home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.lang.c++
- Path: sparky!uunet!uunet.ca!frumious!pat
- From: pat@frumious.uucp (Patrick Smith)
- Subject: Re: Destruction of temporaries
- Message-ID: <1992Sep2.042856.571@frumious.uucp>
- Date: Wed, 2 Sep 1992 04:28:56 GMT
- Reply-To: uunet.ca!frumious!pat
- References: <1992Aug29.184025.328@frumious.uucp> <rmartin.715267769@thor> <23598@alice.att.com>
- Organization: None
- Lines: 34
-
- ark@alice.UUCP () writes:
- |The unfortunate reality is that I have seen a number of development
- |projects that have an inflexible rule -- no warnings allowed! That is,
- |they will not allow their people to ship any code that generates warnings.
-
- I am one of those people who believe it is worth going to a lot of
- effort to ensure that code compiles with no warnings at all.
- That way, new warnings stick out like a sore thumb, instead of
- being lost in a mass of old warnings that you know don't indicate
- real problems.
-
- But promoting this philosophy to an 'inflexible rule' is, as you
- said, unfortunate.
-
-
- |Warnings had better be right almost all the time.
-
- Actually, what I would want is
-
- - Warnings should often indicate real problems.
-
- - Whenever code generates a warning but doesn't contain a real
- problem, it should be possible to rewrite the code so it doesn't
- generate a warning. Preferably with no loss of efficiency.
-
-
- | --Andrew Koenig
- | ark@europa.att.com
-
-
- --
- Patrick Smith
- uunet.ca!frumious!pat
- pat%frumious.uucp@uunet.ca
-