- Boost - lists.boost.org

RE: [boost] Re: recommend logging library?
by Simon J. Julier 19 Mar '04

19 Mar '04

I'm very interested in the question of logging libraries as well. >There are a few log4J ones out there. I've been using Log4cplus from >sourceforge and a been happy enough with it for its simple purpose. I personally agree. I've tried various different logging strategies including various types of macros and globally set log levels. However, I typically find that those approaches don't provide a sufficiently fine-grained approach when I debug code. (e.g., I might want some code to be hyperverbose and other bits to remain quiet). I like the additivity, hierarchical logger strategies of log4J et al. because they given a fairly fine-grained control over what gets logged and how. I've personally been using log4cpp (looks almost like log4cplus). It has two solutions to the issue of the cost of logging a message: 1. You can query the logger to see if the message would be logged if it were generated. Therefore, you can do things like: if (logger.isDebugEnabled()) logger.getDebugStream() << message ; (the stream support leaves a lot to be desired) 2. You can also log messages using printf-style commands: logger.debug("%s", message); These don't get evaluated unless the message is logged. Cheers, Simon

1 0

RE: [boost] Re: Trying to do something like boost::python for Net scapesSpiderMon key JavaScript engine
by Warrick Buchanan 19 Mar '04

19 Mar '04

I also stumbled across CliPP a few days ago and have been examining it somewhat but unfortunately the interpreter is not yet completely finished but I'm certainly going to keep an eye on that one :) Thanks, Warrick. -----Original Message----- From: Fredrik Blomqvist [mailto:fredrik_blomqvist@home.se] Sent: 18 March 2004 23:33 To: boost(a)lists.boost.org Subject: [boost] Re: Trying to do something like boost::python for NetscapesSpiderMon key JavaScript engine Warrick Buchanan wrote: > Hi, > > As the subject says I'm trying to a boost::python like equivalent for > SpiderMonkey (Netscape's JavaScript engine) and I've been trying(!) > to look at boost::python for inspiration. I've got to the stage > where I can quite happily expose a class like so: > > // D3D's Vector class just for example > Declare<D3DXVECTOR>::Property<0>("x", &D3DXVECTOR3::x); // 0 is > 'tinyid' Declare<D3DXVECTOR>::Property<1>("y", &D3DXVECTOR3::y); // 1 > is 'tinyid' Declare<D3DXVECTOR>::Property<2>("z", &D3DXVECTOR3::z); > // 2 is 'tinyid' Declare<D3DXVECTOR>::Definition("Vector3", > jsContext, jsObject, > jsPrototype); > > Just from those four lines you can then use the C++ class in > JavaScript without issue ie: > > var v = new Vector3(); > > v.x = 4.0f; > ...etc... > > Now I want to expose methods and constructors. Originally I was > planning to have a load of templates for each number of arguments a > function could possibly take (which obviously isn't ideal), but > looking at boost::python it seems it has a more elegant but currently > illegible to me way of doing this? Has anyone got any pointers to how > this works or comments that would help my understanding? > This isn't really my domain but I recently stumbled into a lib called CliPP http://clipp.sourceforge.net by Peder Holt. It claims to do Javascript-style C++ bindings. It seems quite "boostified" also, with connection to boost::spirit and, obviously boost::python. Might give you some ideas if nothing else. Regards // Fredrik Blomqvist _______________________________________________ Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost

1 0

binding to collections
by John Torjo 19 Mar '04

19 Mar '04

Dear boosters, Right now, you need to type pretty much if you want to bind, lets say to the key or value from a collection. Example: typedef std::map<int,std::string> map; bind(&map::value_type::first,_1); // bind to key bind(&map::value_type::second,_1); // bind to value I've made two utility classes, in order to lessen the amount of typing. Also, an example is provided. #include <boost/bind.hpp> #include <map> #include <string> #include <iostream> namespace boost { namespace coll { template< class coll_type> struct key { typedef typename coll_type::value_type pair; typedef typename pair::first_type result_type; result_type operator()( const pair & p) { return p.first; } }; template< class coll_type> struct val { typedef typename coll_type::value_type pair; typedef typename pair::second_type result_type; result_type operator()( const pair & p) { return p.second; } }; }} void print_int( int i) { std::cout << i << std::endl; } void print_str( const std::string & s) { std::cout << s << std::endl; } int main(int argc, char* argv[]) { typedef std::map<int,std::string> map; map m; m[1] = "one"; m[2] = "two"; m[3] = "three"; using namespace boost; using namespace boost::coll; // using key & val utility classes std::for_each( m.begin(), m.end(), bind( print_int, bind(key<map>(),_1)) ); // bind to key std::for_each( m.begin(), m.end(), bind( print_str, bind(val<map>(),_1)) ); // bind to val // how you can bind to key or value from a collection, now std::for_each( m.begin(), m.end(), bind( print_int, bind(&map::value_type::first,_1)) ); std::for_each( m.begin(), m.end(), bind( print_str, bind(&map::value_type::second,_1)) ); return 0; } Best, John

1 0

Re: compiling error using boost::fsm by Andreas Huber
by Ahmed Ali 19 Mar '04

19 Mar '04

>boost::fsm currently does not work with any version of MSVC before 7.1 and >probably never will. Very early versions of boost::fsm did support VC7.0 >but I >had to drop it because I kept running into the weirdest compilation >problems. >Most of them were due to the notorious ETI bug. > >Sorry, > >Andreas thats mean I must upgrade? ot it can be fixed with patch? _________________________________________________________________ Tired of spam? Get advanced junk mail protection with MSN 8. http://join.msn.com/?page=features/junkmail

2 1

[filesystem] recursive iteration
by Thorsten Ottosen 19 Mar '04

19 Mar '04

Hi Beman, I just had a look at boost::filesystem, and it felt really good to use. There is one common idiom that might be worth including: recursive traversal. So I would like some like recursive_directory_iterator i( path( "foo" ) ) , end; for( i.set_traversal_mode( files_only ); i != end; ++ i ) // print all file names recursively, default could be both dirs and files cout << i->leaf(); My motivation for this is that 1. the manual implementation of the recursion can require a bit tweaking to get right 2. other languages seem to provide such features, eg. os.walk() in python. If people like the idea, I can provide a starting point implementation. br Thorsten

6 8

RE: [boost] Re: recommend logging library?
by Hurd, Matthew 19 Mar '04

19 Mar '04

I'll add $0.02 to the conversation. Logging for me depends on the infrastructure you need the logs for. The output needs to be suitable. Two main styles I see are SNMP and Log4J style. This means app logging can interoperate. Forgetting about SNMP for the time being, I'd look at the cpp log4J clones. They support things like syslog, NT event log, files, ip messaging output with different styles of formatting. There are a few log4J ones out there. I've been using Log4cplus from sourceforge and a been happy enough with it for its simple purpose. Being able to have zero overhead for compiling it outta-there and having enabling and disabling at run-time would be nice. The other thing I'd like is having support for stack viewing which I've used in the past for helping with production support. Regards, Matt Hurd _______________________ Susquehanna Pacific P/L hurdm(a)sig.com +61.2.8226.5029 _______________________ IMPORTANT: The information contained in this email and/or its attachments is confidential. If you are not the intended recipient, please notify the sender immediately by reply and immediately delete this message and all its attachments. Any review, use, reproduction, disclosure or dissemination of this message or any attachment by an unintended recipient is strictly prohibited. Neither this message nor any attachment is intended as or should be construed as an offer, solicitation or recommendation to buy or sell any security or other financial instrument. Neither the sender, his or her employer nor any of their respective affiliates makes any warranties as to the completeness or accuracy of any of the information contained herein or that this message or any of its attachments is free of viruses.

2 2

Inactive Objects
by scott 19 Mar '04

19 Mar '04

Hi to those folks that contributed to some recent messages related to threads, active objects and reactive objects. Mostly this was Matthew Vogt. So this is mostly for Matthew but anyone else can certainly respond. I have been interested in proposing an alternative threading model to that provided by boost.threads. Quite sensibly somone suggested that any alternative should leverage the existing facillity. J. Garland also provided the reference to the work of Schmidt; the ActiveObject pattern. I have been trying to slice off a piece of a larger library that I have been working on. Defining something that would be appropriate as a proposal to boost is very difficult. I started a requirements document to that end which become more of a research paper. Its convinced me that it is probably a no-go. The fundamental reason would be the scope of anything with pretentions to calling itself "active objects". It requires not only a significant body of supporting code but requires quite a cultural change by users (i.e. developers). I would have posted the document but it is quite long. If anyone is interested in it then just ask for it. Its a rambling piece but it has been reviewed :-) Cheers, Scott

4 4

[OT] AspectC++ 0.8 (Stable)
by Neal Becker 19 Mar '04

19 Mar '04

This may interest some of you: http://freshmeat.net/projects/aspectc/?branch_id=41657&release_id=154775

1 0

Trying to do something like boost::python for Netscapes SpiderMon key JavaScript engine
by Warrick Buchanan 18 Mar '04

18 Mar '04

Hi, As the subject says I'm trying to a boost::python like equivalent for SpiderMonkey (Netscape's JavaScript engine) and I've been trying(!) to look at boost::python for inspiration. I've got to the stage where I can quite happily expose a class like so: // D3D's Vector class just for example Declare<D3DXVECTOR>::Property<0>("x", &D3DXVECTOR3::x); // 0 is 'tinyid' Declare<D3DXVECTOR>::Property<1>("y", &D3DXVECTOR3::y); // 1 is 'tinyid' Declare<D3DXVECTOR>::Property<2>("z", &D3DXVECTOR3::z); // 2 is 'tinyid' Declare<D3DXVECTOR>::Definition("Vector3", jsContext, jsObject, jsPrototype); Just from those four lines you can then use the C++ class in JavaScript without issue ie: var v = new Vector3(); v.x = 4.0f; ...etc... Now I want to expose methods and constructors. Originally I was planning to have a load of templates for each number of arguments a function could possibly take (which obviously isn't ideal), but looking at boost::python it seems it has a more elegant but currently illegible to me way of doing this? Has anyone got any pointers to how this works or comments that would help my understanding? Thanks, Warrick

2 1

mem_fn warnings
by David Abrahams 18 Mar '04

18 Mar '04

I keep getting lots of these: c:\boost\boost\mem_fn.hpp(264) : warning C4180: qualifier applied to function type has no meaning; ignored Would it be possible for the mem_fn header to #pragma push a warning disabler for MSVC? Thanks, -- Dave Abrahams Boost Consulting www.boost-consulting.com

2 4