gram.p6

#÷ [0.2] SPACE (Other) ÷ [999.0] SPACE (Other) ÷ [0.3]

# Default Grapheme Break Test
#
# Format:
# <string> (# <comment>)? 
#  <string> contains hex Unicode code points, with 
#	÷ wherever there is a break opportunity, and 
#	× wherever there is not.
#  <comment> the format can change, but currently it shows:
#	- the sample character name
#	- (x) the Grapheme_Cluster_Break property value for the sample character
#	- [x] the rule that determines whether there is a break or not
grammar GraphemeBreakTest {
    token TOP { [<.ws> [<break-nobreak>] <.ws>]+ % <hex> <comment> }
    token break-nobreak {
        <nobreak> | <break>
    }
    token hex {
        <:AHex>+
    }
    token nobreak {
        '÷'
    }
    token break {
        '×'
    }
    token comment {
        '#' .* $
    }
}
class parser {
    method TOP ($/) {
        say $/.list
    }
    #method hex ($/) {
    #    say :16(~$/);
    #}
}

GraphemeBreakTest.new.parse(
    '÷ 0021 ÷ 0020 ÷	#  ÷ [0.2] SPACE (Other) ÷ [999.0] SPACE (Other) ÷ [0.3]'
    , actions => parser.new
).made;