Re: Patch to move warning for X<=Y<=Z from -W to -Wparentheses

["Joseph S. Myers" <jsm at polyomino dot org dot uk>]

On Fri, 6 Aug 2004, Gerald Pfeifer wrote:

> My reasoning was/is that users may need to adjust their Makefiles to get
> the original behavior they may have come to rely upon.

Are you thinking of users using -Wall but not -W (and not expecting this
warning), or -W but not -Wall (and wanting this warning)?

(As my currently preferred warning options to use with 3.4 for general
purposes are -Wall -W -Wstrict-prototypes -Wmissing-prototypes
-Wold-style-definition -Wshadow -Wpointer-arith -Wcast-align -Wcast-qual
-Wwrite-strings -Wredundant-decls -Wnested-externs -Wmissing-noreturn
-Wformat-security -Wmissing-format-attribute -Winit-self, I tend to think
of moves between -Wall and -W as simply being bug fixes, making the
warnings follow more closely the definition of -Wall, rather than of
greater significance in changing user expectations, in particular assuming
that if -W is used then almost surely -Wall is as well.)

> In my opinion, everything that requires a user to edit a Makefile or 
> source code is a good candidate for being mentioned in the release notes.
> 
> Larger accomplishments, either as part of a specific, focused project, or
> long term committment, merit mention on the front page.  Examples include
> things like tree-ssa, new backends, major advances in optimization or
> standards compliance
> 
> gcc-announce for new releases and announcements of frontends or backends
> being dropped, for example.

These look like reasonable descriptions to go somewhere on the web pages
(contribute.html?).

-- 
Joseph S. Myers               http://www.srcf.ucam.org/~jsm28/gcc/
    jsm@polyomino.org.uk (personal mail)
    jsm28@gcc.gnu.org (Bugzilla assignments and CCs)