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

[Wave] error: ill-formed directive
by Andreas Sæbjørnsen 24 May '06

24 May '06

Using the 'samples/cpp_tokens' example preprocessor with the following code: #define true 1 #define false 0 gives the following error: test.C(1): error: ill formed preprocessor directive: #define Preprocessing the same file with 'cpp' give no error. If any of the two lines in the file is removed the other line will cause an error. I guess the error occurs because 'true' is not an T_IDENTIFIER, but a T_TRUE. Could you imagine a clean fix to this issue? Regards. Andreas

2 2

[regex] "EXC_BAD_ACCESS" when iterating over smatch captures
by Christopher Hart 24 May '06

24 May '06

All: I'm trying to run the captures example included in the Boost.Regex documentation. When running the captures example verbatim, I get: "Program received signal: "EXC_BAD_ACCESS"." This is happening on Mac OS X 10.4.6 (Intel) with Boost 1.33.1. This problem seems to be related to the definition of BOOST_REGEX_MATCH_EXTRA - if I don't use captures and leave the macro undefined, other regex functions work fine. >From gdb: (gdb) backtrace #0 0x00003ad4 in boost::sub_match<__gnu_cxx::__normal_iterator<char const*, std::string> >::str () #1 0x00003b3a in boost::operator<< <char, std::char_traits<char>, __gnu_cxx::__normal_iterator<char const*, std::string> > () #2 0x00001ff1 in print_captures () #3 0x000021e6 in main () The instruction that caused the error: 0x00003ad4 <+0068> movsbl (%eax),%eax I'm way out of my depth on this one. Does anyone have any thoughts on what to try? Has anyone been able to use captures on an Intel Mac system, or could this be a platform specific issue? Thanks, Chris

2 3

[BGL] how does automatic vertex_index creation work?
by Sebastian Weber 23 May '06

23 May '06

Hi folks! I defined a customary graph type which always uses integers in a bound range to adress the vertices. It is therefore as if one would use an adjacency_list with vecS as vertex storage set. Since I want to run an algorithm requiring a vertex_index property to be present on my custom graph, I was looking at the code of the adjacency_list graph. The docs say, that somewhere in the code the vertex_index map does get created automatically if the vertex set is vecS, but where and how is it done? The code is just too big or too advanced for me to find the appropiate lines. Could someone please enlighten me? Thanks in advance, Sebastian Weber

2 3

IBM iSeries support
by MARobertson＠dstsystems.com 23 May '06

23 May '06

I was just wondering if anyone out there has tried using any of the Boost libraies on IBM's iSeries platforms. Its one of the platforms that I have to support. I'd like to use a few of the Boost libraries, but am not sure how to proceed. My main concern is the libraries that require a binary( like Boost.Regex ). The header only libraries that I'm interested in, I've been able to use( so far at least ). Marc Robertson Principal Architect AWD Development DST Technologies, Inc. ----------------------------------------- This e-mail and any attachments are intended only for the individual or company to which it is addressed and may contain information which is privileged, confidential and prohibited from disclosure or unauthorized use under applicable law. If you are not the intended recipient of this e-mail, you are hereby notified that any use, dissemination, or copying of this e-mail or the information contained in this e-mail is strictly prohibited by the sender. If you have received this transmission in error, please return the material received to the sender and delete all copies from your system.

2 1

poll of tool usage and machine spec
by Jan Stetka 23 May '06

23 May '06

Would anyone else be interested in a poll, implemented on a website, of what tools e.g. gcc, visual c++ and machine spec: RAM, operating system, CPU boost users have?

4 3

[iostreams] blocking read() confusion
by Dave Dribin 23 May '06

23 May '06

According to the documentation I/O Streams defines blocking I/O as: <http://www.boost.org/libs/iostreams/doc/concepts/blocking.html> > A Device is Blocking if a read request never produces fewer > characters than requested except at end-of-stream, and if a write > request never consumes fewer characters than requested under any > circumstances. So, if I issue an iostream::read() with 10 characters, and only 5 come back, this is considered non-blocking. To me, this is inconsistent with at least Unix read() semantics. As long as iostream::read() blocks for at least 1 character, it should be considered blocking, no? I don't have a copy of the C++ standard handy, but my Stroustrup book just says that iostream::read(Ch *p, streamsize n) "returns at most n char". Thus it heavily implies that reading less than "n char" is not only possible, but not an error or exceptional condition. But the book really doesn't say anything more than that. I am just curious because I am looking into using Boost Iostreams as a wrapper around some socket I/O that uses standard Unix read() semantics. I don't use the socket in non-blocking mode, but since the Unix read() may return less than "n char", do I need to write a loop that ensures exactly "n char" is read, unless an EOF occurs? Thanks, -Dave

3 2

Re: [Boost-users] [Regex] Replace algorithm
by Andrew McDonald 23 May '06

23 May '06

Hi, The iostreams library has a basic_regex_filter that will do this job. See http://www.boost.org/libs/iostreams/doc/classes/regex_filter.html or in context http://www.boost.org/libs/iostreams/doc/index.html 4. Reference -> 4.2 Classes -> basic_regex_filter The following ctor is the one you probably want basic_regex_filter( const basic_regex<Ch, Tr, Alloc>& pattern, const formatter& replace, regex_constants::match_flag_type flags = regex_constants::match_default ); It takes a regex (and regex flags) and for each match calls the formatter function. The formatter is a boost::function1 that takes the match_results and is expected to return a string that will be used to replace the match. You will probably have to put your src/dst strings into stringstreams to use the filter. Note that the library does support a generous additional variety of sinks and sources , files, arrays, zip archives etc. Please note that I recently found a bug in basic_regex_filter in cases where no match is found. See my earlier posts for a fix. regards, Andrew > -----Original Message----- > From: boost-users-bounces(a)lists.boost.org > [mailto:boost-users-bounces@lists.boost.org]On Behalf Of Johan Nilsson > Sent: Monday, 22 May 2006 7:53 PM > To: boost-users(a)lists.boost.org > Subject: [Boost-users] [Regex] Replace algorithm > > > Hi, > > I'd like to replace all parts of a string with the results of > calling a > function, passing the match as a parameter. Some pseudo-code: > > --- > std::string replacer(smatch& s) > { > return std::getenv(s[1].c_str()); > } > > std::string const input = "This is ${USER} on ${HOST}"; > boost::regex const re("\\$\\{([^\\}]+)\\}"); > std::string result = boost::regex_replace(input, re, > boost::bind(&replacer)); > > assert("This is Me on MyWorkstation" == result); > --- > > (I'm sure there are errors above, but I hope you get the > general idea). > > Any easy way to do something similar? > > // Johan > > > > _______________________________________________ > Boost-users mailing list > Boost-users(a)lists.boost.org > http://lists.boost.org/mailman/listinfo.cgi/boost-users >

1 0

Boost Format Problem with empty strings
by Catalin Pitis 23 May '06

23 May '06

Hi I'm using Visual C++ 8.0 and I tried to run the following code, which seems to crash with assertion (NULL POINTER) at debug time: (boost::format( "Something %1%") % std::string()).str() It seems that it doesn't like empty strings in placeholders at all. Is it a bug or is it as designed? Thanks Catalin

2 1

Installing boost 1.33.1 with STLPort 5.0.2 and vc8
by jon 23 May '06

23 May '06

Hello All, For VC7.1 there is a tool set for 'normal' vc7.1 and one for vc-7_1-stlport. With my 'old' vc 7.1 things worked out fine with the STLPort version. Now I'm switching to vc8 and was wondering how to get boost to compile with STLPort instead of the standard vc8 implementation. I assume that boost will compile against MS-STL implementation if I do something like: bjam "-sTOOLS=vc-8_0" install Thanks in advance... Jón

2 1

Non-intrusive serialization on derived objects without default constructor?
by Will Lee 22 May '06

22 May '06

I've tried to searched for this in the archive and in google, but I can't seem to find the answer. I'm having trouble using boost::serialization with its non-intrusive. In parituclar, I can't really figure out how to serialize derived classes that do not have default constructors in an non-intrusive way. class A; class B : A; class C : A; Suppose B and C do not have a default constructor. And I would like to serialize: A* b = new B(10); A* c = new C("something"); Due to the restriction of the API that I can not change, all I have is the pointer to A. I'm puzzled on how this can be done. From reading the documentation and source code, I think I need to override the load_construct_data and save_construct_data, such as: template<class Archive> inline void load_construct_data(Archive &ar, A *t, const unsigned int vers) { } template<class Archive> inline void save_construct_data(Archive &ar, A *t, const unsigned int vers) { } However, it seems like the memory is allocated already by the archiver with the size of type A before calling the load_construct_data. How then can I allcoate an object with type B or C? There are examples on how this can be done intrusively, but since I don't have control of the source, I really want to do it non-intrusively. Any help is greatly appreciated. Will

1 0