March 2005 - Boost - lists.preview.boost.org

Possible solution to making boost::regex link under debug, static runtime library for .NET 2003 (VC7)
by Robert Mathews 03 Mar '05

03 Mar '05

Currently, if you try to compile a program that uses boost::regex (either boost versions 1.31 and 1.32) under the debug static runtime library(libboost_regex-vc71-sgd-1_32.lib), you'll get a bunch of multiply defined symbols something like this: LIBCMTD.lib(_wctype.obj) : error LNK2005: _iswlower already defined in c_regex_traits.obj and so on for all of the character type macros: _iswalpha _iswupper _iswlower _iswdigit _iswxdigit _iswspace _iswpunct _iswalnum _iswprint _iswgraph _iswcntrl _iswascii This has been noticed before && stems from the fact that the inline directive is ignored for debug mode, causing those inline functions to exist both in the C runtime library and in the regex lib. I found a note going back to 2002 - http://lists.boost.org/MailArchives/boost-users/msg01398.php - which indicates that this problem is longstanding. Unfortunately, I'd like to be able to use boost::regex, and I'd like to be able to use static runtime libraries. I doubt I'm the only person who'd like be able to do that. Fortunately, the references to those macros/inlines seems to be localized to only one file, libs\regex\src\c_regex_traits.cpp, so I thought I'd try fixing that file. Digging around a bit, I modified libs\regex\src\c_regex_traits.cpp to compile and link correctly for this model. I attached the diffs below; I'm sure that there was a more elegant way, but at least this is a solution that allows the regex library to link under all runtime libraries for VC7. Maybe somebody could boostify the solution and incorporate this into the next release of the boost libraries? Thanks, Rob. --- diff follows 45a46,65 > # ifdef BOOST_MSVC > // Workaround for .NET2003 (VC7) linkage problem. > // This prevents inline functions from existing in both the static debug multithreaded unicode runtime library and in this library, a problem > // which causes link errors like: > // LIBCMTD.lib(_wctype.obj) : error LNK2005: _iswlower already defined in c_regex_traits.obj > > #define iswalpha(_c) ( iswctype(_c,_ALPHA) ) > #define iswupper(_c) ( iswctype(_c,_UPPER) ) > #define iswlower(_c) ( iswctype(_c,_LOWER) ) > #define iswdigit(_c) ( iswctype(_c,_DIGIT) ) > #define iswxdigit(_c) ( iswctype(_c,_HEX) ) > #define iswspace(_c) ( iswctype(_c,_SPACE) ) > #define iswpunct(_c) ( iswctype(_c,_PUNCT) ) > #define iswalnum(_c) ( iswctype(_c,_ALPHA|_DIGIT) ) > #define iswprint(_c) ( iswctype(_c,_BLANK|_PUNCT|_ALPHA|_DIGIT) ) > #define iswgraph(_c) ( iswctype(_c,_PUNCT|_ALPHA|_DIGIT) ) > #define iswcntrl(_c) ( iswctype(_c,_CONTROL) ) > #define iswascii(_c) ( (unsigned)(_c) < 0x80 ) > # endif > 520,521c540,545 < lower_case_map[i] = (char)std::tolower(i); < } --- > # ifdef BOOST_MSVC > lower_case_map[i] = (char)tolower(i); > # else > lower_case_map[i] = (char)std::tolower(i); > # endif > } 915a940,967 > # ifdef BOOST_MSVC > bool BOOST_REGEX_CALL c_regex_traits<regex_wchar_type>::do_iswclass(regex_wchar_type c, boost::uint_fast32_t f) > { > BOOST_RE_GUARD_STACK > if((c & ~0xFF) == 0) > return BOOST_REGEX_MAKE_BOOL(re_detail::wide_unicode_classes[(uchar_type)c] & f); > if((f & char_class_alpha) && iswalpha(c)) > return true; > if((f & char_class_cntrl) && iswcntrl(c)) > return true; > if((f & char_class_digit) && iswdigit(c)) > return true; > if((f & char_class_lower) && iswlower(c)) > return true; > if((f & char_class_punct) && iswpunct(c)) > return true; > if((f & char_class_space) && iswspace(c)) > return true; > if((f & char_class_upper) && iswupper(c)) > return true; > if((f & char_class_xdigit) && iswxdigit(c)) > return true; > if(f & char_class_unicode) > return true; > return false; > } > > #else 940a993 > # endif // BOOST_MSVC

3 6

Boost.org Site Suggeswtion was [RE: [boost] Re: Adobe Open Source Library]
by Brian Braatz 03 Mar '05

03 Mar '05

Suggestion: Put something on the web site with links for projects which use boost ASL would be one of those links > -----Original Message----- > From: boost-bounces(a)lists.boost.org [mailto:boost-bounces@lists.boost.org] > On Behalf Of Hartmut Kaiser > Sent: Wednesday, March 02, 2005 11:57 PM > To: boost(a)lists.boost.org > Subject: RE: [boost] Re: Adobe Open Source Library > > > David B. Held wrote> > > > > I just saw this on Slashdot and thought the boosters might find it > > > interesting: > > > > > > Adobe has apparently open sourced some libraries- I (very > > brief) quick > > > look shows me: > > > > > > 1- They seem to be heavily using template techniques > > > [...] > > > > That doesn't surprise me at all. If I recall, Mat Marcus > > works for Adobe and is also a big Boost advocate. > > This is from the docs of the ASL (Adobe Source Libraries): > > <quote> > ASL is being developed in C++, and relies heavily on the Boost libraries > <http://www.boost.org/> which are required for building ASL. > </quote> > > Regards Hartmut > > > _______________________________________________ > Unsubscribe & other changes: > http://lists.boost.org/mailman/listinfo.cgi/boost

3 2

RE: [boost] Re: ANN: review of FSM library starts
by Augustus Saunders 03 Mar '05

03 Mar '05

Darryl Green writes: >Why the interest in miminimizinghe effort of writing custom > reactions? They are a bit of a low level access method, and result > (in less trivial, non-ininlinexamples) in hiding what the > transitions destination state is. Normally, you wouldn't use many > custom reactions. HmHmmI'm getting the feeling that I need to completely disregard the tutorial and sample code and try and actually write my own code, just relying on the reference manual. I was under the impression that custom reactions were basically event handlers, and that you effectively wind up with this processing chain: Event Happens -> custom reaction -> current state exit handler -> new state entrance handler. The idea being that over-literal seseparationf states (so that all handling is done in entrance/exit handlers) is too tedious. I guess, then, I'm confused about their purpose. >I'm not seeing what there is that I would want hidden. What do you > think needs hiding? >> I think it's a distraction. Sure, it's great that you've got this >> nice cocorrespondanceith UMUMLbut that's unimportant *in the >> tutorial*. Mention it on the about page, have an appendix all >> about it, go into detail in the references, etc. > I disagree. There needs to be some model, language etc to base the > tutorial description on. You have already said that you don't think > writing a guide to FSFSMss the right thing to do. There are any > number of books and even a standard ;-) that describe the UMUMLSFSM> concepts so that Andreas doesn't need to. It would have been > possible to pick some other complex FSFSMescription language, but > UMUMLs supposed to be universal, and gets plenty of books, tools, > courses etc built around it, so why not use it? Hold on, I think you're reading a bit much into what I'm saying. I don't give rip how you graphically represent the FSFSMs The pictures in the current tutorial are great. People who are familiar with state machines at all will understand the pretty pictures. So go nuts, use UMUMLiagrams until you're blue in the face :) However, in the tutorial, don't mention it. People who know UMUMLill instantly recognize it, other people won't care. People who specifically care about UMUMLocompatibilityeserve their own appendix or section in the reference manual with a point by point breakdown of what is and is not supported from UMUMLomenclature. Scanning through the tutorial, I don't see any terminology that strikes me as UMUMLpecific. In short, as far as I can tell, the UMUMLocorrespondences a detail of little consequence when reading the tutorial. >Ah. So you probably want an example that is a rerecogniseror some > regular language. Sure, but not too complicated. > I'd be very susuprisedf this was an area where boost fsfsms a good > choice, it would be interesting to see how you go. Also, if you want > to coconstuctsfsmso do this sort of thing quickly and easily, > wouldn't you just use a regexp, or one of the plethora of > lexer/parser construction tools and let it figure out the tedious > details for you? I would expect (haven't got around to looking at it > yet) expressive to be just the thing for that, but spirit would do > the job too with a lot less typing than using boost fsfsm Besides the point. boost::fsfsms also not a good choice for Hello World. But if you're going to tell me that you've got a FSFSMibrary, well, I expect to see such an example. Helps me get my bearings. Anyway, boost::fsfsms compile time only, so it's obviously not ususefulor this kind of stuff. Purely didactic purpose here. >This isn't very useful if the fsfsmses more complex event types. And >conversely the dedespatchethod used by boost fsfsms less than >spectacularly efficient if your events really are simple values. I think we might be forgetting all this was discussing the documentation. I want concise examples that A) compile and run, and B) hide pointless infrastructure. The StStopWatchpcppxample file has about 130 lines not counting comments and #includes, of which about 35-40 are pointless infrastructure. I'd like to see that example in 35 lines *total* beyond the #includes. > Think of a state as a scope. Think RARAII >If you don't have nested states then your machine is flat and you > would almost certainly have empty states and place all your actions > and data in the machine. What is the problem? You will have to convince me that persisting state across state changes is inherently unsafe or invokes undefined behavior. It seems obvious to me that you might stick counters, file handles, sockets, database connections, etc in state local storage, or my original example of window contents. To me, this is "obvious." In fact, I can't come up with an example where I *want* this stuff destroyed evevery time change state. Sticking everything on the machine is a non-solution, and creating extra levels of state-nesting and crossing your fingers that you never have to leave the parent state seems pretty error prone. I guess this is what "deep history" is for, but that seems like a lot of work, and you'd have to convince me that whatever mechanism you use for saving/rerestoringistory was safe. > What do you want the accept state to do that is so special? Nothing, really. It just seems like it should be so simple that we might as well have it for completeness's sake. It's not like there's an endless stream of different types of FSFSMso far as I see. > I'm more familiar with the use of the terms rerecogniserr scanner > rather than rereferingo the fsfsms an acceptor fsfsmbut yes, in > lnlnaguagerocessing those terms are standard. Unfortunately, I think > you are going to be disappointed if you try to use boost fsfsmn that > sort of application. Not that that wouldn't be useful review > feedback. Boost fsfsms much better for what wiwikipedianforms me > are called transducer fsfsmsa term I am unfamiliar with, but a class > of fsfsmhat I am very familiar with. Looking at the UMUMLocs, it is clearly based on the "transducer" mold. It would be nice if there were an appendix or footnote mentioning this more academic stuff. I'm guessing that there are plenty of people in my shoes, who are more familiar with the academic stuff than the UMUMLtuff. Thanks- Augustus __________________________________ Celebrate Yahoo!'s 10th Birthday! Yahoo! Netrospective: 100 Moments of the Web http://birthday.yahoo.com/netrospective/

2 1

Re: [boost] Re: RPC library
by Iain Hanson 03 Mar '05

03 Mar '05

On Tue, 2005-03-01 at 19:27 +0000, Jarl Lindrud wrote: > Iain Hanson <Iain.Hanson <at> videonetworks.com> writes: > > [snip] > You're absolutely right that it's in no way a replacement for CORBA, but then > its not meant to be either. I wrote it for those situations where you want > something simple and easy to use (there are many such situations, no?), where > using CORBA feels like swinging a sledgehammer at a walnut. No. CORBA is very like C++ you pay for what you use. Simple rpc schemes are a misnomer and dangerous because they don't scale and users end up re-inventing the wheel such as naming service, security etc etc. Just because you can do something does not mean you should. I would classify SOAP and your RPC lib in the clever idea but you should not do it category. /ikh

2 1

[multi_index][rfc] hashed indices, new implementation
by Joaquín Mª López Muñoz 03 Mar '05

03 Mar '05

Just a quick note to announce, for those whoe are interested in hashed indices, that I've uploaded a new implementation on the Sandbox Files arena: multi_index_030305_plus_hash_part_1.zip multi_index_030305_plus_hash_part_2.zip whit significantly less space overhead (1 pointer per node + 1 pointer per bucket, just like SGI's implementation.) Code and docs are 100% complete, comments are much appreciated before I commit this stuff to the CVS anytime soon. Thank you, Joaquín M López Muñoz Telefónica, Investigación y Desarrollo

1 0

Adobe Open Source Library
by Brian Braatz 03 Mar '05

03 Mar '05

I just saw this on Slashdot and thought the boosters might find it interesting: Adobe has apparently open sourced some libraries- I (very brief) quick look shows me: 1- They seem to be heavily using template techniques 2- They have a "concepts Library" -look familiar?: Assignable ConvertibleToBool ConvertibleToFunction ConvertibleToRange DefaultConstructible EqualityComparable LessThanComparable RegularType TransformFunction 3- They have a GUI Library (appears to have a similarity to the GUI discussion on this list awhile back) Anyways- I am just throwing this out there for discussion. SlashDot article: http://slashdot.org/articles/05/03/02/1449240.shtml?tid=152 Adobe Open Source Library http://opensource.adobe.com/

3 2

[Wiki] trouble restoring libraries under construction
by Jonathan Turkanis 03 Mar '05

03 Mar '05

Hi, I was trying to despam the Wiki, which I have done many times, but for some reason got confused (or maybe I am permanently confused). I ended up removing the spam but not restoring the previous content. It should be revision 90, I believe. Jonathan

1 0

[Boost-bugs] [ boost-Bugs-1155556 ] parse_date asserts on empty string
by SourceForge.net 03 Mar '05

03 Mar '05

Bugs item #1155556, was opened at 2005-03-02 19:27 Message generated for change (Tracker Item Submitted) made by Item Submitter You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=107586&aid=1155556&group_… Category: date_time Group: None Status: Open Resolution: None Priority: 5 Submitted By: Nobody/Anonymous (nobody) Assigned to: Jeff Garland (az_sw_dude) Summary: parse_date asserts on empty string Initial Comment: the following sample code fails with assert : (Assertion failed: valid_, file c:\dev\externallibs\boost_1_32_0\boost\token_iter ator.hpp, line 57) >>> #include <boost/date_time/posix_time/posix_time.hpp> #include <iostream> using namespace boost::gregorian; int main( int, char** ) { try { date d = from_simple_string( "" ); // never reached std::cout << to_simple_string( d ) << std::endl; } catch( std::exception & ) { // no std::exception caught ! } return 0; } >>> The error seems to be in date_time\date_parsing.hpp, line 119 : for(boost::tokenizer<>::iterator beg=tok.begin(); beg!=tok.end(), pos < spec_str.size(); ++beg, ++pos) {... the comparison "beg!=tok.end()" has no effect because the value of the expression is only determined by "pos < spec_str.size()". Therefore the string iteration goes beyond the end of the string and...assert... The function also fails if the given string is not a complete date, ie: "2004", "2004-Jan" I changed the "," to "&&" and it seems to perform better (though i didn't run the boost regression tests). I also checked the CVS repository to confirm the bug is still there. Have a nice day ! and thanks for this great library, it is really good to develop with it. Here is my email : guillaume.souchet(a)bnpparibas.com ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=107586&aid=1155556&group_… ------------------------------------------------------- SF email is sponsored by - The IT Product Guide Read honest & candid reviews on hundreds of IT Products from real users. Discover which products truly live up to the hype. Start reading now. http://ads.osdn.com/?ad_id=6595&alloc_id=14396&op=click _______________________________________________ Boost-bugs mailing list Boost-bugs(a)lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/boost-bugs

1 0

RE: ANN: review of FSM library starts
by Augustus Saunders 03 Mar '05

03 Mar '05

I hesitate to call this a review; it's more my reactions looking through the fsm documentation and examples--I have not actually used it yet. I feel like these comments should have been given a long time ago, but I guess I missed all the pre-review discussion. I'm not an expert in this area, but I feel comfortable with the ideas and have fooled around with making my own state machines in the past. My main interest in a library like fsm lies in correctness and simplicity, not performance. Before I delve into answering the "official review questionare," I want to make some overall comments. First, it is obvious to me that this library has a lot of potential. The authors have thought through a lot of issues that I've never considered. My problems mainly lie with the syntax and documentation, and I think problems with the latter stem from the former. I'd like to see some discussion on the matter. Overall, I feel the syntax is too burdensome and the organization of code that I have to write is a hopeless mess (at least, looking at the examples). For simple cases, I really want to see a way of defining a fsm with less line noise and less indirection. I haven't thought through the ramifications, but it just seems to me that some mpl, lambda, and BOOST_TYPEOF loving should make something roughly akin to the following possible: { using namespace boost::fsm; using namespace boost::lambda; // I guess there's no way to avoid declaring // these ahead of time?? class Machine, State1, State2, Event1, Event2; struct S1Data { int i; S1Data() : i(0) {} }; typedef simple_state< Machine, S1Data, BOOST_TYPEOF( cout << constant("Entering State1.") ), BOOST_TYPEOF( cout << constant("Exiting State1.") ), on_event<Event1, BOOST_TYPEOF( _1->i++ ) >, transition<Event1, State2> > State1; // the "data" member of State1 has type S1Data, and is // passed into event handlers struct S2Data { int j; S2Data() : j(0) {} }; typedef simple_state< Machine, S2Data, BOOST_TYPEOF( cout << constant("Entering State2.") ), BOOST_TYPEOF( cout << constant("Exiting State2.") ) > on_event<Event2, BOOST_TYPEOF( _1->j++ )>, transition<Event2, State1> > State2; // the "data" member of State2 has type S2Data, and is // passed into event handlers // It seems that in the propsed fsm library, // state composition is specified bottom up, // and I'm not sure what the motivation is. // We're all used to writing classes top-down. typedef state_machine< State1, State2 > Machine; Machine m; m.initiate(); // process events while (whatever) m.process_event(blah); // I don't quite understand why state_cast is necessary // versus just overloading regular casting. cout << "State1 had Event1 called " << ((State1)m).data.i << " times." << endl; cout << "State2 had Event2 called " << ((State2)m).data.j << " times." << endl; } I'm sure I've messed up some details sine I don't really get why some of the fsm mechanisms are there. Also, I'm ignoring for the moment that member dereferencing is a little uglier with lambda, so _1->i++ would have to be (_1->* &S1Data::i)++. I assume there exists a way to make state-specific data available to a lambda expression in a clean way. I've also ignored sticking in lots of mpl::vector or mpl::lists, as it's my impression that with mpl one can iterate through the template arguments and figure out if they are events or transitions or whatever. So I smushed together the different elements, and presumed for < 10 arguments (or whatever, using PREPROCESSOR lib to configure that) the mpl::vector could be omitted. Finally, even though I've not really tested BOOST_TYPEOF, they use lambda expressions as an example, so I don't think it's entirely out of line of think that it might work. Now, I don't know if this example has really reduced the total number of characters required to be typed, but it is more clear to me. I realize that I've seperated the state data out into a seperate struct; personally, I like that. More importantly to me, I can look at those typedefs and reasonably guess what's being defined. If you made FSM_ENTER and FSM_EXIT macros to alias BOOST_TYPEOF you would make it clear what those entries were for. In general, you don't have to keep refering back and forth to the reference manual to figure out what each entry is. Looking at the examples in the tutorial just made my head spin, especially trying to figure out how states and machines get tied together (this is still unclear to me). This kind of "inline" code would be great for lots of cases where you just want to increment/decrement some counter or push/pop some variable. Maybe I'm hoping for too much magic, but I feel that at least little magic or secret sauce is needed here. * What is your evaluation of the documentation? How easy (or hard) it is to understand library features? What can be improved? All the other reviewers have said nice things about the documentation, but I had a lot of trouble with the tutorial. As already mentioned by others, totally targetted at the wrong people. In general, I would drop all mention of UML except for some appendix somewhere (and the reference manual too). "Getting Started" and "Target Audience" should be in a differenct document, and "How to read this tutorial" should be removed. Then, before you get into the C++ code for "Hello, World!", you should work through "Hello, World!" and maybe "StopWatch" using pseudo-code. Even if you can work out some magic like I hope for, I think you need to map out the basic process with pseudo-code, like this: // Hello World example // First, we have to declare all of our states // and events ahead of time. This is because // all states can transition to each other, // so you can't necessarily define them in order. class mystate1, mystate2, event1, event2, etc; // Next, we're going to define some states // for our machine. Each state definition // indicates what other states it transitions // to on each input, buried in "bunch_of_stuff" typedef simple_state< bunch_of_stuff > mystate; typedef simple_state< bunch_of_stuff > mystate2; // etc // Then, we're going define a machine with // states. The first listed state is the default typdef state_machine<mystate, mystate2> Machine; // etc, in this style In general, all of your "remarks" and "comments" need to be in the code! I think the example state machines are basically OK, but I think that it is absolutely essential that you include a simple example that is alphabet-based. In fact, some helper functions for alphabet-based languages are probably a good idea, just to make the tutorial easier. In particular, you should be able to create letters-as-events, and dispatch them easily. The switch statement for dispatch in the Camera example is scandalous ;) The reference manual seems ok, although I didn't really use it. The rationale tended to feel unconvincing, although I'm not well informed on the issues. * What is your evaluation of the design? What features are supported better by alternative FSMs? What could be added (or removed) from the library? I'm confused about why entering/exiting states causes construction / destruction. It's taken as a given, but I would tend to assume that state-local-storage is maintained throughout unless I explicitly reset it. I'm also surprised by the lack of "accept" states. Maybe it's not universal, but my theory certainly had them, and it's just disconcerting for them not to be there. * The library documentation contains few not yet solved issues (name, separating the library into two parts, exception handling). What is you opinion here? boost::fsm is good, IMO. Don't know about seperating libraries. * What is your evaluation of the implementation? Are there parts of code too obscure or duplicating exiting Boost functionality? Can something be factored out to standalone library or among general utilities? Didn't look. * Are there performance bottlenecks? Does the library design fit requirements of real-time systems? How is it useable for embedded systems? Is the documentation clear on performance tradeoffs? No comment. * What is your evaluation of the potential usefulness of the library? Can you compare this FSM to other implementations? Tremendous potential. I can't compare directly to other FSM libraries/tools, but just like spirit, eliminating the need for extra tools is a great benefit. * Did you try to use the library? With what compiler? Did you have any problems? Do you have tips for better support of older compilers? What are possible portability problems? No. * How much effort did you put into your evaluation? A glance? A quick reading? In-depth study? Looked through docs and examples, thought carefully about what I wanted, and poked around to see what might be feasible. Many hours. * Are you knowledgeable about the problem domain? Not an expert, not completely ignorant. And finally, every reviewer should answer this question: * Do you think the library should be accepted as a Boost library? Be sure to say this explicitly so that your other comments don't obscure your overall opinion. I think that boost::fsm will benefit greatly from some further consideration. The Serialization library took two tries and was much improved, to good effect. I can be accused of hoping for too much "magic" syntax, but it'll take a lot of convincing that the current way is best. I'm not unreasonable, but I haven't seen any discussion on it, and Spirit, Lambda, Serialization, and others given me high expectations for usability. And lastly, thanks to all involved in bringing us boost::fsm. Despite my critique, this is first-rate work. Thanks- Augustus __________________________________ Celebrate Yahoo!'s 10th Birthday! Yahoo! Netrospective: 100 Moments of the Web http://birthday.yahoo.com/netrospective/

4 5

Re: Re: multi_index_container
by Wusel 03 Mar '05

03 Mar '05

Hi Joaquín M López Muñoz, thank you for your quick answer! I appreciate that multi_index_container exist and I can use it. In my opinion this library is very useful and facilitate many programs. >> 44 //version B: DOES compile >> 45 typedef index<employee_set,id >::type employee_set_by_id; >> 46 typedef index<employee_set,age>::type employee_set_by_age; > > OK, fine. > >> 48 //remove the name index (lines 31, 32) for following versions >> 49 >> 50 //version C: DOES compile >> 51 typedef index<employee_set,id >::type employee_set_by_id;; > > OK, too. > >> 52 >> 53 //version D: does NOT compile >> 54 //causes >> ...\include\boost-1_32\boost\mpl\vector\aux_\preprocessed\no_ctps\vector10.hpp(398) >> : >> 55 // error C2039: 'item3' : is not a member of >> 'boost::mpl::vector2<T0,T1>' >> 56 typedef index<employee_set,age>::type employee_set_by_age; > > !! Here I'm missing somethig... Isn't version D exactly the same > as version B, which does work? Exact this I wanted point out. Version D differs from version B in removing two lines 31, 32 only (i.e. the name index). Version C is nearly the same, too. Nevertheless there is a difference when compiling. best regards Volker Vieth

2 3