home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!mcsun!uknet!cf-cm!ted
- From: Ted.W.Lawson@cm.cf.ac.uk (Ted Lawson)
- Newsgroups: comp.lang.eiffel
- Subject: Re: "Rename" considered hazardous
- Message-ID: <1992Nov23.210505.14436@cm.cf.ac.uk>
- Date: 23 Nov 92 21:05:03 GMT
- References: <1992Nov23.135527.7035@th-ilmenau.RZ.TH-Ilmenau.DE>
- Sender: news@cm.cf.ac.uk (Network News System)
- Organization: University of Wales College of Cardiff
- Lines: 19
-
- In article <1992Nov23.135527.7035@th-ilmenau.RZ.TH-Ilmenau.DE> huebner@PrakInf.TH-Ilmenau.DE (Huebner) writes:
- >
- >Handling the problem of name conflicts arising from multiple inheritance
- >by renaming could be a nice solution. But it seems that there should be
- >some restrictions to ensure the reliability of the produced software.
-
- (original example omitted for brevity)
-
- > z.foo(x); -- o.k.
- > z.foo(y); -- Error: message "f" not understood by object of Typ B
-
- "It is important to understand that renaming does not change any of the
- inherited features, but simply changes the names under which those features
- will be known by the _clients_ and _descendents_ of a class." [ETL p 83]
-
- The above error doesn't happen because C is neither a client nor a descendant
- of B.
-
- Ted
-