- Boost-users - lists.boost.org

RE: [Boost-users] Trouble linking regex library on Solaris machine
by Ben Hutchings 19 Feb '04

19 Feb '04

Justin Adam Miller <millerj(a)wam.umd.edu> wrote: > Howdy. As I am only interested in using the regex library right now, I > followed the instructions for building that library only, using dmake. > Compilation went fine. When I tried using the library in a little test > program with the command: > > g++ regextest.cxx -I./boost_1_31_0 > -L./boost_1_31_0/libs/regex/build/sunpro -lboost_regex -o > regextest.cxx > > I get many undefined symbol errors. If requested, I can provide the > entire error output. It looks like you are trying to use the wrong library. SunPro refers to the Sun C++ compiler and tools whereas you are using GNU C++. You have to build the library and application with the same compiler.

2 1

Trouble linking regex library on Solaris machine
by Justin Adam Miller 19 Feb '04

19 Feb '04

Howdy. As I am only interested in using the regex library right now, I followed the instructions for building that library only, using dmake. Compilation went fine. When I tried using the library in a little test program with the command: g++ regextest.cxx -I./boost_1_31_0 -L./boost_1_31_0/libs/regex/build/sunpro -lboost_regex -o regextest.cxx I get many undefined symbol errors. If requested, I can provide the entire error output. Justin Miller

2 2

RE: [Boost-users] Re: Semaphore and barriers for boost-threads?
by Ben Hutchings 19 Feb '04

19 Feb '04

Craig Rodrigues <rodrigc(a)crodrigues.org> wrote: > It is possible to implement semaphore's with mutexes and > condition variables. I looked at the source code for > FreeBSD 5.2's implementation of semaphores to see how this was > done: > > http://www.freebsd.org/cgi/cvsweb.cgi/~checkout~/src/lib/libpthread/thre ad/thr_sem.c > > I even took a whack at writing one based on Boost (it may be buggy): > > #include <boost/thread.hpp> > #include <boost/thread/mutex.hpp> > > struct mysem { > mysem(unsigned int count): count_(count) {} > > void post() { > boost::mutex::scoped_lock l(sem_mutex_); > ++count_; > sem_cond_.notify_all(); notify_one() should be safe and more efficient here, because any of the waiters can be satisfied by the change you have made. Use notify_all() where waiters are waiting for different conditions on the same condition variable. (I'm not sure I expressed that exactly correctly - can anyone improve on that?) > } > > void wait() { > boost::mutex::scoped_lock l(sem_mutex_); > for(;;) { > > if( count_ == 0 ) sem_cond_.wait(l); This is wrong - you must always use while rather than if to test the inverse of the condition you are waiting for because it is possible that the condition is no longer true when wait() returns. (There is also a function template for wait() that tests a predicate for you.) It is even possible on some systems for wait() to return even though no-one called notify_one() or notify_all(). > else { --count_; break; } > } > } > > private: > unsigned int count_; > boost::mutex sem_mutex_; > boost::condition sem_cond_; > }; > > > What kinds of things can go wrong if I use this class? <snip> The other potential problem I see is overflow of count_, but it is exceedingly unlikely that you will be able to create enough threads to cause that.

1 0

TimePropertyMap
by Abhishek Parajuli 18 Feb '04

18 Feb '04

2 1

Help required with tokenizer class
by Matthew Delves 18 Feb '04

18 Feb '04

Greetings, Have been experimenting with the tokenizer class('s) in Boost and have come to a bit of an impass. What I am attempting to do is write an email client. Communication is okay, though what I am stuck on is breaking apart the header of the email into meaningful information. As according to RFC822 each header field is sepperated by \r\n (CRLF) I want to use this as a deliminator for the email message. Unfortunantly each field in the header can be spread over multiple lines ~ and thereby when I use \r\n as a deliminator with boost it will seperate strings if their is a \r or a \n. This is not what I want to happen. What I would like to do is sepperate on \r\n and only on that string and if there is a "\n " (\n then a space) ignore it (or preferably get rid of the \n and join the two strings). The code that I am using to create the tokenizer is as follows. boost::char_separator<char> deliminator(crlf, "", boost::drop_empty_tokens); boost::tokenizer<boost::char_separator<char> > tokenizer(recieved, deliminator); If anyone has any suggestions it would be of great help. Thanks, Matthew Delves http://www.webmastermattd.net (Please send replies directly to me and not to the list... thanks)

3 3

Mathematical expressions with containers
by Russell Hind 18 Feb '04

18 Feb '04

Currently I'm doing more and more mathematical calculations on containers, and wondered if there is any library out there that allows containers to be placed in expressions without having to use transform/for_each etc. e.g. std::vector<std::complex<double> > V1; std::vector<std::complex<double> > V2; double C1, C2; ... std::vector<std::complex<double> > Result; Result = C1 + V2 / (V1 * C2); while would be equivalent to (assuming Result has been resized) (I'm currently using a mixture of transforms and spirit::phoenix (because lambda doesn't work with my compiler)) transform(V1.begin(), V1.end(), V2.begin(), Result.begin(), C1 + phoenix::arg2 / (phoenix::arg1 * C2)); Or something to this effect that will aid with this. Cheers Russell

1 0

spirit max_limit_d min_limit_d limit_d directives
by Владимир Красовский 18 Feb '04

18 Feb '04

Hello everyone Is it expected behavior, that code like this: typedef uint_parser<double> uint_double_parser; my_rule_t max_59_uint = max_limit_d<double>(59.) [ uint_double_parser() ] [max_59_uint.value = arg1] ; causes assertion on input like "" or "aaa"? Investigating problem, I found that in function parse(ScannerT const& scan) const hit.value() used directly, without checking if it valid (see original code). If it is expected use, then what function parser_result::has_valid_attribute() for ? I've modified function parse(ScannerT const& scan) const (see my code) and everything works fine, but I'm not sure if such modification is "legitimate" and does not break anything else. original code (<boost_1_31_0>\boost\spirit\core\composite\directives.hpp): /////////////////////////////////////////////////////////////////////////// // // max_bounded class // /////////////////////////////////////////////////////////////////////////// template <typename BoundsT> struct max_bounded_gen; template <typename ParserT, typename BoundsT> struct max_bounded : public unary<ParserT, parser<max_bounded<ParserT, BoundsT> > > { typedef max_bounded<ParserT, BoundsT> self_t; typedef unary_parser_category parser_category_t; typedef max_bounded_gen<BoundsT> parser_generator_t; typedef unary<ParserT, parser<self_t> > base_t; template <typename ScannerT> struct result { typedef typename parser_result<ParserT, ScannerT>::type type; }; max_bounded(ParserT const& p, BoundsT const& max__) : base_t(p) , max_(max__) {} template <typename ScannerT> typename parser_result<self_t, ScannerT>::type parse(ScannerT const& scan) const { typedef typename parser_result<self_t, ScannerT>::type result_t; result_t hit = this->subject().parse(scan); if (hit.value() > max_) // ^^^^^^^^^^^ no check for validity return scan.no_match(); return hit; } BoundsT max_; }; my code: /////////////////////////////////////////////////////////////////////////// // // max_bounded class // /////////////////////////////////////////////////////////////////////////// template <typename BoundsT> struct max_bounded_gen; template <typename ParserT, typename BoundsT> struct max_bounded : public unary<ParserT, parser<max_bounded<ParserT, BoundsT> > > { typedef max_bounded<ParserT, BoundsT> self_t; typedef unary_parser_category parser_category_t; typedef max_bounded_gen<BoundsT> parser_generator_t; typedef unary<ParserT, parser<self_t> > base_t; template <typename ScannerT> struct result { typedef typename parser_result<ParserT, ScannerT>::type type; }; max_bounded(ParserT const& p, BoundsT const& max__) : base_t(p) , max_(max__) {} template <typename ScannerT> typename parser_result<self_t, ScannerT>::type parse(ScannerT const& scan) const { typedef typename parser_result<self_t, ScannerT>::type result_t; result_t hit = this->subject().parse(scan); if (!hit.has_valid_attribute() || hit.value() > max_) // ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ is this right ? return scan.no_match(); return hit; } BoundsT max_; }; -- Best regards, Vladimir mailto:vkrasovsky@yandex.ru

1 0

RE: [Boost-users] tips from anyone using boost with Qt?
by Tom Matelich 18 Feb '04

18 Feb '04

We actually primarily use ACE threads, mainly because we started our development before Boost or Qt had threads. The only point I might make is that Qt currently doesn't support threads on HP. For the next independent thing I write needing threads, I'll probably use the Boost thread API, only because it seems like an implementation that was approached from a very C++ POV with a lot of really smart people giving input into the design. I love Qt for its UI, but much of their cross-platform wrapping code leaves something to be desired for me. -----Original Message----- From: Paul Miller [mailto:paul@fxtech.com] Sent: Tue 2/17/2004 12:32 PM To: boost-users(a)lists.boost.org Cc: Subject: [Boost-users] tips from anyone using boost with Qt? I'm in the interesting situation of having to decide whether to use Qt threads or boost threads in this application I am developing. boost threads sound nice because they do not require Qt. Qt threads sound nice because they are integrated more with Qt, and might interroperate better when combined with a UI. Anyone have any advice on the matter? _______________________________________________ Boost-users mailing list Boost-users(a)lists.boost.org http://lists.boost.org/mailman/listinfo.cgi/boost-users ----------------------------------------------------------------------- DISCLAIMER: Information contained in this message and/or attachment(s) may contain confidential information of Zetec, Inc. If you have received this transmission in error, please notify the sender by return email. -----------------------------------------------------------------------

1 0

tips from anyone using boost with Qt?
by Paul Miller 18 Feb '04

18 Feb '04

I'm in the interesting situation of having to decide whether to use Qt threads or boost threads in this application I am developing. boost threads sound nice because they do not require Qt. Qt threads sound nice because they are integrated more with Qt, and might interroperate better when combined with a UI. Anyone have any advice on the matter?

1 0

retrieving boost::variant values to different types
by Andrew Chapman 18 Feb '04

18 Feb '04

Hey everyone, I was just having a look at the new boost::variant vlass, and I was expecting to be able to do something like: boost::variant<int, float> var; var = 5; int i = boost::get<int>(var); // fine float f = boost::get<float>(var); // throws As the comments show, asking for the variant's value using get<int> works fine but get<float> throws an exception as it is an unmatched type. So it seems that before you can retrieve the value of the variant you need to know its type? How do you check what type the variant is currently holding? This would work but is inefficient/ugly: boost::variant<int, float> var; var = 5; std::stringstream str; str << var; int i = atoi(str.str()); float f = atof(str.str()); I was hoping to use the variant class to implement scripting language style variables where you can get/set them to be any type (which seems to work fine) and also retrieve them to any type (though retrieving a string as a bool may not always make too much sense). Any suggestions? -- Andrew Chapman 3D Technical Director - Framestore CFC

1 0