- Boost-users - lists.boost.org

help building shared library for lib/filesystem on RHEL4 with gcc 3.4.4
by William Deegan 03 Aug '06

03 Aug '06

Greetings, When I do the following: pushd libs/filesystem/build bjam "-sTOOLS=gcc" I only end up with the /boost-1.30.2/libs/filesystem/build/bin/libboost_filesystem.a/gcc/release/runtime-link-dynamic/libboost_filesystem.a Built and no .so What am I missing? Thanks, Bill

2 3

[test] how to use register_exception_translator with auto_test?
by Jörg Becker 03 Aug '06

03 Aug '06

How can I use the the register_exception_translater with the Auto unit test framework? With boost 1.31 / VC7.1 I wrote a dummy test only containing the register_exception_translator call. This works fine, but with boost 1.33 / gcc3.3 this workaround does not work. Any Suggestions? Jörg

3 6

checking validity
by Ovanes Markarian 03 Aug '06

03 Aug '06

Hello all, I have one specific question. Which I am uncertain if it can be solved anyway. Imagine the following class structure: struct A { some_field* x; }; struct B { optional<A> a_; }; struct C { optional<B> b_; }; Now I have an instance of struct C and would like to know if x is a valid pointer. this code is really ugly: if(c.b_) if(c.b_.a_) if(c.b_.a_.x) //do smth ... This will be really unpleasent if the structure is more deep then this one. Which is the case in our project. The intention would be to write everything in one if-statement. It is not possible to implement a bool or ! operator in these structures, since these are not the only fields and these operators would not know which field should be checked... Currently I am thinking about boost parameter library, but still uncertain if this problem can be solved using it... Can some of you make a suggestion what is the best suitable way of resolving this issue and if this issue resolvable anyway. With Kind Regards, Ovanes Markarian

2 1

Re: [Boost-users] checking validity
by james.jones＠firstinvestors.com 03 Aug '06

03 Aug '06

From: "Ovanes Markarian" <om_boost(a)keywallet.com> > Now I have an instance of struct C and would like to know if x is a valid pointer. > this code is really ugly: > if(c.b_) > if(c.b_.a_) > if(c.b_.a_.x) > //do smth ... > The intention would be to write everything in one if-statement. It is not possible > to implement a bool or ! operator in these structures, since these are not the > only fields and these operators would not know which field should be checked... How about just: if(c.b_ && c.b_.a_ && c.b_.a_.x) //do smth ... Is that cheating? You can do it all in one 'if' statement this way. I suppose you could write a macro to do something like this: #define CHECK(x,y,z,w) (x.y && x.y.z && x.y.z.w) and then use it like this: if(CHECK(c, b_, a_, x)) //do smth ... And maybe you could write the macro in such a way as to take arbitrary lists of arguments. But that's the best I can come up with. - James Jones Administrative Data Mgmt. Webmaster 375 Raritan Center Pkwy, Suite A Data Architect Edison, NJ 08837

2 1

[preprocessor] Enumerating sequence of pairs
by Roman Perepelitsa 03 Aug '06

03 Aug '06

Hello. I want to enumerate sequence of pairs but can't figure out how to do it using boost.preprocessor. Here is example. // sequence of pairs #define SEQ (a, b) (c, d) // How can I implement this macro? #define MAGIC(S) ... MAGIC(SEQ) expands to abcd. If I had possibility to convert SEQ to ((a, b)) ((c, d)) problem would be solved easily because it's standard sequence and thus can be enumerated using with BOOST_PP_SEQ_FOLD_LEFT or another boost.preprocessor algorithm. I have implemented macro which converts SEQ to ((a, b)) ((c, d)) but implementation is huge (I copy-pasted a lot of code from boost.preprocessor). Now I need to enumerate sequence of triples and I don't want to copy-paste again. Any tips? Roman Perepelitsa.

2 2

string_algo and utf8 locales
by Maik Beckmann 03 Aug '06

03 Aug '06

Hello list, this is my first post to you :) I've got a problem with handling locales and string_algo. As an example I want so make this string "hää?" uppercase (->"HÄÄ?") by using this piece of code: <code> #include <iostream> #include <locale> #include <string> #include <boost/algorithm/string.hpp> using namespace std; using namespace boost; int main(int argc, char** argv) { string loc_name = "de_DE.utf8"; string str("hää?"); str = to_upper_copy(str,locale(loc_name.c_str())); // ...should give "HÄÄ?" cout << str << endl; // but prints "Hää?" return 0; } </code> As you see I'm trying to use locales, but it doesn't work. What is my mistake?? Thanks in advance, Maik PS: - I'm using gentoo-linux with gcc-4.1.1 - My locales list: LANG=de_DE.utf8 LC_CTYPE="de_DE.utf8" LC_NUMERIC="de_DE.utf8" LC_TIME="de_DE.utf8" LC_COLLATE="de_DE.utf8" LC_MONETARY="de_DE.utf8" LC_MESSAGES="de_DE.utf8" LC_PAPER="de_DE.utf8" LC_NAME="de_DE.utf8" LC_ADDRESS="de_DE.utf8" LC_TELEPHONE="de_DE.utf8" LC_MEASUREMENT="de_DE.utf8" LC_IDENTIFICATION="de_DE.utf8" LC_ALL=de_DE.utf8

2 3

shared_ptr in API
by Philippe Lambert 03 Aug '06

03 Aug '06

Hello, We are a large team of C++ developers, using Boost at large. We have a philosophical issue about shared_ptr usage and I'm looking for external inputs. The 2 clans are opposed about a simple question: "Should we use shared_ptr in the API function signatures of a module (dll or else)". Both clan use a lot of Sutter/Meyers/other gurus and authors citation to support their view. 1- Pros: shared_ptr are everywhere in the current code. It is the best way to prevent resources leaks, it is easy to use, convenient, it is mostly safe. A lot of gurus are saying "use shared_ptr". Preventing them in API will means a lot of .get() in the existing code. 2- Cons: A shared_ptr in the API give the right to the called function to keep a copy (virtually permanently) of the object. It is a bad indication in the contract if the function doesn't need to keep a copy. It tell to the user "take care, I may keep a copy of this and influence his life cycle". If it is not the case, you should tell it in the signature by asking for a const reference or something like it. Also, if the caller have a stack object and want to use an API function which ask for a shared_ptr, he will need to create a copy, or create a shared_ptr with a deactivated destructor... It is not a beautiful thing, both ways. Of course, I thing that the correct solution is in between, but how to state it? If it is a real issue, I would like to read other team opinion, standard, decision algo or policy... If we are technically totally disoriented, I would like to be redirected... Thanks to all Philippe Lambert

3 2

Re: [Boost-users] replace_all_regex on c-string help
by llwaeva＠21cn.com 03 Aug '06

03 Aug '06

> > > I got the string (char*s), which is too big (>10M), from an external > > library, it's too slow to convert the raw string to std::string before > > it is handled by replace_all_regex > > Here's your fundamental problem: regex does not do in-place search and > replace. It is boost::algorithm::replace_all_regex, not boost:regex_replace

3 2

regular expression on table data
by Winson Yung 03 Aug '06

03 Aug '06

Hello all, I have a follow table: OPERATING REVENUES: publishing $ 42,419 $ 44,754 $ 46,203 collegiate marketing and production services 97 ASSOCIATION MANAGEMENT SERVICES 16 wireless 8,883 8,129 7,507 51,302 52,883 53,823 All I know is that OPERATING REVENUES: will be always there, question is how to write a regular expression to capture the total (which is 51,302 here) There might be more/less than four rows in the table. Would really appreciated if anyone has good suggestion on this. /Winson

2 5

[date_time] Warning in date_time/constrained_value.hpp
by Vsevolod Vlaskin 03 Aug '06

03 Aug '06

Hi, We run into a strange compilation warning, which we see on and off in one particular spot in our code: std::string str( /* initialized with a sane value */ ); ... double dbl( boost::lexical_cast<double>( str ) ); return dbl; The warning shows up only in release mode, NOT in debug: /usr/include/boost/date_time/constrained_value.hpp: In function int main(int, char**): /usr/include/boost/date_time/constrained_value.hpp:64: warning: year$D120988$value_ is used uninitialized in this function If we comment definition of dbl, the warning disappears. It is not a big deal, but maybe, someone could shed light on it. Best regards, Vsevolod Vlaskine __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com

1 0