On Tue, 2019-07-02 at 10:08 +0100, Sean Young wrote:
> On Mon, Jul 01, 2019 at 06:38:13PM +0200, Bastien Nocera wrote:
> > Do our best to remove comments from each line we process from the
> > keymap
> > sources, so as to avoid commented duplicates and false positives
> > sneaking in to the keymap definitions.
> >
> > Signed-off-by: Bastien Nocera <had...@hadess.net>
> > ---
> > utils/keytable/gen_keytables.pl | 3 +++
> > 1 file changed, 3 insertions(+)
> >
> > diff --git a/utils/keytable/gen_keytables.pl
> > b/utils/keytable/gen_keytables.pl
> > index 3dc74ba6..d73daf58 100755
> > --- a/utils/keytable/gen_keytables.pl
> > +++ b/utils/keytable/gen_keytables.pl
> > @@ -138,6 +138,9 @@ sub parse_file($$)
> > }
> >
> > if ($read) {
> > + # Remove comments
> > + ~ s#/\*.*?\*/##sg;
> > + ~ s#.*\*/##sg;
>
> This doesn't solve the
>
> /* { 0x800ff40b, KEY_ENTER },
> not used */
>
> case. Or // comments.
Those weren't problems I was aware of, or that I was trying to solve...
The presence of a "0x800ff40b = KEY_ENTER" won't stop the remote
definition from loading, or the remote to work correctly, and //
comments usually don't get merged into the kernel sources.
> > if (m/(0x[\dA-Fa-
> > f]+)[\s\,]+(KEY|BTN)(\_[^\s\,\}]+)/) {
> > $out .= "$1 = \"$2$3\"\n";
> > next;
> > --
> > 2.21.0
