home *** CD-ROM | disk | FTP | other *** search
- From: hough@aai.com (Al Hough)
- Message-ID: <9601242252.AA08579@fermat.aai.com>
- X-Original-Date: Wed, 24 Jan 96 17:52:15 EST
- Path: in2.uu.net!bounce-back
- Date: 25 Jan 96 05:29:36 GMT
- Approved: fjh@cs.mu.oz.au
- Organization: -
- Newsgroups: comp.std.c++
- Subject: STL sets have mutable iterators?
- Reply-To: hough@aai.com
- X-Auth: PGPMoose V1.1 PGP comp.std.c++
- iQBFAgUBMQcVS+EDnX0m9pzZAQEnvgF6AuTVk3AyTG/L4cYlHN1a4smuc8e2WEja
- +9MKk/OaH2FT9dBuEYdu/LoLtWLWlHSX
- =vAcz
-
- In my April 95 copy of the DWP, I couldn't find any requirement
- that a set iterator be a const_iterator. Previous STL documentation
- required that for sets, an iterator was a const_iterator, thus
- insuring that the return type from dereferencing a set<T>::iterator
- was a const T& and not a T&. I agreed with the behavior that a
- set iterator should yield const T&, because yielding a T& would
- allow users to break the "uniqueness" of the set by allowing the
- possiblility that a user could modify an existing value in the
- set so that it became equal to another value in the set.
-
- What was intended here?
-
- --
- Al Hough email: hough@aai.com
- Amerinex Artificial Intelligence phone: (413) 256-8941
- ---
- [ comp.std.c++ is moderated. Submission address: std-c++@ncar.ucar.edu.
- Contact address: std-c++-request@ncar.ucar.edu. The moderation policy
- is summarized in http://dogbert.lbl.gov/~matt/std-c++/policy.html. ]
-