April 2006 - Boost-users - lists.preview.boost.org

[spirit] thread safety / lock freedom
by Tim Blechmann 13 Apr '06

13 Apr '06

hi all, i'm currently looking into the spirit library to create a parser ... it seems to be threadsafe, but it's using thread locks. is it possible to use the library in a lockfree manner? can i be sure, that there will be no lock misses, when i'm not sharing parsers between the threads? thanks in advance ... tim -- TimBlechmann(a)gmx.de ICQ: 96771783 http://www.mokabar.tk All we composers really have to work with is time and sound - and sometimes I'm not even sure about sound. Morton Feldman

2 1

[Xpressive] Baffled about results of search
by Lynn Allan 13 Apr '06

13 Apr '06

<alert comment="xpressive newbie"> vc7.1 xpressive from "vault" as of today (April 11) With the statements below (more or less based on Example 1 and Example 2 in the sample code provided with xpressive), I'm attempting to detect which "group" of subexpressions matched, and getting results that I don't understand (not unexpected for a newbie <g>). std::string altDow( "Alternate days of the week are Tue and Thursday and Sat and Monday." ); sregex rex = sregex::compile( "((Sunday|Sun)" "|(Monday|Mon)" "|(Tuesday|Tue)" "|(Wednesday|Wed)" "|(Thursday|Thu)" "|(Friday|Fri)" "|(Saturday|Sat))" ); The first regex_search sort of works, but I'm unclear what field in the smatch:what variable to use to tell what matched, and some oddities are happening. regex_search( altDow, what, rex ) Length(0): 3 Position(0): 31 what[0]: Tue what[1]: Tue what[2]: what[3]: what[4]: Tue The subsequent searches give valid information about the length and position, but other fields in the smatch:what variable seem invalid. regex_search( altDow.substr(32), what, rex) regex_search( altDow.substr(45), what, rex) Regardless, I'm unclear how to tell which of the "groups" of subexpressions was "hit". With boost::regex there was a way to loop thru the boolean field: what[i].matched and see what "hit", but I am not seeing a comparable accessor. Am I doing something wrong, leaving something out, completely clueless on using xpressive, or is xpressive not intended for this kind of usage? (and if xpressive is appropriate, what statements will provide the intended information?) The intention is to be able to to replace/encode the altDay string to be: Alternate days of the week are <dow=2>Tue</dow> and <dow=4>Thursday</dow> and <dow=6>Sat</dow> and <dow=1>Monday</dow>." ); If it will help, here is a link to the vc7.1 project. http://cleanspeech.sf.net/misc/DayOfWeekTest.zip </alert>

4 10

catch(...) in thread_proxy
by Jeremy Stanley 13 Apr '06

13 Apr '06

I find the catch(...) in thread_proxy to be a real nuisance. The only thing I can do to prevent threads from silently slipping away is to put a catch(...) inside my thread procedure, but when this is hit I have no idea what the exception is or where it was thrown. I could configure the debugger (msvc71) to break whenever an exception is thrown, but then I have to wade through handled exceptions in order to find the real error. There was a discussion on this subject about eight months ago where the consensus seemed to be get rid of the catch(...)--any progress on this since then? I notice the catch(...) is still present in CVS, which isn't encouraging. As a refresher, this approach won't help... try { thread_proc(); } catch(...) { exception_filter(); // ... because I don't have a stack trace here. } This sort of approach could be okay though: if ( swallow_exceptions ) { try { thread_proc(); } catch(...) { whatever(); } } else { thread_proc(); } If swallow_exceptions were on by default, it wouldn't change behavior unless the programmer asked for it, perhaps using an optional parameter in the thread and thread_group constructors.

2 1

High memory requirement of adjacency_list
by Pratyush 13 Apr '06

13 Apr '06

Hi, It seems that adjacency list takes up lot of memory than what I was expecting. Below is the code that I used for testing. int main() { typedef boost::adjacency_list<> graph_t; graph_t g(6045064); } The graph populated is taking about 190MB of memory, i.e., 32 bytes/vertex approx. I was expecting it to take 4 bytes/vertex. Am I doing something wrong? My intent is to use "vec_adj_list_impl" having no properties at all. Thanks & regards, Pratyush

5 8

Release Candiate Download Location?
by Olaf van der Spek 13 Apr '06

13 Apr '06

Hi, What is the location where the current release candidate can be downloaded?

1 0

[format] %d, %u etc. with (unsigned) char types
by Luke Elliott 13 Apr '06

13 Apr '06

I couldn't find anything on this in the list archives or the format docs, but it seems likely to have been asked before so apologies if that is the case. I would expect something like: boost::format fmt("%d"); fmt % char(0); std::cout << fmt; to output "0" given that %d is explicitly specified. This isn't the case (with VC++ 7.1 and GCC 3.4 at least, and presumably all others given this is normal stream behaviour). Obviously it can be worked around by casting to a non-char type, but that seems a shame given the nature of the library and not having to worry about type-safety too much. So, is this by design or accident? If not by design, would it be possible to change the behaviour (specialisation somewhere for (unsigned) char types or something??)? Thanks in advance Luke Elliott.

1 0

Validating file names using regex
by Jean-Sebastien Vachon 13 Apr '06

13 Apr '06

Hi there, I am sure what I am looking for exist somewhere but I haven't been able to find anything that works in boost::regex. What I am looking for is a regular expression that will validate a filename according to the specification of the OS (UNIX systems and Windows). I would be very grateful if someone could provide me with these two expressions (one for Linux and one for Windows). Later,

4 7

[xpressive] Installation prepares .lib files?
by Lynn Allan 13 Apr '06

13 Apr '06

WinXp-SP2 vc7.1 Would it be feasible/advantageous for xpressive to have an installation procedure a'la regex to build a whole series of .lib libraries? That seems to speed up builds quite a bit.

2 1

Iostreams and Tarballs?
by Gareth Foster 13 Apr '06

13 Apr '06

Hello list, I am wanting to manage a file format used by another application in my own application. They have a tar.gz file, renamed to something else. In it, there is an XML file and some .wav sound files. Can I use Boost to handle this? I would want to parse the XML from the archive, and then load a few of the sound files. What I don't want to do is something cheesy like call the tar program from mine! That's'orrible :) Cheers for any help, and code (pseudo or otherwise) would be great. Or just a yes or no you can/can't do that would be a start. Cheers all, Gaz

3 2

[review] promotion traits - result
by Tobias Schwinger 12 Apr '06

12 Apr '06

Dear Boost community, Alexander Nasonov's Promotion Traits have been accepted conditionally into Boost with 4/4 positive votes. The conditions are - an example must be added to the documentation (plus accompanying source code for the example folder), - the reference should be clarified (cv-qualification, bit fields) - further conditions might arise from the merger with the extended library (TypeTraits) and are subject to synchronization with the library maintainer (John Maddock). Thanks to Alexander for the submission and to our reviewers for their work! Regards, Tobias (Review Manager)

1 0