Although it seems impossible to create a non-intrusive weak pointer class that can track the liveness of an arbitrary class of object, it is possible to create...
478
Dave Abrahams
abrahams@...
Aug 1, 1999 5:54 pm
... Fiendishly clever! Questions/observations 1. How do you create a weak_ptr? I presume its constructor would need to take a shared_ptr parameter... 2. This...
479
Greg Colvin
gcolvin@...
Aug 1, 1999 6:24 pm
From: Dave Abrahams <abrahams@...> To: <boost@egroups.com> Sent: Sunday, August 01, 1999 11:58 AM Subject: [boost] Re: live_ptr ... Yes. ... Right,...
480
Alan Griffiths
alan@...
Aug 1, 1999 9:14 pm
In message <000801bedc3e$fb008170$37781990@...>, Greg Colvin <gcolvin@...> writes ... There is an existing, tested implementation along...
481
Andy Glew
glew@...
Aug 2, 1999 12:57 am
Although it seems impossible to create a non-intrusive weak ... Good solution. ... However, I am somewhat astonished by this statement. My background is in...
482
Greg Colvin
gcolvin@...
Aug 2, 1999 1:08 am
From: Andy Glew <glew@...> ... So was I. The tradeoff seems to be that initial access to the pointed to object is slower, but the shared_ptr is...
483
Andy Glew
glew@...
Aug 2, 1999 3:28 am
... Not much. P6 systems only lose 30 cycles to 100 cycles on a pointer chasing miss. In future systems this cost will be much higher. I.e. I do not believe...
484
Greg Colvin
gcolvin@...
Aug 2, 1999 3:45 am
... Not a bad idea. Let me know when you have one we can use. In the mean time, I'm left with guessing about the typical uses of smart pointers, and the...
485
Beman Dawes
beman@...
Aug 3, 1999 12:06 pm
... death ... outlined ... editing it a ... library. ... My first response to Kevlin was perhaps unfairly negative. Maybe we ought to explore the alternatives...
486
Kevlin Henney
Kevlin.Henney@...
Aug 3, 1999 7:57 pm
Beman Dawes wrote: While the template approach seems conceptually more elegant, I find the resulting user code unbearably ugly. The typedef and constant ...
487
Dietmar Kuehl
dietmar.kuehl@...
Aug 3, 1999 9:03 pm
Hi, ... While this is different in C, I think the template approach is the only viable one in C++! Otherwise, how would I write generic code where the number...
488
Greg Colvin
gcolvin@...
Aug 3, 1999 9:19 pm
From: Dietmar Kuehl <dietmar.kuehl@...> ... I assume that numeric_limits<int32>::max() == INT32_MAX? If not, I agree with Deitmar....
489
Jens Maurer
jmaurer@...
Aug 3, 1999 10:27 pm
... We should probably use int_t<> only for getting the correct type and use numeric_limits<> for getting max/min values. The last line would read ...
490
Andy Glew
glew@...
Aug 3, 1999 11:15 pm
... I think the template approach is "better", and "more in the spirit of C++". My only concern is in template depth. ANSI C++ requires implementations nest...
491
Beman Dawes
beman@...
Aug 4, 1999 1:43 am
... advantange. ... many ways ... standard ... That strikes me as better than having a single type with exact/least/fast/etc types embedded. Not sure exactly...
492
Beman Dawes
beman@...
Aug 4, 1999 1:43 am
... My guess is that real compilers will handle any depth we are likely to use, but I don't have any data points to back that up. ... Yes you could, but do you...
493
Beman Dawes
beman@...
Aug 4, 1999 1:43 am
... Yes, good point. ... Another good point! If for 8, 16, 32, [and soon 64] bit types we include also typedefs, then people who want/need shorter forms can...
494
Valentin Bonnard
Bonnard.V@...
Aug 4, 1999 6:09 am
Please have a look at: http://www.eleves.ens.fr:8080/home/bonnard/NewLibs/NumericalTypes/int_chooser.h ...
495
Kevlin Henney
Kevlin.Henney@...
Aug 4, 1999 6:20 am
... No, it is a different type, like uint_least<32>::type, IMO. Agreed. This was exactly the convention I used. Kevlin...
496
Kevlin Henney
Kevlin.Henney@...
Aug 4, 1999 6:20 am
... That strikes me as better than having a single type with exact/least/fast/etc types embedded. Not sure exactly why, but my gut feeling is it would extend...
497
Valentin Bonnard
Bonnard.V@...
Aug 4, 1999 6:26 am
... I am sure why a single type with typedef for (least or fast) -- Valentin Bonnard...
498
Valentin Bonnard
Bonnard.V@...
Aug 4, 1999 6:30 am
I am sure that having a working X server is better than this broken Sun X server, which sends random keypresses to Netscape. ... I am sure why a single type...
499
Andy Glew
glew@...
Aug 5, 1999 3:07 pm
Should Boost be added to the list of C++ libraries at http://www.trumphurst.com/cpplibs/cpplibs.phtml periodically posted as a FAQ to comp.lang.c++? It doesn't...
500
Beman Dawes
beman@...
Aug 5, 1999 3:23 pm
... Yes. There should be a general entry for www.boost.org I suppose. Should there also be entries for individual libraries? That might be nice. ... Would...
501
Andy Glew
glew@...
Aug 5, 1999 3:44 pm
... Sure. That way I'll keep up to date as to what is around. I'd have done it by now, but I wanted to make sure it was okay with y'all....
502
Thomas Plum
tplum@...
Aug 5, 1999 6:34 pm
I want to provide some arguments for making use of, and encouraging the use of, the <stdint.h> as defined in C9X (or indirectly thru a boost-specific...
503
Nathan Myers
ncm@...
Aug 5, 1999 7:34 pm
... I agree with Tom: if boost defines something to do a job similar to C's <stdint.h>, it should do it markedly better, or not at all. An example would be a...
504
Beman Dawes
beman@...
Aug 5, 1999 8:59 pm
... where ... I agree with you as regards the C stdint.h typedefs. In addition to your arguments, it seems to me that the int32_t, int_least32, etc. typedefs...
505
Dietmar Kuehl
dietmar.kuehl@...
Aug 5, 1999 8:59 pm
Hi, ... I have no problem if this header is used to get a fairly portable implementation of the C++ version. I even think that it can be useful for people...
506
Andy Glew
glew@...
Aug 6, 1999 12:39 am
... I have little or no opinion about whether traits or macros should be used, but say this: If standard C uses macros and Boost uses traits then Boost would...
