Recently I posted a comment on comp.lang.c++.moderated about how I used boost::call_traits to rewrite the standard binders and member function adapters, and...
2984
Dave Abrahams
abrahams@...
May 2, 2000 10:41 am
... go for it....
2985
John Maddock
John_Maddock@...
May 2, 2000 10:54 am
Mark, ... used boost::call_traits to rewrite the standard binders and member function adapters, and thus avoided the problem of references to references. ...
2986
John Maddock
John_Maddock@...
May 2, 2000 10:56 am
Dave, ... arguments: it's backward compatibility and some as-yet-unidentified usefulness for generic programming vs. simplicity and predictability. < As I see...
2987
Dave Abrahams
abrahams@...
May 2, 2000 12:10 pm
... I guess I didn't make the arguments clearly enough. The choices are between what Alexy supplied and something similar but slightly simpler. The simpler ...
2988
Gary Powell
Gary.Powell@...
May 2, 2000 2:23 pm
... The lambda library (LL) function "bind" does this, and more. While I am interested in other solutions, you ought to take a look at this one first. There is...
2989
Beman Dawes
beman@...
May 2, 2000 3:31 pm
Nine people from six countries (Canada, Denmark, Germany, Japan, UK, US) discussed boost topics during dinner at a Russian restaurant in Tokyo, Japan, on 18...
2990
Gary Powell
Gary.Powell@...
May 3, 2000 4:39 pm
Question: Should a library in the vault or on the boost web page in general require other boost libraries? Other than reusing boost/config.hpp, as a library...
2991
jsiek@...
May 3, 2000 6:04 pm
I think this is a very important issue, because reuse is very important. That's the whole point of boost and writing libraries in general. If you don't reuse...
2992
Kevlin Henney
kevlin@...
May 3, 2000 8:05 pm
In message <8epko4+im7r@eGroups.com>, Gary Powell <Gary.Powell@...> writes ... I'm not sure that this is such a useful way of thinking about it. High ...
2993
Daniel Heck
dheck@...
May 3, 2000 8:10 pm
Reply-To: Hi everyone, I'm new to this mailing list and relatively to the STL. To get a little more acquainted with the STL I just wrote a POSIX style command...
2994
Dave Abrahams
abrahams@...
May 4, 2000 2:47 am
... I haven't looked at the code, but I think boost could really benefit from more libraries of a similar scope: handle the common, day-to-day business of ...
2995
Thomas Holenstein
tholenst@...
May 4, 2000 9:41 am
Hello, I've written an alternative implementation of the std::vector class. It provides all fundamental operations (push_back, pop_back, operator[]) in...
2996
Daniel Heck
dheck@...
May 4, 2000 12:34 pm
... I wrote the program under Linux; it seems the STL that comes with gcc is not fully standard compliant. For some strange reason, the STL implementation that...
2997
Daniel Heck
dheck@...
May 4, 2000 12:40 pm
Hi, ... Only one question: How does your class compare to the standard deque<> class? It seems optvec and deque are quite similar in their advantages and ...
2998
Bill Wade
bill.wade@...
May 4, 2000 1:41 pm
... A typical deque will have O(N) worst-case push_back. However the techniques described in http://daisy.uwaterloo.ca/~eddemain/papers/WADS99a/ could be ...
2999
Tim Sharrock
tim@...
May 4, 2000 2:10 pm
... For large collections they are probably similar, but the design goal of low wasted space is not explicit in deque, so the "page size" may well be optimised...
3000
Beman Dawes
beman@...
May 4, 2000 7:31 pm
... command ... below; ... Daniel, There might well be boost interest in a command line parser, but I doubt boost people will take your submission seriously...
3001
Dietmar Kuehl
dietmar_kuehl@...
May 4, 2000 9:13 pm
Hi, I haven't paid close attention to the various discussions on smart pointers but I want to use one in the rewrite of the directory iterator I'm currently...
3002
Jens Maurer
jmaurer@...
May 4, 2000 9:32 pm
... In general, I think a boost command-line parser would be helpful. However, command-line options are usually a once-only thing, so using POSIX getopt (with...
3003
Daniel Heck
dheck@...
May 5, 2000 12:46 am
... You're completely right, but I certainly don't feel like spending too much time on a project noone is interested in. I did that once on another mailing...
3004
Daniel Heck
dheck@...
May 5, 2000 12:46 am
... Sure; it may not be inconvenient, but vanilla POSIX getopt doesn't even support long options. ... I played with several different designs before settling...
3005
Dave Abrahams
abrahams@...
May 5, 2000 1:56 am
... This shouldn't be neccessary. If I understand you correctly, I do this all the time with boost::shared_ptr. I assume you have a class directory_iterator...
3006
Dave Abrahams
abrahams@...
May 5, 2000 5:00 am
Please briefly check out <http://www.python.org/doc/current/lib/module-getopt.html> For an interface which has proven itself to be useful and powerful Regards,...
3007
Kevlin Henney
kevlin@...
May 5, 2000 10:21 am
In message <20000504211352.18595.qmail@...>, Dietmar Kuehl <dietmar_kuehl@...> writes [...] ... [...] ... [...] ... An alternative...
3008
Thomas Holenstein
tholenst@...
May 5, 2000 11:01 am
Hi, ... I think most of the differences are described in Bill's answer. Especially the point that a deque has worst-case O(N) push_back. (With a very low...
3009
Tim Sharrock
tim@...
May 5, 2000 11:30 am
... I see! My interests tend to lie in lots of small collections, rather than a few big ones, so the asymptotic case is only occasionally of interest to me...
3010
Beman Dawes
beman@...
May 5, 2000 3:23 pm
What process should we use for accepting a new boost library? Currently the author posts an preliminary submission in the boost egroups files section (formerly...
3011
John Maddock
John_Maddock@...
May 6, 2000 10:56 am
Beman, ... Sounds like a sound policy, however, I think we should avoid too many proposals at once, how about if there was a "discussion of the week/month", ...
3012
Dave Abrahams
abrahams@...
May 6, 2000 5:40 pm
... I like Beman's general approach. I also agree with John that too many simultaneous libs under consideration could pose problems. Additionally, we need some...
