abydos.stemmer._snowball.sb_dutch()

Complexity

Conditions

66

Size

Total Lines	137
Code Lines	93

Duplication

Lines	0
Ratio	0 %

Code Coverage

Tests	74
CRAP Score	66

Importance

Changes

0

# -*- coding: utf-8 -*-
def sb_dutch(word):
    """Return Snowball Dutch stem.

    The Snowball Dutch stemmer is defined at:
    http://snowball.tartarus.org/algorithms/dutch/stemmer.html

    :param str word: the word to calculate the stem of
    :returns: word stem
    :rtype: str

    >>> sb_dutch('lezen')
    'lez'
    >>> sb_dutch('opschorting')
    'opschort'
    >>> sb_dutch('ongrijpbaarheid')
    'ongrijp'
    """
    _vowels = {'a', 'e', 'i', 'o', 'u', 'y', 'è'}
    _not_s_endings = {'a', 'e', 'i', 'j', 'o', 'u', 'y', 'è'}

    def _undouble(word):
        """Undouble endings -kk, -dd, and -tt."""
        if (
            len(word) > 1

Wrong hanging indentation before block (add 4 spaces).

            and word[-1] == word[-2]

Wrong hanging indentation before block (add 4 spaces).

            and word[-1] in {'d', 'k', 't'}

Wrong hanging indentation before block (add 4 spaces).

        ):
            return word[:-1]
        return word

    # lowercase, normalize, decompose, filter umlauts & acutes out, and compose
    word = normalize('NFC', text_type(word.lower()))
    _accented = dict(zip((ord(_) for _ in 'äëïöüáéíóú'), 'aeiouaeiou'))
    word = word.translate(_accented)

    for i in range(len(word)):

Consider using enumerate instead of iterating with range and len

        if i == 0 and word[0] == 'y':
            word = 'Y' + word[1:]
        elif word[i] == 'y' and word[i - 1] in _vowels:
            word = word[:i] + 'Y' + word[i + 1 :]
        elif (
            word[i] == 'i'

Wrong hanging indentation before block (add 4 spaces).

            and word[i - 1] in _vowels

Wrong hanging indentation before block (add 4 spaces).

            and i + 1 < len(word)

Wrong hanging indentation before block (add 4 spaces).

            and word[i + 1] in _vowels

Wrong hanging indentation before block (add 4 spaces).

        ):
            word = word[:i] + 'I' + word[i + 1 :]

    r1_start = max(3, _sb_r1(word, _vowels))
    r2_start = _sb_r2(word, _vowels)

    # Step 1
    if word[-5:] == 'heden':
        if len(word[r1_start:]) >= 5:
            word = word[:-3] + 'id'
    elif word[-3:] == 'ene':
        if len(word[r1_start:]) >= 3 and (
            word[-4] not in _vowels and word[-6:-3] != 'gem'

Wrong hanging indentation before block (add 4 spaces).

        ):
            word = _undouble(word[:-3])
    elif word[-2:] == 'en':
        if len(word[r1_start:]) >= 2 and (
            word[-3] not in _vowels and word[-5:-2] != 'gem'

Wrong hanging indentation before block (add 4 spaces).

        ):
            word = _undouble(word[:-2])
    elif word[-2:] == 'se':
        if len(word[r1_start:]) >= 2 and word[-3] not in _not_s_endings:
            word = word[:-2]
    elif word[-1:] == 's':
        if len(word[r1_start:]) >= 1 and word[-2] not in _not_s_endings:
            word = word[:-1]

    # Step 2
    e_removed = False
    if word[-1:] == 'e':
        if len(word[r1_start:]) >= 1 and word[-2] not in _vowels:
            word = _undouble(word[:-1])
            e_removed = True

    # Step 3a
    if word[-4:] == 'heid':
        if len(word[r2_start:]) >= 4 and word[-5] != 'c':
            word = word[:-4]
            if word[-2:] == 'en':
                if len(word[r1_start:]) >= 2 and (
                    word[-3] not in _vowels and word[-5:-2] != 'gem'

Wrong hanging indentation before block (add 4 spaces).

                ):
                    word = _undouble(word[:-2])

    # Step 3b
    if word[-4:] == 'lijk':
        if len(word[r2_start:]) >= 4:
            word = word[:-4]
            # Repeat step 2
            if word[-1:] == 'e':
                if len(word[r1_start:]) >= 1 and word[-2] not in _vowels:
                    word = _undouble(word[:-1])
    elif word[-4:] == 'baar':
        if len(word[r2_start:]) >= 4:
            word = word[:-4]
    elif word[-3:] in ('end', 'ing'):
        if len(word[r2_start:]) >= 3:
            word = word[:-3]
            if (
                word[-2:] == 'ig'

Wrong hanging indentation before block (add 4 spaces).

                and len(word[r2_start:]) >= 2

Wrong hanging indentation before block (add 4 spaces).

                and word[-3] != 'e'

Wrong hanging indentation before block (add 4 spaces).

            ):
                word = word[:-2]
            else:
                word = _undouble(word)
    elif word[-3:] == 'bar':
        if len(word[r2_start:]) >= 3 and e_removed:
            word = word[:-3]
    elif word[-2:] == 'ig':
        if len(word[r2_start:]) >= 2 and word[-3] != 'e':
            word = word[:-2]

    # Step 4
    if (
        len(word) >= 4

Wrong hanging indentation before block (add 4 spaces).

Too many boolean expressions in if statement (6/5)

        and word[-3] == word[-2]

Wrong hanging indentation before block (add 4 spaces).

        and word[-2] in {'a', 'e', 'o', 'u'}

Wrong hanging indentation before block (add 4 spaces).

        and word[-4] not in _vowels

Wrong hanging indentation before block (add 4 spaces).

        and word[-1] not in _vowels

Wrong hanging indentation before block (add 4 spaces).

        and word[-1] != 'I'

Wrong hanging indentation before block (add 4 spaces).

    ):
        word = word[:-2] + word[-1]

    # Change 'Y' and 'U' back to lowercase if survived stemming
    for i in range(0, len(word)):
        if word[i] == 'Y':
            word = word[:i] + 'y' + word[i + 1 :]
        elif word[i] == 'I':
            word = word[:i] + 'i' + word[i + 1 :]

    return word