Re: [PATCH] Makefile: fix error when gen TAGS

From: Yang Bai
Date: Mon Mar 12 2012 - 03:51:26 EST


On Mon, Mar 12, 2012 at 3:34 PM, Stephen Boyd <sboyd@xxxxxxxxxxxxxx> wrote:
>
> Ah ok. How about this patch instead? Then we don't duplicate the sed
> statement.
>
> ---->8-------
>
> diff --git a/scripts/tags.sh b/scripts/tags.sh
> index 833813a..b390096 100755
> --- a/scripts/tags.sh
> +++ b/scripts/tags.sh
> @@ -168,7 +168,7 @@ exuberant()
>
>        --regex-dotconfig='/^#?[[:blank:]]*(CONFIG_[[:alnum:]_]+)/\1/'
>
>        # Remove structure forward declarations.
> -       LANG=C sed -i -e '/^\([a-zA-Z_][a-zA-Z0-9_]*\)\t.*\t\/\^struct
> \1;.*\$\/;"\tx$/d' tags
> +       LANG=C sed -i -e '/^\([a-zA-Z_][a-zA-Z0-9_]*\)\t.*\t\/\^struct
> \1;.*\$\/;"\tx$/d' $2
>  }
>
>  emacs()
> @@ -213,9 +213,9 @@ emacs()
>  xtags()
>  {
>        if $1 --version 2>&1 | grep -iq exuberant; then
> -               exuberant $1
> +               exuberant $1 $2
>        elif $1 --version 2>&1 | grep -iq emacs; then
> -               emacs $1
> +               emacs $1 $2
>        else
>                all_sources | xargs $1 -a
>         fi
> @@ -244,11 +244,11 @@ case "$1" in
>
>        "tags")
>                rm -f tags
> -               xtags ctags
> +               xtags ctags $1
>                ;;
>
>        "TAGS")
>                rm -f TAGS
> -               xtags etags
> +               xtags etags $1
>                ;;
>  esac
>
>
> But then with your next patch maybe it would be better to just have a flag?
>
> ----8<-----
>
> diff --git a/scripts/tags.sh b/scripts/tags.sh
> index 833813a..4b1755d 100755
> --- a/scripts/tags.sh
> +++ b/scripts/tags.sh
> @@ -166,9 +166,6 @@ exuberant()
>        all_defconfigs | xargs -r $1 -a                         \
>        --langdef=dotconfig --language-force=dotconfig          \
>
>        --regex-dotconfig='/^#?[[:blank:]]*(CONFIG_[[:alnum:]_]+)/\1/'
> -
> -       # Remove structure forward declarations.
>
> -       LANG=C sed -i -e '/^\([a-zA-Z_][a-zA-Z0-9_]*\)\t.*\t\/\^struct
> \1;.*\$\/;"\tx$/d' tags
>  }
>
>  emacs()
> @@ -233,6 +230,7 @@ if [ "${ARCH}" = "um" ]; then
>        fi
>  fi
>
> +remove_structs=
>  case "$1" in
>        "cscope")
>                docscope
> @@ -244,11 +242,18 @@ case "$1" in
>
>        "tags")
>                rm -f tags
> -               xtags ctags
> +               xtags ctags $1

Why we need a $1 here??
> +               remove_structs=y
>                ;;
>
>        "TAGS")
>                rm -f TAGS
> -               xtags etags
> +               xtags etags $1
The same.
> +               remove_structs=y
>                ;;
>  esac
> +
> +# Remove structure forward declarations.
> +if [ -n $remove_structs ]; then
> +       LANG=C sed -i -e '/^\([a-zA-Z_][a-zA-Z0-9_]*\)\t.*\t\/\^struct
> \1;.*\$\/;"\tx$/d' $1
> +fi
>

This way is better but I don't understand why a $1 is there.
I think it is useless and if you agree with me, i will send a fine patch to you.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/