This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: named warnings & individual warning control
- From: Mark Mitchell <mark at codesourcery dot com>
- To: DJ Delorie <dj at redhat dot com>
- Cc: gcc at gcc dot gnu dot org
- Date: Fri, 02 Jul 2004 07:53:51 -0700
- Subject: Re: named warnings & individual warning control
- Organization: CodeSourcery, LLC
- References: <200406211908.i5LJ8mCX027121@greed.delorie.com> <40D7CF2B.2030405@codesourcery.com> <200406221400.i5ME0QjE002663@greed.delorie.com> <40D85A29.9070503@codesourcery.com> <200406221714.i5MHEM6i005590@greed.delorie.com> <40D86E05.8070805@codesourcery.com> <200406221850.i5MIo2S0007045@greed.delorie.com> <40D88578.6080306@codesourcery.com> <200406242210.i5OMAwbp023722@greed.delorie.com> <40DBB1D7.6070108@codesourcery.com> <200406282056.i5SKu4oL029790@greed.delorie.com> <200407012126.i61LQfvU001865@greed.delorie.com>
DJ Delorie wrote:
Take our existing functions, like warning(), and add a parameter that
indicates the warning group the message is in.
Technical question: There are a couple of places in the source where
we have a function pointer, and set it to one of {warning, error,
pedwarn}. Thus, to add a parameter to warning() means adding a
parameter to pedwarn() (probably a good idea) and error() (probably a
bad idea). Further, we have pedwarn_c90() and pedwarn_c99() which
should probably have the same function signature as pedwarn().
I think passing the parameter to "error" is a good idea too. For
example, in the case that the user requests that diagnostics contain
links back to the relevant standard, you could use the tag passed in to
find that information. (At first, we will not have enough specificity
to do that, but later...)
--
Mark Mitchell
CodeSourcery, LLC
(916) 791-8304
mark@codesourcery.com