? 2. Is there a way to tell at compile-time whether a particular parameter exists in an ArgumentPack? I currently have a metafunction that uses a special default type and tests for that; is there a more "official" way? 3. I have a case where the default type for a named parameter cannot be instantiated in some cases where that parameter is given explicitly. The operator||() lazy defaults don't seem to work for that because they get the result_type member from my default right away, even if it won't be used. I currently have a hack to work around this using my parameter_exists test and some metaprogramming. The reason I am working on this low of a level is that I am playing with converting Boost.Graph named parameter structures into ArgumentPacks to have access to Boost.Parameter's nicer capabilities and syntax. I thus need to build all of the argument data structures and such within another function and then access them later. Thank you for your help. -- Jeremiah Willcock

2 5

[Modularization] A new approach to header modularization
by Beman Dawes 03 Jun '09

03 Jun '09

I've started to put together a wiki page based on our ongoing Header Modularization discussions. See http://svn.boost.org/trac/boost/wiki/HeaderModularization I listed eight approaches to meeting the Goals/Objectives/Needs/Wants, and was about to identify the weaknesses of each approach when I hit on approach nine: Move (permanently) each library's root/libs/libname contents into root/boost/libname for the library. In other words, instead of moving the boost/... tree headers to the libs/... tree, merge the libs/... tree into the boost/... tree. For example, Boost.Filesystem is currently organized like this: root/ boost/ filesystem.hpp filesystem/ config.hpp convenience.hpp exception.hpp fstream.hpp operations.hpp path.hpp libs/ filesystem/ build/ ... CMakeLists.txt doc/ ... example/ ... index.html ... module.cmake src/ ... test/ ... After the reorganization, filesystem would be organized like this: root/ boost/ filesystem.hpp filesystem/ config.hpp convenience.hpp exception.hpp fstream.hpp operations.hpp path.hpp build/ ... CMakeLists.txt doc/ ... example/ ... index.html ... module.cmake src/ ... test/ ... In theory, the root/libs/... tree isn't needed anymore. In practice we might want to keep a root/libs/libname/index.html entry for each library so docs links don't break. Apologies if someone already suggested this approach. It is so simple I can't believe someone didn't already suggest it. Unless I'm missing something, this is the only approach that appears to do fairly well for all of the Goals/Objectives/Needs/Wants, and very well for most of them. It would require some rework of scripts and other setups that depend on the current organization of libs/..., but that seems minor compared to the long-term benefits. Comments? --Beman

18 50