def test_init_charmap_invalid6(self):
"""Test that an invalid key (neither a single Unicode character nor a
Unicode character range) in charMap raises a InvalidCharMapKeyError.
"""
with pytest.raises(InvalidCharMapKeyError):
CharMapper({u'a--': u'Hello'})
def test_init_charmap_invalid1(self):
"""Test that an invalid key (byte string) type in charMap raises a
TypeError.
"""
with pytest.raises(TypeError):
CharMapper({b'a': 'Hello'})
def test_init_charpap_invalid7(self):
"""Test that an invalid key (neither a single Unicode character nor a
Unicode character range) in charMap raises a InvalidCharMapKeyError.
"""
with pytest.raises(TypeError):
CharMapper({'--a': b'Hello'})
def test_mapstring_empty_string(self):
"""Test that an empty string causes the map_string method to return an
empty string.
"""
mapper = CharMapper(VALID_MAP)
assert mapper.map_string('') == ''
def test_init_default_not_valid2(self):
"""Test that an invalid type (byte string) for default raises a
TypeError.
"""
with pytest.raises(TypeError):
CharMapper({}, b'Hello')
def test_init_charmap_invalid2(self):
"""Test that an invalid value type (byte string) for a valid key
(single Unicode character) in charMap raises a TypeError.
"""
with pytest.raises(TypeError):
CharMapper({'a': b'Hello'})
def test_init_charmap_invalid3(self):
"""Test that an invalid value type (byte string) for an invalid key
(Unicode character range with wrong order) in charMap raises a
InvalidCharMapKeyError.
"""
with pytest.raises(InvalidCharMapKeyError):
CharMapper({'c-a': b'Hello'})
def test_mapstring_not_unicode(self):
"""Test that a non-unicode string causes the map_string method to raise
a TypeError.
"""
with pytest.raises(TypeError):
mapper = CharMapper(VALID_MAP)
mapper.map_string(b'Hello, world!')
def test_mapstring_none(self):
"""Test that a None value causes the map_string method to raise a
TypeError.
"""
with pytest.raises(TypeError):
mapper = CharMapper(VALID_MAP)
mapper.map_string(None)
def test_init_charmap_invalid4(self):
"""Test that an invalid value type (byte string) for an invalid key
(neither a single Unicode character nor a Unicode character range) in
charMap raises a InvalidCharMapKeyError.
"""
with pytest.raises(InvalidCharMapKeyError):
CharMapper({'cdsn': b'Hello'})
def test_init_none(self):
"""Test that init with None raises a TypeError.
"""
with pytest.raises(TypeError):
CharMapper(None)
def test_mapstring_arabic(self):
"""Test that a map_string properly maps an Arabic unicode string.
"""
mapper = CharMapper(VALID_MAP)
assert mapper.map_string('٠١٢٣٤٥٦٧٨٩') == '012---++++'
def test_mapstring_english(self):
"""Test that a map_string properly maps an English unicode string.
"""
mapper = CharMapper(VALID_MAP)
assert mapper.map_string('Hello, world!') == 'Hu**o, wor*m!'
# Identify No Analysis marker
_NOAN_RE = re.compile(u'NOAN')
_COPY_FEATS = frozenset([
'gloss', 'atbtok', 'atbseg', 'd1tok', 'd1seg', 'd2tok', 'd2seg', 'd3tok',
'd3seg'
])
_UNDEFINED_LEX_FEATS = frozenset(['root', 'pattern', 'caphi'])
DEFAULT_NORMALIZE_MAP = CharMapper({
u'\u0625': u'\u0627',
u'\u0623': u'\u0627',
u'\u0622': u'\u0627',
u'\u0671': u'\u0627',
u'\u0649': u'\u064a',
u'\u0629': u'\u0647',
u'\u0640': u''
})
""":obj:`~camel_tools.utils.charmap.CharMapper`: The default character map used
for normalization by :obj:`CalimaStarAnalyzer`.
Removes the tatweel/kashida character and does the following conversions:
- 'إ' to 'ا'
- 'أ' to 'ا'
- 'آ' to 'ا'
- 'ٱ' to 'ا'
- 'ى' to 'ي'
- 'ة' to 'ه'
def test_init_default_valid2(self):
"""Test that a Unicode string type for default doesn't raise an
Exception.
"""
assert CharMapper({}, 'Hello')
def test_init_not_dict(self):
"""Test that a non-dict object (list) raises a TypeError.
"""
with pytest.raises(TypeError):
CharMapper([])
from __future__ import absolute_import
import pytest
from camel_tools.utils.charmap import CharMapper
from camel_tools.utils.transliterate import Transliterator
# A mapper that translates lower-case English characters to a lower-case x and
# upper-case English characters to an upper-case X. This makes it easy to
# predict what the transliteration should be.
TEST_MAP = {
u'A-Z': u'X',
u'a-z': u'x',
}
TEST_MAPPER = CharMapper(TEST_MAP, None)
class TestTransliteratorInit(object):
"""Test class for Transliterator.__init__.
"""
def test_init_none_mapper(self):
"""Test that init raises a TypeError when given a mapper that is None.
"""
with pytest.raises(TypeError):
Transliterator(None)
def test_init_invalid_type_mapper(self):
"""Test that init raises a TypeError when given a mapper that is not a
CharMapper instance.
def test_init_charmap_valid3(self):
"""Test that a valid charMap doesn't raise an Exception.
"""
assert CharMapper({u'a-f': u''})
def test_init_charmap_valid5(self):
"""Test that a valid charMap doesn't raise an Exception.
"""
assert CharMapper({'--a': ''})
def test_init_charmap_valid4(self):
"""Test that a valid charMap doesn't raise an Exception.
"""
assert CharMapper({'a-f': '', 'b': None}, 'Hello')
def test_init_empty_dict(self):
"""Test that init with an empty dict doesn't raise an exception.
"""
assert CharMapper({})
def test_init_charmap_valid2(self):
"""Test that a valid charMap doesn't raise an Exception.
"""
assert CharMapper({u'a': None})
def test_init_dictlike_object(self):
"""Test that init with an dict-like object doesn't raise an exception.
"""
assert CharMapper(AnotherMapping())
def test_init_default_valid1(self):
"""Test that a None type for default doesn't raise an Exception.
"""
assert CharMapper({}, None)
def test_init_default_not_valid1(self):
"""Test that an invalid type (list) for default raises a TypeError.
"""
with pytest.raises(TypeError):
CharMapper({}, [])
"""This module provides functions for normalizing Arabic text.
"""
import re
import unicodedata
from camel_tools.utils.charmap import CharMapper
_ALEF_NORMALIZE_BW_RE = re.compile(u'[<>{|]')
_ALEF_NORMALIZE_SAFEBW_RE = re.compile(u'[IOLM]')
_ALEF_NORMALIZE_XMLBW_RE = re.compile(u'[IO{|]')
_ALEF_NORMALIZE_HSB_RE = re.compile(u'[\u0102\u00c2\u00c4\u0100]')
_ALEF_NORMALIZE_AR_RE = re.compile(u'[\u0625\u0623\u0671\u0622]')
_UNICODE_CHAR_FIX = CharMapper({
'\ufdfc': 'ريال',
'\ufdfd': 'بسم الله الرحمن الرحيم',
})
def normalize_unicode(s, compatibility=True):
"""Normalize Unicode strings into their canonically composed form or
(i.e. characters that can be written as a combination of unicode characters
are converted to their single character form).
Note: This is essentially a call to :func:`unicodedata.normalize` with
form 'NFC' if **compatibility** is False or 'NFKC' if it's True.
Args:
s (:obj:`str`): The string to be normalized.
compatibility (:obj:`bool`, optional): Apply compatibility
decomposition. Defaults to True.
