On Mon, Feb 13, 2012 at 2:37 PM, Richard Trieu <[email protected]> wrote: > On Mon, Feb 13, 2012 at 1:59 PM, David Blaikie <[email protected]> wrote: >> >> On Mon, Feb 13, 2012 at 1:52 PM, Richard Trieu <[email protected]> wrote: >> > On Mon, Jan 30, 2012 at 10:40 AM, Richard Trieu <[email protected]> >> > wrote: >> >> >> >> On Wed, Jan 25, 2012 at 5:24 PM, Richard Trieu <[email protected]> >> >> wrote: >> >>> >> >>> This patch adds a warning to catch semi-colons after function >> >>> definitions. The motivation of removing extraneous semi-colons is to >> >>> make >> >>> it easier to see which scopes are being ended by a brace. Function >> >>> scopes >> >>> will end with a plain brace. Class scopes will end with brace and >> >>> semi-colon. >> >>> >> >>> class C { >> >>> void F() {}; >> >>> }; >> >>> >> >>> extra-semi.cc:2:14: warning: extra ';' after function definition >> >>> [-Wextra-semi] >> >>> void F() {}; >> >>> ^ >> >>> 1 warning generated. >> >> >> >> >> >> Ping. >> > >> > Ping. >> >> Not to derail this review - but just out of curiosity: do we have a >> more general 'unnecessary ;' warning? or could we make a general one >> that covers all cases where ';' is just redudndant & suggests a >> removal? > > There's a few in the parse diagnostics. They are: > > ext_top_level_semi > warn_cxx98_compat_top_level_semi > ext_extra_struct_semi > ext_extra_ivar_semi > > warn_cxx98_compat_top_level_semi is a warning that is in a diagnostic group > so it must be left separate. The other 3 are Extensions and could probably > be folded in with my diagnostic if I switched mine to Extension as well and > added a select into the message.
Oh, sorry - I wasn't so much concerned with the diagnostic text, as the logic - is there some way we could, in a relatively universal/general way, catch all/most unnecessary semicolons rather than adding it on a case-by-case basis like this? _______________________________________________ cfe-commits mailing list [email protected] http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits
