maori.sbl

routines (
    R1 R2 RV
    mark_regions
    verb_sfx
)

externals ( stem )

integers ( pV p1 p2 )

groupings ( v )

stringescapes {}

/* special characters (long vowels using diacritics) */

stringdef a_    '{U+0101}'  // a-macron
stringdef e_    '{U+0113}'  // e-macron
stringdef i_    '{U+012B}'  // i-macron
stringdef o_    '{U+014D}'  // o-macron
stringdef u_    '{U+016B}'  // u-macron
stringdef a"    '{U+00E4}'  // a-diaeresis
stringdef e"    '{U+00EB}'  // e-diaeresis
stringdef i"    '{U+00EF}'  // i-diaeresis
stringdef o"    '{U+00F6}'  // o-diaeresis
stringdef u"    '{U+00FC}'  // u-diaeresis

/* define vowels */

define v 'aeiou{a_}{e_}{i_}{o_}{u_}{a"}{e"}{i"}{o"}{u"}'

define mark_regions as (

    $pV = limit
    $p1 = limit
    $p2 = limit  // defaults

    do (
        gopast v setmark pV
    )
    do (
        gopast v gopast non-v setmark p1
        gopast v gopast non-v setmark p2
    )
)

backwardmode (

    define RV as $pV <= cursor
    define R1 as $p1 <= cursor
    define R2 as $p2 <= cursor

    /* strip passive verb endings */

    define verb_sfx as (
        [substring] among (
            'tia' 'hia' 'ngia' 'kia' 'mia' 'ria' 'whia' 'wh{a_}ia' 'wh{a"}ia' 'ia'
            'ina' 'whina' 'kina'
            'nga' 'na' 'a'
            (RV delete)
        )
    )

)

define stem as (
    do mark_regions
    backwards (
        do verb_sfx
    )
)