On Tue, 01 Dec 2009 16:10:18 -0800, Garrett D'Amore <[email protected]> wrote: > James Carlson wrote: >> Rafael Vanoni wrote: >> >>> If the ident's you're referring to are of the form >>> >>> #pragma ident "%Z%%M% %I% %E% SMI" >>> >>> then yes. These are leftovers from Teamware, before we moved to >>> Mercurial. Removal of these are part of our pre integration check (hg >>> pbchk), they are being deleted as files are modified. >>> >> >> I think he's referring to the even older stuff that looks like this: >> >> #ident "%Z%%M% %I% %E% SMI" /* SunOS 4.1 1.6 */ >> >> It's astounding that they'd bother to error out on #ident (why not just >> ignore it already? do you just hate your user base?), but unthinkable >> that they'd do this to the still-in-use #pragma. >> >> If so, then that should go as well. But I'd count any compiler that >> errors out on it without even a command line option to fix things as >> hopelessly broken. >> >> > That said, IMO, we really ought to be nuking these #ident's. > > But yeah, the fact that gcc barfs all over this by default is IMO a bug. > > - Garrett >
Likewise, if cdm's pbchk is already catching #pragma ident, it's clearly a bug for it to ignore #ident. Or are the remaining hundreds of instances actually indicative of files left in dusty corners? In that case, they'd conveniently mark prime targets for a spring cleaning. -Albert _______________________________________________ opensolaris-code mailing list [email protected] http://mail.opensolaris.org/mailman/listinfo/opensolaris-code
