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

bjam build question
by Joe McCay 05 Apr '06

05 Apr '06

Is it possible to specify a clean build from scratch, or do you have manually delete the object files and stuff? jmccay

2 1

'Function String' parser question
by K M 05 Apr '06

05 Apr '06

Hi, I have this parser that parses string that represents a function, (a c++ function prototype to say) CONSIDER: bool parse_info<> x_parser4(char const* str, FunctionSpec& spec) { // some rules... rettype = lexeme_d[ (alpha_p) >> *(alnum_p | space_p | ch_p('*')) ][list_actor(spec)]; // assign function name when match... function = lexeme_d[ (alpha_p | ch_p('_')) >> *(alnum_p | ch_p('_')) ][assign_a(spec.m_Name)]; // other rules.... .... return parse(str,top); } // Here's what my struct looks like: struct FunctionSpec { std::string m_Name; enum eVarType { VAR_VOID, VAR_BOOL, VAR_INT... } eVarType m_RetType; }; In the function rule what i did was just to assign the name to the m_Name of spec In the rettype tule what i did there was to pass the match string along with the struct (spec) to a class (list_actor) using the operator()... Problem is that i can't jsut assign m_RetType to the 'matched string' delivered to list_actor's operator() since this one is an enumerated type; In the code i attached, there is a function that 'maps' a string to eVarType, but i can't call this function within the list_actor's operator() body? i.e. template <typename ActionIterT> operator(...,...) { // can't do here? // tried to bind the function but it didn't work as well... } In summary what i need to do is to assign value for m_RetType and other members of FunctionSpec using the strings i get from parsing, e.g. when i get the string "void" for a match in the rettype rule, m_RetType will be assigned to VAR_VOID. etc. ----- krby

1 0

Failed installing on VC8. Namespace 'boost' not found.
by �� 05 Apr '06

05 Apr '06

Hi. I'm newbie in C++ and i can't install the Boost library. Maybe i'm just doing somethng wrong ? Hepl plz. I have have A Visual Studio 2005 installed into default location. I have python 2.4.3 installed at default location (c:\Python24) => trying to install boost : I'm launching bjam.exe as bjam "-sTOOLS=vc-8_0" install it produces 168 files in c:\boost\lib with total size ~1Gig and produces the c:\Boost\include\boost-1_33_1\boost\ with .hpp's inside. //======================================= then setting VC++ Directorise in Tools>Options>Projects and Solutions in in VC++ 8.0 IDE - adding "c:\Boost\include" and "c:\Boost\include\boost-1_33_1\boost" to Include files - adding "c:\Boost\lib" Library files //======================================= then creating a simple VC++ console project and adding some using/include from examlles #include "stdafx.h" #include <boost/config.hpp> #include <iostream> #include <fstream> #include <boost/graph/graph_traits.hpp> #include <boost/graph/adjacency_list.hpp> #include <boost/graph/dijkstra_shortest_paths.hpp> using namespace boost; int _tmain(int argc, _TCHAR* argv[]) { return 0; } ==> Build == > fatal error C1083: Cannot open include file: 'boost/config.hpp': No such file or directory //================================================= Or in this manner : #include "stdafx.h" using namespace boost; int _tmain(int argc, _TCHAR* argv[]) { return 0; } ==> Build == > error C2871: 'boost' : a namespace with this name does not exist //============================== Any ideas how to make that thing work ?

3 2

Re: [Boost-users] Failed installing on VC8. Namespace 'boost' not found
by Alex Korban 04 Apr '06

04 Apr '06

Hi Just a note about Vinzenz's advice: make sure you don't put the trailing slash into the include path, otherwise Visual Studio might produce weird and seemingly unrelated errors. E.g. put c:\Boost\include\boost-1_33_1 into the include paths, not c:\Boost\include\boost-1_33_1\ Hope this helps. Regards Alex Korban.

1 0

Re: [Boost-users] Idea for equation library
by Sohail Somani 04 Apr '06

04 Apr '06

> -----Original Message----- > From: boost-users-bounces(a)lists.boost.org > [mailto:boost-users-bounces@lists.boost.org] On Behalf Of > Stephen Gross > Is it > worthwhile for me to post the toolkit code on this group? Heck yes.

1 0

Idea for equation library
by Stephen Gross 04 Apr '06

04 Apr '06

I've got an idea for a toolkit that uses boost::lambda to let you write 'equations' on the fly in c++. For instance, consider the following pseudo-code: ================== Equation e = x + y; print e(x -> 5, y -> 10) // Prints 15 e = x - y print e(x -> 5, y -> 10) // Prints -5 Equation d = eq(e) + z print d(x -> 1, y -> 2, z -> 3) // Prints 2 ================== The idea is that you can encapsulate an algebraic function in a standalone 'Equation' instance, and dynamically reassign that equation. It's built off of boost::lambda (of course), but it lets you name the variables rather than just refer to them with the standard placeholders (_1, _2, etc.). Is it worthwhile for me to post the toolkit code on this group? Thanks, --Steve Stephen Gross Case Western School of Medicine Cleveland, OH "By Grabthar's hammer, by the sons of Worvan, you shall be avenged." - Dr. Lazarus

1 0

using random library
by Kamil Renczewski 04 Apr '06

04 Apr '06

Hello I have a problem. I need to write program, whitch interprets arguments given at runtime and create generators from this arguments. Then i have to use this generators (by calling operator() method) to generate numbers. Problem is that i dont know type of generators at compile time and i dont know how to store them in struct or class or even pass to a function. Here is en example of my thinking: i run program like this: generators N 10 2 E 2.3 and i would like to get some ramdom numbers from a normal_distribution<>(10, 2) and exponential_distribution<>(2.3) another run of a program might be: generators U -10 30 G 0.3 whitch should give me an uniform_int<>(-10, 30) and geometric_distribution<>(0.3) . Any ideas how to do this? thanks

2 1

Building 1.33.1 with Sun Studio 11
by paul.floyd 04 Apr '06

04 Apr '06

Hi all I've almost managed to get Boost 1.33.1 to build on Solaris with Sun Studio 11. I'm not quite up to speed yet with all of the Boost MACROS (in particular the WORKAROUND one). Here's what I've done so far: a) Apply all of the latest patches to Sun Studio 11 b) Apply all of the patches at http://blogs.sun.com/roller/page/sga?catname=%2FBoost c) I had a compile problem with readdir_r not being seen. To fix this I changed my bjam command line to add -sSUNPRO_CXX="CC -D_XOPEN_SOURCE=500" (on Solaris 10 you may prefer -D_XOPEN_SOURCE=600) d) In boost/test/impl/unit_test_suite.ipp, the code makes use of std::random_shuffle. AFAICS this uses rand48(), and rand48() is in <stdlib.h> but not in <cstdlib> and std::. To fix this I added #if defined(__SUNPRO_CC) #include <stdlib.h> #endif at line 42. (this looks like a compiler/std lib issue to me) e) I had some errors with size_t not being in the global scope (Sun Studio is the pickiest compiler I know of for this). This comes from boost/config/posix_features.hpp including unistd.h My fix for this was to add #include <cstdlib> using std::size_t; at line 17 of boost/config/platform/solaris.hpp f) I had two problems with libs/iostreams/src/bzip2.cpp f.1) bzlib.h uses FILE as if it were in the global scope, but as per point e), with Sun Studio, when including <cstdio>, it is in std::, so just before including "bzip2.h" (line 18) I added #include <boost/iostreams/filter/bzip2.hpp> #if defined(__SUNPRO_CC) using std::FILE; #endif f.2) Solaris 8 ships with a version of libbz2 that is too old to be compatible with the Boost.Iostreams code. Iostreams doesn't seem to do any checking of the libbz2 version (if any version is available to check). To fix this I downloaded libbz2 from http://www.bzip.org/downloads.html, untarred it to a '3rdparty' directory, built it (make CC=cc CFLAGS=-xO5) and then changed my bjam command line to use -sSUNPRO_CXX="CC -D_XOPEN_SOURCE=500 -I/home/paulf/scratch/boost_1_33_1/3rdparty/bzip2-1.0.3" This leaves me with (for the moment) one compile error: boost/spirit/core/non_terminal/impl/grammar.ipp, line 279 the compiler complains about Error: Initializing definition_t &def = get_definition<DerivedT, ContextT, ScannerT>(self); requires an lvalue At first I thought this might be related to Sun Studio having an issue with name mangling typedefs, but this doesn't seem to be the case. A+ Paul Accédez au courrier électronique de La Poste : www.laposte.net ; 3615 LAPOSTENET (0,34 /mn) ; tél : 08 92 68 13 50 (0,34/mn)

1 0

Re: [Boost-users] Perl Regex Question
by paul.floyd 04 Apr '06

04 Apr '06

> admin(a)geocodenet.com wrote : > > > When I use this regular expression: > > > > ^(*.)a(*.) [snip] > > 1==>M > > 2==>nagement > > This is what you get if you use (.*?) (ungreedy) Or alternatively (and more portable to different regex engines which may not have non-greedy options): ^([^a]*)a(.*) A+ Paul Accédez au courrier électronique de La Poste : www.laposte.net ; 3615 LAPOSTENET (0,34 /mn) ; tél : 08 92 68 13 50 (0,34/mn)

1 0

Confusion on The Directory Structure For serialization library
by Cheok Yan Cheng 04 Apr '06

04 Apr '06

Hello all, I just download boost 1.33.1 and I am building using VC++ 7.1. I am a bit confused on the directory structure of serialization library (1) \bin\boost\libs\serialization\build\libboost_serialization.lib \bin\boost\libs\serialization\build\libboost_wserialization.lib May I know the differences among libboost_serialization.lib and libboost_wserialization.lib directory? What is the meaning of 'w'? (2) \bin\boost\libs\serialization\build\libboost_serialization.lib \bin\boost\libs\serialization\build\boost_serialization.dll First, I thought libboost_serialization.lib are mean for static library, boost_serialization.dll are mean for dynamic library. However, inside the libboost_serialization.lib directory, they are further divided into libboost_serialization.lib\vc-7_1\release\runtime-link-static\threading-multi libboost_serialization.lib\vc-7_1\release\threading-multi <---- (Does this mean this is a dynamic library since it doesn't contains runtime-link-static??? How about boost_serialization.dll) Thank you very much! __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com

2 1