May 2013 - Boost - lists.boost.org

[build] Conditional sources and compiler options
by Andrey Semashev 06 May '13

06 May '13

Hi, I'm trying to add some SIMD support to my library and I'm having problems with setting up my Jamfile properly. Basically, I have a number of source files that only need to be compiled when the compiler is of a specific version or later (say, gcc >= 4.7 or msvc >= 12 or intel >= X.Y on x86 but not for other compilers/architectures and not for older versions of the mentioned compilers). Next, I need to add special compiler options (specific for each compiler) just for these files. Natuarlly, I also have other source files that need not these tricks and should be compiled normally, without any special options. All source files have to be compiled into a single library (static or dynamic). Is there any way to do it in Boost.Build? I've read about alternatives in the docs, but they require duplicating the library target, which I'd like to avoid. I also read about conditional requirements, but as I understand they don't allow to specify version conditions (like <toolset>gcc >= 4.7). Also I didn't find any way to specify special build options for particular files. Thanks in advance.

2 2

[boost.preprocessor.iteration] relative file iteration not supported
by Adrian H 05 May '13

05 May '13

Hello, I have used Boost in the past and do like the use of the preprocessor as a way of making repetitious code. However, I found that in a project that I was working on, I needed the ability to iterate over a file relative to the file I was iterating from, and that file wasn't relative to any of the include directories. I thought about this and it makes sense to why it didn't work, but I can see that it can prove to be inconvenient sometimes when someone wishes to make library header files relative to each other to allow for flexible positioning of the related files in the source tree or not have to add new items to the include directory set. I have written a preprocessor library to fill this need, but was wondering if such a library would be of interest to the Boost community. I deliberately tweaked it to be similar to the Boost Preprocessor Iterator library, and can be acquired at https://sourceforge.net/projects/relativeiterate/, including documentation and an example located in the README.txt file. I would appreciate any feedback you may have on this and would be more than willing to make changes to accommodate what would be needed for it to be included into the Boost set. Thank you for your time. Adrian -- ======================================== Adrian Hawryluk BSc. Computer Science ---------------------------------------- Specialising in: OOD Methodologies in UML OOP Methodologies in C, C++ and more RT Embedded Programming GUI Development ========================================

3 4

[chrono/date] constructor, factories and validation
by Vicente J. Botet Escriba 05 May '13

05 May '13

Hi all, Two approaches to checking date validation have been proposed Howard approach: * date constructors build unchecked dates (provide the same order of parameters that the date factories. * date factories (expression using operator/()) build checked dates. * could be completed with make_date that will build checked dates. Advantages: * avoids the ugly no_check (that could be replaced by no_throw) Drawbacks * The 3 sets of date constructor factories must provide all the same order of parameters. * Creating a new factory will mean creating a new constructor on each concrete date class. * The factories must be friend of the concrete dates. My original approach complemented with make_date factory: * constructors without no_check build checked dates, and with no_check parameter build well unchecked ones. Provides only decreasing lexicographic order of arguments e.g. date(year,month,day), date(year,month,day,no_check_t). * make_date provides the usual combinations as provided by factory using operator/() and no_check. * date factories (expression using operator/()) build checked dates Advantages: * There are less date constructors. * We can add as many factories as we are able to construct with checking or not. Drawbacks * the no_throw/no_check/unchecked parameter is ugly and a source of conflict. Alternative approach (reverse the defaults): * constructors without check build unchecked dates, and with check parameter build well checked ones. Provides only decreasing lexicographic order of arguments e.g. date(year,month,day), date(year,month,day,check_t). * make_date provides the usual combinations as provided by factory using operator/() and check. * a make_checked_date could have also a sens. * date factories (expression using operator/()) build checked dates Best, Vicente

1 0

[operators] The Dangling Reference Polarization
by Daniel Frey 05 May '13

05 May '13

I see some good discussion going on, thanks to everyone participating! Rest assured that I value both side's contributions, it is important for me (and those agreeing with my perspective) to be able to defend this position and also to keep looking for improvements. If our side can not be defended, it's probably not good enough :) That said, here's what I hope can serve as a first summary, let's first look at it from a technical perspective: If an operator returns an rvalue reference (T&&), it is safe for all use-cases except binding the result to an rvalue- or const-lvalue-reference: T r = a + b + c; // no problem const T r = a + b + c; // no problem T& r = a + b + c; // compile-time error const T& r = a + b + c; // problem (P1) T&& r = a + b + c; // problem (P2) It seems that this is consistent behavior across at least GCC, Clang and VC++. If a compiler shows a different behavior, its likely a compiler-bug. Technically, there is no doubt that the above is a bug, but that doesn't answer the question of whether this is a bug of the user who wrote the above code or if it's a bug of the operator to return an rvalue reference instead of an rvalue. This is where we leave the technical-aspects-land and enter the minefield of opinions, expectations and decisions. I am of the opinion that the above problematic use-cases are a bug on the users side, my line-of-thoughts and my arguments are: a) The above code is explicit, you can spot it in the users code. It's not something that happens silently. b) The above code always looked suspicious to me anyways, it has a code-smell that I just don't like. c) It rely's on the fact that it extends the lifetime of a returned temporary value. The rules of the C++ standard that allow this are quite clear, and hence it's the users responsibility to make sure that there actually is a temporary value whose lifetime can be extended. It has always been an error if the expression returned a reference and not a value. Consider const std::string& oops = std::vector<std::string>(1, "hello, world" ).front(); This compiles are creates a dangling reference, as the standard says that front() returns a reference and not a value. It's not 100% comparable as in this case the standard guarantees a bug, but the main point for C++ has always been which guarantees are given and not about a certain expectation one might have. As an example: static_assert( std::is_empty< std::tuple<> >::value, "Expectation 1 broken" ); static_assert( std::is_empty< std::array< T, 0 > >::value, "Expectation 2 broken" ); // with any valid T The first case is certainly a reasonable expectation and it currently works for all implementations of std::tuple that I know of, but it is not guaranteed! If your code relies on it and it breaks in the future, you can not blame the person who modified std::tuple. The second case is also a reasonable expectation, but it breaks everywhere and AFAIK there is no standard-conforming way to implement std::array so that std::array<T,0> is empty. For the curious, here's the StackOverflow question I asked about it: <http://stackoverflow.com/q/15512827/2073257>. Another example about how to properly handle expectations vs. guarantees: <http://stackoverflow.com/q/14882588/2073257>. d) I fail to see any valid use-case for binding the result of the expression to a reference. Can someone please provide a convincing example of why (P1) or (P2) are needed/useful? Keep in mind that in the context of operator+ (or any other operator in question), we already require that the type T is copy-/moveable and that copy-elision most likely takes place if you use "T r = …" instead of "const T& r = …". e) The alternative needs to prove its claimed/theoretical efficiency in practice. Too often I have heard about "sufficiently smart compilers", but I'm more the prove-it-in-the-real-world type. So summarizing it once again: For a) and b), this is just an observation and my personal opinion. c) means it's important to talk about guarantees, not just about expectations (no matter how reasonable they are). d) poses the question if we loose something if we don't guarantee that (P1) and (P2) are allowed. I currently have several ideas/option of how we could improve Boost.Operators, some of which I already mentioned, some which only recently occurred to me. Before I will discuss these, I'd like to resolve (if possible) the Dangling Reference Polarization and gather more opinions/feedback. So far, I found the discussion both challenging and very helpful, again thanks to everyone participating! Best regards, Daniel

5 13

[gsoc] Boost.Multiprecision Proposal
by Saurav 05 May '13

05 May '13

Hi, I have attached my proposal for the Boost.Multiprecision project. It has been provided in PDF format for better readability. Please have a look and give your feedback. Best regards, Saurav

3 4

[Boost.ThreadPool] GSOC 2013 Proposal
by Ian Forbes 04 May '13

04 May '13

Hi, I am submitting my Google Summer of Code proposal for review. My proposal is for the Boost.ThreadPool library. It will be based on work done here: https://svn.boost.org/trac/boost/wiki/SoC2013#Boost.ThreadThreadPool The current proposal can be found at: http://cs.mcgill.ca/~iforbe/gsoc2013.html The original proposal can be found here: http://www.google-melange.com/gsoc/proposal/review/google/gsoc2013/en4bz/1 , although it has broken HTML tags and is ill formatted. Thanks goes to Vicente J. Botet Escriba for pointing out that Boost does in fact has LockFree concurrent data structures. Please feel free to respond with any questions or criticism. I will try to respond to the best of my ability. Thanks, Ian Forbes

1 0

[chrono/date] year/day/week literals
by Vicente J. Botet Escriba 04 May '13

04 May '13

Hi, on the GSoC discussion about Boost.Chrono/Date proposal we were discussing about date construction. Some of us think that we need to use named types for day, month, year and week so that the date constructors are not ambiguous. Everyone agree with the constant object for month. date dt(year(2013), may, day(3)); But having to use day(3) or year(2013) seems to wordy. I was wondering if we can not add some literals for day, year and week so that we can just write date dt(2013y, may, 3d); The advantage I see in addition to been less wordy, is that we will have a compile error when the year, day or week is out of range. What do you think? Vicente

5 11

Live read-only GIT mirrors of Boost trunk SVN
by Niall Douglas 04 May '13

04 May '13

The live GIT mirrors of Boost trunk SVN have become a bit stale in recent months: https://github.com/ryppl/boost-svn, long time favourite of many, now 4 months stale. https://github.com/Beman/boost-trunk-git-svn is now 2 months stale. https://github.com/alexanderkjeldaas/boost-mirror is fresh, but has no branches nor tags. They all generally use git-svn, which tends to be brittle and not translate commits perfectly unfortunately e.g. SVN eol properties and other such metadata. So I decided to try out a new Subversion <=> Git mirroring tool, SubGit v2.0 EAP (http://subgit.com/eap/) whose author very kindly gave me a free open source licence and the result is at: https://github.com/ned14/boost-trunk All branches and tags, as far as I can tell, are mirrored correctly. I've set a cron job every 24 hours. Theoretically if we had an authors mapping file it would work in the opposite direction too i.e. be read-write. My next task is to live mirror a github repo into Boost Sandbox. Currently SubGit segfaults on this due to the complex SVN properties auto-translation which requires a specially marked up .gitattributes, but the author is very keen to fix the problem. In theory then it ought to just work. Any mistakes in my Boost mirror, or any problems, or even thoughts are gratefully received. Niall

4 6

Specific-Width Floating-Point Typedefs
by Paul A. Bristow 04 May '13

04 May '13

There has recently been some discussion on improving portability of programs using floating-point. Drawing on the benefits from definition of typedefs for int32_t, int64_t, int128_t . we have drawn up a proposal to add similar typedefs for floating-point. Our draft document (in both html and pdf) is at https://svn.boost.org/svn/boost/sandbox/precision/libs/precision/doc/html/i… https://svn.boost.org/svn/boost/sandbox/precision/libs/precision/doc/precis… We hope to propose this to the SG6 Numerics subgroup of WG21 ISO C++ standards body. But would welcome comments from Boosters to help us improve this proposal. Thank you. Paul --- Paul A. Bristow, Prizet Farmhouse, Kendal LA8 8AB UK +44 1539 561830 07714330204 pbristow(a)hetp.u-net.com

8 17

date libary construction
by Shamsher Siddiqui 04 May '13

04 May '13

i am deeply regretting for not responding your Question and for not posting the google summer of code Proposal . my pre semester exams are going on so i was busy in preparing which is ending on 7 may which is coming tuesday so i will be always not active during this period but i will try to respond to discussion which happening on the mailing list . i am finally sending only the proposal part . Proposal for DATE LIBRARY 1)Construction of Date: According to paper [2] there were many benchmark set but date should be represented in both ways i.e field and serial .Field and serial representation have their advantage over each other under certain operation . No invalid dates can be entered considering range cheaking ,non existent date if entered bad_date exception is thrown during the time of constructing the date. Constructing constant object for day,week_day. Date construction concrete structure <template class date_concrete> class date_pre { //year, month,week day //constructors //basic accessors declaration //conversion accessors //airthematic operation //camparison operator for date camparison } 2) Arithematic Operation On date Date airthematic can be done on three units i.e day,month,year.In chrono date ::calendar(Gregorian calendar) length of month and year is not fixed : day arithematic Adding number of days to date until range doesn’t exceed Implementing last day arithmetic to avoid the bad date excepti Eg. aug/last/2011 +months(1)= sep/30/2011 aug/31/2000 +months(1)=sep/31/2000 //invalid date bad date excep thrown month arithmetic Month arithmetic is quite complex compared to day arithmetic since number of are not fixed . (month day <=28 || month day =last day ) always add month month day =29 cheak for leap year and than add month Month day>=30 add but don’t add if resulting month is feb Using constant objects for month arithmetic. year arithematic Always add year to date considering the condition of leap year Reversal of date back to original date Usage of last day during year arithematic 3) InterConversion field representation & serial representation Some operation are quite fast on field but airthematic operation are much faster on serial campared field representation .So it is necessary to provide the inter-coversion so that it can be easily converted from one representation to another. Constructing bi-directional conversion of serial and field by making use of memoization optimization techinique. 4)Gregorian calendar implementation This Gregorian calendar will be a hybrid calendar that support both Julian and Gregorian calendar with a single discontinuity(the day on which Gregorian calendar was adopted ).The only difference between gregorian and julian is leap year calculation and before day of adoption of Gregorian calendar ,Julian calendar is implemented in this date library. The concrete draft for implementation for calendar template< typename ymd_type ,type name date _int _type > class gregorian_calendar { public : typedef ymd_type ymd _ type ; typedef typename ymd_type : : month_type month_ type ; typedef typename ymd_type : :day_type day_type ; typedef typename ymd_type : : year_type year_type ; typedef date_ int _type date_int_type ; static unsigned short day_of_week (const ymd _type &ymd ) ; static int week_number (const ymd_type &ymd ) ; static date_int_type day_number(const ymd_type &ymd ) ; static ymd_type from_day_number(date_int_type) ; static date_int_type julian_day_number(const ymd_type &ymd) ; static bool is_leap_year(year_type) ; static unsigned short end_of _month_day( year_type y ,month_type m ) ; static ymd _ type epoch() ; static unsigned short days _ i n _week(); }; 5)Building date iterator ,date output and date input Date Iterator will generate the date based on initial condition and end.Bidirectional iterator will do all the generation of series of date.Various iterators which will be implemented are day ,week ,month ,year Iterator. date output same as H. Hinnant. Questions: 1)could you explain us why do you think a class template date is a good alternative? i am using the template class since it will implement the user extensibility .This is very important feature for developer but not for normal user because developer likes to customise code it will give them to add or modify without removing features. 2) What is the expected service of the concrete classes? Which concrete representations would you provide? i have made the concrete class to show some of details of the date library which i am going to implement . <template date_concrete> class date_pre { /*year, month,week day construction of date_field i.e serial and field based reprasentation*/ /*details of constructor 1)date_pre(); //default constructor to initialise the date with the initial default value set 2)date_pre(year_type year,month_type month,day_type day); /*this constructor will initialize the date with values given by the user . but the constructor argument arrangement can be differ according to user or the country date format used */ 3)date_pre(year_type year,month_type month,week_of_month_type week_number,weekday_type wd); //this will constructs the date like monday in week 20 of 2013 4)date_pre(special day sp); //this will day construct the date with the special values like mothers day ,christmas in year //basic accessors declaration //conversion accessors //airthematic operation //camparison operator for date camparison //Coversion of calendar }; these all same as in H. Hinnant proposal except the conversion of Julian calendar to gregorian calendar . 3) How would you implement the memoizer for conversions? Would it be thread safe? Could you post an example how your date_pre and gregorian claendar classes are used? in summary, how your proposal improc-ves H.H implementation or my prototype? since pre semester are going so can i answer to this 2 question after coming tuesday ie 7 may 2013 sorry again for not answering ur questions i will try to be active on the mailing list during my pre semester xam Thanking You Shamsher Ahmed

1 0