Re: [PATCH 1/2] completion: take into account the formatting backticks for options

> > backticked in their definition lists. This patch updates the generation of
> > the completion list to take into account this new format.
> > 
> > Signed-off-by: Jean-Noël Avila <jn.avila@free.fr>
> > ---
> > 
> >  generate-configlist.sh | 4 ++--
> >  1 file changed, 2 insertions(+), 2 deletions(-)
> > 
> > diff --git a/generate-configlist.sh b/generate-configlist.sh
> > index dffdaada8b5..802178daad4 100755
> > --- a/generate-configlist.sh
> > +++ b/generate-configlist.sh
> > @@ -13,8 +13,8 @@ print_config_list () {
> > 
> >  	cat <<EOF
> >  
> >  static const char *config_name_list[] = {
> >  EOF
> > 
> > -	grep -h '^[a-zA-Z].*\..*::$' "$SOURCE_DIR"/Documentation/

> > "$SOURCE_DIR"/Documentation/config/*.adoc | -	sed '/deprecated/d; s/::$//;
> > s/,  */\n/g' |
> > +	grep -h '^`\?[a-zA-Z].*\..*`\?::$'
> > "$SOURCE_DIR"/Documentation/*config.adoc
> > "$SOURCE_DIR"/Documentation/config/*.adoc |
> Using \? in BRE as a short-hand for \{0,1\} (or trying to use any
> single regexp magic in BRE by adding a backslash when that magic is
> only available in ERE) is a GNUism, isn't it?
> 
> We probably should rewrite the thing as ERE, perhaps like
> 
> 	grep -E -h '^`?[a-zA-Z].*\..*`?::$' ...
> 
> Also, if we can avoid piping grep into sed or awk, we should do so,
> when it does not lose readability.  Perhaps we can do something like
> 
> 	sed -E -n -e '/... that pattern .../{
> 		/deprecated/d;
> 		s/::$//;
> 		...
> 	}' "$SOURCE_DIR"/Documentation/*config.adoc ... |
>         sort |
> 	...
> 
> in this case?