gmane.comp.parsers.spirit.general

Re: problem with simple grammar action.

Joel de Guzman — 2008-12-01T23:27:42

No. They get dropped. There is no _3 there. Only _1 and _2 (assuming there's a >> after '+'). There's no plan to eliminate this behavior. I think I'm lost. I need more explanation. Could you please clarify? Simple examples would help. Regards,

Re: multi_pass and operator--

Carl Barron — 2008-12-01T20:50:11

On Dec 1, 2008, at 3:43 PM, Jeff Flinn wrote: if you are going to convert (begin,end ) to an std::string its easier to do this std::string result(++begin.end); result.erase(result.length()-1); // erase final '"' done!!! ------------------------------------------------------------------------- This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK & win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100&url=/

Re: multi_pass and operator--

Jeff Flinn — 2008-12-01T20:43:04

Yes, your getting an input_iterator from the istream, which doesn't support those operators. You might be able to replace: assert( end - str >= 2 ); with: assert( std::distance(str, end) >= 2 ); You'd have to forego the decrement of the end iterator, and return a substring excluding the final quote. Based on your use of 1.34.1, I assume you're using classic spirit. And based on your choice of var names that you are extracting a quoted string w/out the qoutes, if so you this is what I do: struct QS { typedef std::string result_t; template< typename ScannerT > int operator() ( ScannerT const& scan, result_t& result )const { match<> m = ( lexeme_d[( '"' >> !( (*(c_escape_ch_p-'"'))[assign(result)] ) >> '"' )] ).parse( scan ); return m? m.length() : -1; } }; const functor_parser QS_p; Then use it with a semantic action like so, with the assign actor: some_rule = ... >> QS_p[assign(some_string)] >> ... Jeff ------------------------------------------------------------------------- This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK & win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100&url=/

multi_pass and operator--

John Wilkinson — 2008-12-01T20:14:43

------------------------------------------------------------------------- This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK & win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100&url=/_______________________________________________ Spirit-general mailing list Spirit-general< at >lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/spirit-general

Re: rule actions

ian eyberg — 2008-12-01T19:22:14

wow! thanks a lot! I can't believe I missed that -- I had tried assign(x,y) and that wasn't working but assign_a(x,y) works perfectly fine. Thanks again! Ian On 14:10 Mon 01 Dec , Carl Barron wrote: ------------------------------------------------------------------------- This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK & win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100&url=/

Re: problem with simple grammar action.

Carl Barron — 2008-12-01T19:15:01

On Nov 30, 2008, at 7:01 AM, Joel de Guzman wrote: Do you count the unused_types in determinining the position k for _k ? that is to get stynthesised(c) = stnthesised(b) where c = (a >> '+' b) [_val = _3 /* _val = _2 */] if you eliminate this behavior you lose the abiliity to get partial results without local varaibles, but if ig you keep this and add _syn so that it is the synthetic return value regardles if it is a sequence, or not then the user most often just needs _stn but if h4 wishes the current behavior he uses _0, etc. Seems like the best of both worlds then _1 does not have two meanings anymore and the confusion can be eliminated. [I assume _syn is not in namespace bosst::spirit somewhere else] ------------------------------------------------------------------------- This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK & win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100&url=/

Re: parser compiled with msvc9 is four to fivetimes slower relative to msvc8

Michael Weber — 2008-12-01T19:12:58

It turns out that _SECURE_SCL was indeed the main culprit, as you suggested. Apparently the _SECURE_SCL implementation in VS2008 prevents container constructors from being inlined, leading to the performance decrease. The remainder of my performance degradation was due to a VS2008 project file conversion bug. When it converts a VS2005 project to 2008, it drops your optimization option (http://support.microsoft.com/kb/958148). Once you enable optimization again and force _SECURE_SCL=0, then you get back to VS2005 performance. Of course, as you point out, there is the nasty ABI impact to deal with when setting _SECURE_SCL=0. thx for your help, Michael ------------------------------------------------------------------------- This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK & win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100&url=/

Re: rule actions

Carl Barron — 2008-12-01T19:10:40

On Dec 1, 2008, at 12:42 PM, ian eyberg wrote: Assign or assign_a ?? assign_a has two forms assign_a(x) and assign_a(x,y) // last one does x = y. see the complete list in the docs. Beware that they are depreciated I believe. ------------------------------------------------------------------------- This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK & win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100&url=/

rule actions

ian eyberg — 2008-12-01T17:42:22

Hi, I've been going through the documentation on the website on rule actions, and still am not understanding them fully. I have an rule that is something like this: myrule = str_p("blahtxt")[assign(self.action.name)]; this correctly throws 'blahtxt' into my action.name variable I'm looking to do something like this: myrule = str_p("blahtxt")[self.action.name = "myblah"]; where self.action.name would now hold "myblah" Surely, I'm just overlooking something obvious here. Thanks, -ian ------------------------------------------------------------------------- This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK & win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100&url=/

[spirit2] compile time

Joel de Guzman — 2008-12-01T15:50:38

Hi, I'm sure you all know that I am tweaking spirit2 inside out to squeeze more compile time performance. Here are some numbers so far. The most complex example in the spirit2 examples folder is the mini-C compiler (compiles a very simple C like language to a virtual machine). It is as close to a real world application while still maintaining simplicity for the purpose of instruction. Some numbers: spirit2 (in boost trunk): 2 minutes, 37.9 seconds spirit2x (experimental in spirit svn): 1 minute, 46.2 seconds What's interesting about the mini-C compiler is that I tried as much as possible to apply all tricks (***) to make it build faster, like e.g. composing separate grammars, separating the grammar definitions (the constructors) in separate cpp files, explicit instantiation of the templates, etc. So, while the original 2 minutes, 38 seconds is not bad at all to begin with (considering the /relative/ complexity of the grammar(s)), now we have a definite speedup. I'm quite happy. Let's hope we gain some more. Hardware Specs: Core-2 Duo, 2.4Ghz, 4GB memory Software Specs: MSVC 9.0 on Windows XP (***) I'll try to document the tricks. Call it "spirit: best practice", if you will. The Mini_C compiler has 25 rules, complete with semantic actions. The rules have varying complexity from the simplest one-liner to a 28 line rule. Regards,

Re: [karma] streaming empty containers

Hartmut Kaiser — 2008-12-01T13:34:21

Ohh, that's probably a bug. I'll look into it asap. Regards Hartmut ------------------------------------------------------------------------- This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK & win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100&url=/ _______________________________________________ Spirit-general mailing list Spirit-general< at >lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/spirit-general

Re: [karma] streaming empty containers

Hartmut Kaiser — 2008-12-01T13:31:37

Yeah, sorry. Was too tired yesterday... Regards Hartmut ------------------------------------------------------------------------- This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK & win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100&url=/

Re: [karma] streaming empty containers

Matthias Vallentin — 2008-12-01T07:12:08

Yeah, haven't even noticed it. There is one other issue I ran into when applying the optional operator. Simply prepending the '-' results in the following compile error: [...]/boost-1_37/boost/spirit/home/karma/operator/list.hpp:62: error: call of overloaded ‘generate(const boost::spirit::component, boost::fusion::nil>&, boost::spirit::karma::detail::output_iterator >, void>&, const boost::fusion::unused_type&, const boost::fusion::unused_type&, boost::spirit::unused_type)’ is ambiguous using the following construct: std::cout << boost::spirit::karma::format( '{' << -(stream % ", ") << '}', v ) << std::endl; Any ideas? Matthias

Re: [karma] streaming empty containers

Joel de Guzman — 2008-12-01T06:02:34

Just to clarify, I think Hartmut meant: -(a % ',') Cheers!

Re: [karma] streaming empty containers

Matthias Vallentin — 2008-12-01T05:47:09

Finally I see the connection to Spirit.Qi. Thanks for these examples. Matthias

Re: [karma] streaming empty containers

Hartmut Kaiser — 2008-12-01T04:05:27

The list generator a << ',' is equivalent to a << *(',' << a), so if you need support for empty containers just make it optional: -(a << ','). The generated output is cut off because the generator failed and it's not possible to unroll output which already has been generated. But the stream should have gotten its ios_base::badbit flag set (or for that matter, if you enable exceptions on the stream, it should have thrown). Regards Hartmut ------------------------------------------------------------------------- This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK & win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100&url=/

Re: [karma] streaming a std::map

Matthias Vallentin — 2008-12-01T02:21:56

Yeah, works like a charm, thanks! The appropriate header is by the way: Matthias

[karma] streaming empty containers

Matthias Vallentin — 2008-12-01T02:02:07

Hi there, how do I reliably stream containers regardless of their size? The following example only works with non-empty containers: std::vector v; std::cout << boost::spirit::karma::format( '{' << (stream % ", ") << '}', v ) << std::endl; If the container is empty the output is {% and does not include a closing bracket. In fact, the entire stream is cut off at the point of the first '{'. Of course, I could manually check if the container is empty but it would be sweet if karma can handle empty containers as well (output: "{}"). Could it be that I am misusing karma here? Matthias

Re: Simple suggestion for BOOST_FUSION_ADAPT_STRUCT

Joel de Guzman — 2008-12-01T00:33:31

Sure. That's planned. I'm thinking of a simpler name though: BOOST_FUSION_STRUCT. Patches welcome. This should be easy to do ;-) Regards,

Simple suggestion for BOOST_FUSION_ADAPT_STRUCT

Raindog — 2008-12-01T00:10:11

Perhaps for structures whose main purpose is to be used w/ fusion, a macro can be used that will both define the structure and adapt it so that: struct foo { int var1; int var2; std::string name; std::string name2; }; BOOST_FUSION_ADAPT_STRUCT( foo, (int, var1) (int, var2) (std::string, name) (std::string, name2) ) Can be shorted to something like: BOOST_FUSION_DEFINE_AND_ADAPT_STRUCT( foo, (int, var1) (int, var2) (std::string, name) (std::string, name2) ) ------------------------------------------------------------------------- This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK & win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100&url=/

Re: [karma] streaming a std::map

Hartmut Kaiser — 2008-11-30T22:10:53

If you want to use the stream primitive you need to provide a operator<<() for the map's value_type. But, since std::pair is a fusion sequence (at least if you include the proper fusion header for that), you could write for instance: out << karma::format( '{' << ((int_ << lit) % ", ") << '}', container ); if the type of the map is map. HTH Regards Hartmut ------------------------------------------------------------------------- This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK & win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100&url=/