Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]
Groups > comp.lang.c++ > #12857 > unrolled thread
| Started by | izhak <iz.hak@mail.ru> |
|---|---|
| First post | 2011-12-29 16:55 +0000 |
| Last post | 2012-01-07 23:19 -0800 |
| Articles | 5 — 3 participants |
Back to article view | Back to comp.lang.c++
Why the iterator_traits is needed? izhak <iz.hak@mail.ru> - 2011-12-29 16:55 +0000
Re: Why the iterator_traits is needed? "Bo Persson" <bop@gmb.dk> - 2011-12-29 17:58 +0100
Re: Why the iterator_traits is needed? izhak <iz.hak@mail.ru> - 2011-12-29 17:10 +0000
Re: Why the iterator_traits is needed? 88888 Dihedral <dihedral88888@googlemail.com> - 2011-12-29 16:42 -0800
Re: Why the iterator_traits is needed? 88888 Dihedral <dihedral88888@googlemail.com> - 2012-01-07 23:19 -0800
| From | izhak <iz.hak@mail.ru> |
|---|---|
| Date | 2011-12-29 16:55 +0000 |
| Subject | Why the iterator_traits is needed? |
| Message-ID | <jdi623$8s6$1@speranza.aioe.org> |
Why do we need the iterator_traits if it's just a simple redefinition of some of the Iterator concept's members?
[toc] | [next] | [standalone]
| From | "Bo Persson" <bop@gmb.dk> |
|---|---|
| Date | 2011-12-29 17:58 +0100 |
| Message-ID | <9m3kgvF7qfU1@mid.individual.net> |
| In reply to | #12857 |
izhak wrote: > Why do we need the iterator_traits if it's just a simple > redefinition of some of the Iterator concept's members? It's not just that. There is also a specialization for pointer, which doesn't have the concept members. Pointers are also iterators! Bo Persson
[toc] | [prev] | [next] | [standalone]
| From | izhak <iz.hak@mail.ru> |
|---|---|
| Date | 2011-12-29 17:10 +0000 |
| Message-ID | <jdi6u6$brl$1@speranza.aioe.org> |
| In reply to | #12858 |
On Thu, 29 Dec 2011 17:58:14 +0100, Bo Persson wrote: > There is also a specialization for pointer, which doesn't have the > concept members. > > Pointers are also iterators! > > > Bo Persson Yes, now I understand... And for any other type, we my just specialize it, to make that type usable in stl. Clear, thanks.
[toc] | [prev] | [next] | [standalone]
| From | 88888 Dihedral <dihedral88888@googlemail.com> |
|---|---|
| Date | 2011-12-29 16:42 -0800 |
| Message-ID | <28877841.460.1325205752939.JavaMail.geo-discussion-forums@prgi2> |
| In reply to | #12857 |
izhak於 2011年12月30日星期五UTC+8上午12時55分31秒寫道： > Why do we need the iterator_traits if it's just a simple redefinition of > some of the Iterator concept's members? I think to understand an iterator is important. There are implementations of iterators. Iterators are different from recursions in implementations. In the abstract level the two does not differ too much.
[toc] | [prev] | [next] | [standalone]
| From | 88888 Dihedral <dihedral88888@googlemail.com> |
|---|---|
| Date | 2012-01-07 23:19 -0800 |
| Message-ID | <15697836.1161.1326007163127.JavaMail.geo-discussion-forums@prlh23> |
| In reply to | #12869 |
88888 Dihedral於 2011年12月30日星期五UTC+8上午8時42分32秒寫道： > izhak於 2011年12月30日星期五UTC+8上午12時55分31秒寫道： > > Why do we need the iterator_traits if it's just a simple redefinition of > > some of the Iterator concept's members? > > I think to understand an iterator is important. > > There are implementations of iterators. > > Iterators are different from recursions in implementations. > > In the abstract level the two does not differ too much. But iterator in STL is more powerful. For example, there could be many instances of timers in different classes of objects. Each timer is defined clearly for the actions to be triggered in objects that can access the timer in the next timer update event. Thinking in the iterator way and the message driven mechanism is better than the MS call back mechanism.
[toc] | [prev] | [standalone]
Back to top | Article view | comp.lang.c++
csiph-web