def setUp(self):
session = (
'@UTF8\n'
'@Begin\n'
'@Languages:\tsme\n'
'@Date:\t12-SEP-1997\n'
'@Participants:\tMEM Mme_Manyili Grandmother , '
'CHI Hlobohang Target_Child\n'
'@ID:\tsme|Sesotho|MEM||female|||Grandmother|||\n'
'@ID:\tsme|Sesotho|CHI|2;2.||||Target_Child|||\n'
'@Birth of CHI:\t14-JAN-2006\n'
'@Birth of MEM:\t11-OCT-1974\n'
'@Media:\th2ab, audio\n'
'@Comment:\tall snd kana jmor cha ok Wakachi2002;\n'
'@Warning:\trecorded time: 1:00:00\n'
'@Comment:\tuses desu and V-masu\n'
'@Situation:\tAki and AMO preparing to look at book , '
'"Miichan no otsukai"\n'
'*MEM:\tke eng ? 0_8551\n'
'%gls:\tke eng ?\n'
'%cod:\tcp wh ?\n'
'%eng:\tWhat is it ?\n'
'%sit:\tPoints to tape\n'
'%com:\tis furious\n'
'%add:\tCHI\n'
'*CHI:\tke ntencha ncha . 8551_19738\n'
'%gls:\tke ntho e-ncha .\n'
'%cod:\tcp thing(9 , 10) 9-aj .\n'
'%eng:\tA new thing\n'
'%com:\ttest comment\n'
'*MEM:\tke eng ntho ena e? 19738_24653\n'
'%gls:\tke eng ntho ena e ?\n'
'%cod:\tcp wh thing(9 , 10) d9 ij ?\n'
'%eng:\tWhat is this thing ?\n'
'%sit:\tPoints to tape\n'
'*CHI:\te nte ena . 24300_28048\n'
'%gls:\tke ntho ena .\n'
'%cod:\tcp thing(9 , 10) d9 .\n'
'%eng:\tIt is this thing\n'
'*MEM:\tke khomba\n'
'\tkhomba . 28048_31840\n'
'%gls:\tkekumbakumba .\n'
'%cod:\tcp tape_recorder(9 , 10) .\n'
'%eng:\tIt is a stereo\n'
'@End')
self.reader = InuktitutReader(io.StringIO(session))
self.reader.load_next_record()
self.maxDiff = None
def test_get_actual_alternative_two_alternatives(self):
"""Test get_actual_alternative with 2 alternatives."""
utterance = ('nuutuinnaq [=? nauk tainna] hela nuutuinnaq '
'[=? nauk tainna] .')
actual_output = InuktitutReader.get_actual_alternative(utterance)
desired_output = 'nauk tainna hela nauk tainna .'
self.assertEqual(actual_output, desired_output)
def test_iter_morphemes_multiple_morphemes(self):
"""Test iter_morphemes with 3 morphe-words."""
morph_words = 'VR|malik^follow+VV|liq^POL+VI|gitsi^IMP_2pS'
actual_output = list(InuktitutReader.iter_morphemes(morph_words))
desired_output = [('VR', 'malik', 'follow'),
('VV', 'liq', 'POL'),
('VI', 'gitsi', 'IMP_2pS')]
self.assertEqual(actual_output, desired_output)
def test_get_actual_alternative_single_alternative(self):
"""Test get_actual_alternative with 1 alternative."""
utterance = 'nuutuinnaq [=? nauk tainna]'
actual_output = InuktitutReader.get_actual_alternative(utterance)
desired_output = 'nauk tainna'
self.assertEqual(actual_output, desired_output)
class TestInuktitutReader(unittest.TestCase):
"""Class to test the InuktitutReader."""
def setUp(self):
session = (
'@UTF8\n'
'@Begin\n'
'@Languages:\tsme\n'
'@Date:\t12-SEP-1997\n'
'@Participants:\tMEM Mme_Manyili Grandmother , '
'CHI Hlobohang Target_Child\n'
'@ID:\tsme|Sesotho|MEM||female|||Grandmother|||\n'
'@ID:\tsme|Sesotho|CHI|2;2.||||Target_Child|||\n'
'@Birth of CHI:\t14-JAN-2006\n'
'@Birth of MEM:\t11-OCT-1974\n'
'@Media:\th2ab, audio\n'
'@Comment:\tall snd kana jmor cha ok Wakachi2002;\n'
'@Warning:\trecorded time: 1:00:00\n'
'@Comment:\tuses desu and V-masu\n'
'@Situation:\tAki and AMO preparing to look at book , '
'"Miichan no otsukai"\n'
'*MEM:\tke eng ? 0_8551\n'
'%gls:\tke eng ?\n'
'%cod:\tcp wh ?\n'
'%eng:\tWhat is it ?\n'
'%sit:\tPoints to tape\n'
'%com:\tis furious\n'
'%add:\tCHI\n'
'*CHI:\tke ntencha ncha . 8551_19738\n'
'%gls:\tke ntho e-ncha .\n'
'%cod:\tcp thing(9 , 10) 9-aj .\n'
'%eng:\tA new thing\n'
'%com:\ttest comment\n'
'*MEM:\tke eng ntho ena e? 19738_24653\n'
'%gls:\tke eng ntho ena e ?\n'
'%cod:\tcp wh thing(9 , 10) d9 ij ?\n'
'%eng:\tWhat is this thing ?\n'
'%sit:\tPoints to tape\n'
'*CHI:\te nte ena . 24300_28048\n'
'%gls:\tke ntho ena .\n'
'%cod:\tcp thing(9 , 10) d9 .\n'
'%eng:\tIt is this thing\n'
'*MEM:\tke khomba\n'
'\tkhomba . 28048_31840\n'
'%gls:\tkekumbakumba .\n'
'%cod:\tcp tape_recorder(9 , 10) .\n'
'%eng:\tIt is a stereo\n'
'@End')
self.reader = InuktitutReader(io.StringIO(session))
self.reader.load_next_record()
self.maxDiff = None
def test_get_start_time_start_time_present(self):
"""Test get_start_time for a case a start time existing."""
self.reader.record.dependent_tiers['tim'] = '19301'
actual_output = self.reader.get_start_time()
desired_output = '19301'
self.assertEqual(actual_output, desired_output)
def test_get_start_time_start_time_absent(self):
"""Test get_start_time for a case no start time existing."""
self.reader._dependent_tiers = {
'utt': 'ha be'
}
actual_output = self.reader.get_start_time()
desired_output = ''
self.assertEqual(actual_output, desired_output)
def test_get_end_time(self):
"""Test get_end_time. Result should be empty string."""
actual_output = self.reader.get_end_time()
desired_output = ''
self.assertEqual(actual_output, desired_output)
# Tests for the get_target_alternative-method.
def test_get_actual_alternative_single_alternative(self):
"""Test get_actual_alternative with 1 alternative."""
utterance = 'nuutuinnaq [=? nauk tainna]'
actual_output = InuktitutReader.get_actual_alternative(utterance)
desired_output = 'nauk tainna'
self.assertEqual(actual_output, desired_output)
def test_get_actual_alternative_two_alternatives(self):
"""Test get_actual_alternative with 2 alternatives."""
utterance = ('nuutuinnaq [=? nauk tainna] hela nuutuinnaq '
'[=? nauk tainna] .')
actual_output = InuktitutReader.get_actual_alternative(utterance)
desired_output = 'nauk tainna hela nauk tainna .'
self.assertEqual(actual_output, desired_output)
def test_get_actual_alternative_no_alternatives(self):
"""Test get_actual_alternative with 2 alternatives."""
utterance = 'nuutuinnaq hela nuutuinnaq .'
actual_output = InuktitutReader.get_actual_alternative(utterance)
desired_output = 'nuutuinnaq hela nuutuinnaq .'
self.assertEqual(actual_output, desired_output)
def test_get_actual_alternative_empty_string(self):
"""Test get_actual_alternative with 2 alternatives."""
utterance = ''
actual_output = self.reader.get_actual_alternative(utterance)
desired_output = ''
self.assertEqual(actual_output, desired_output)
# Tests for the get_target_alternative-method.
def test_get_target_alternative_single_alternative(self):
"""Test get_target_alternative with 1 alternative."""
utterance = 'nuutuinnaq [=? nauk tainna]'
actual_output = InuktitutReader.get_target_alternative(utterance)
desired_output = 'nuutuinnaq'
self.assertEqual(actual_output, desired_output)
def test_get_target_alternative_two_alternatives(self):
"""Test get_target_alternative with 2 alternatives."""
utterance = ('nuutuinnaq [=? nauk tainna] hela nuutuinnaq '
'[=? nauk tainna] .')
actual_output = InuktitutReader.get_target_alternative(utterance)
desired_output = 'nuutuinnaq hela nuutuinnaq .'
self.assertEqual(actual_output, desired_output)
# Test for the get_actual_utterance-method.
def test_get_actual_utterance_one_per_inference_type(self):
"""Test get_actual_utterance with 1 example per inference type.
Inference types:
- alternatives
- fragments
- replacement
- shortening
"""
self.reader.record.utterance = \
'nuutuinnaq [=? nauk tainna] mu:(ğ)ça &ab yarasam [: yorosom] .'
actual_output = self.reader.get_actual_utterance()
desired_output = 'nauk tainna mu:ça ab yarasam .'
self.assertEqual(actual_output, desired_output)
def test_get_actual_utterance_two_per_inference_type(self):
"""Test get_actual_utterance with 2 examples per inference type.
Inference types:
- alternatives
- fragments
- replacement
- shortening
"""
self.reader.record.utterance = (
'&ab mu:(ğ)ça nuutuinnaq [=? nauk tainna] mu:(ğ)ça &ab yarasam '
'[: yorosom] yarasam [: yorosom] nuutuinnaq [=? nauk tainna] .')
actual_output = self.reader.get_actual_utterance()
desired_output = ('ab mu:ça nauk tainna mu:ça ab '
'yarasam yarasam nauk tainna .')
self.assertEqual(actual_output, desired_output)
def test_get_actual_utterance_empty_string(self):
"""Test get_actual_utterance with an empty string."""
self.reader.record.utterance = ''
actual_output = self.reader.get_actual_utterance()
desired_output = ''
self.assertEqual(actual_output, desired_output)
# Tests for the get_target_utterance-method.
def test_get_target_utterance_one_per_inference_type(self):
"""Test get_target_utterance with 1 example per inference type.
Inference types:
- alternatives
- fragments
- replacement
- shortening
"""
self.reader.record.utterance = \
'nuutuinnaq [=? nauk tainna] mu:(ğ)ça &ab yarasam [: yorosom] .'
actual_output = self.reader.get_target_utterance()
desired_output = 'nuutuinnaq mu:ğça xxx yorosom .'
self.assertEqual(actual_output, desired_output)
def test_get_target_utterance_two_per_inference_type(self):
"""Test get_target_utterance with 2 examples per inference type.
Inference types:
- alternatives
- fragments
- replacement
- shortening
"""
self.reader.record.utterance = (
'&ab mu:(ğ)ça nuutuinnaq [=? nauk tainna] mu:(ğ)ça &ab yarasam '
'[: yorosom] yarasam [: yorosom] nuutuinnaq [=? nauk tainna] .')
actual_output = self.reader.get_target_utterance()
desired_output = ('xxx mu:ğça nuutuinnaq mu:ğça xxx '
'yorosom yorosom nuutuinnaq .')
self.assertEqual(actual_output, desired_output)
def test_get_target_utterance_empty_string(self):
"""Test get_target_utterance with an empty string."""
self.reader.record.utterance = ''
actual_output = self.reader.get_target_utterance()
desired_output = ''
self.assertEqual(actual_output, desired_output)
# Tests for the get_morph_tier-method.
def test_get_morph_tier_morph_tier_present(self):
"""Test get_morph_tier with test.cha."""
self.reader.record.dependent_tiers['xmor'] = 'test mor tier'
actual_output = self.reader.get_morph_tier()
desired_output = 'test mor tier'
self.assertEqual(actual_output, desired_output)
def test_get_morph_tier_morph_tier_absent(self):
"""Test get_morph_tier with test.cha."""
self.reader.record.dependent_tiers['xmor'] = ''
actual_output = self.reader.get_morph_tier()
desired_output = ''
self.assertEqual(actual_output, desired_output)
# Tests for the iter_morphemes-method.
def test_iter_morphemes_standard_case(self):
"""Test iter_morphemes with a morph in the expected format."""
morph_word = 'WH|nani^whereat'
actual_output = list(InuktitutReader.iter_morphemes(morph_word))
desired_output = [('WH', 'nani', 'whereat')]
self.assertEqual(actual_output, desired_output)
def test_iter_morphemes_multiple_morphemes(self):
"""Test iter_morphemes with 3 morphe-words."""
morph_words = 'VR|malik^follow+VV|liq^POL+VI|gitsi^IMP_2pS'
actual_output = list(InuktitutReader.iter_morphemes(morph_words))
desired_output = [('VR', 'malik', 'follow'),
('VV', 'liq', 'POL'),
('VI', 'gitsi', 'IMP_2pS')]
self.assertEqual(actual_output, desired_output)
def test_iter_morphemes_no_match(self):
"""Test iter_morphemes with a morpheme that yields no match."""
morph_word = 'WH|nani|whereat'
actual_output = list(InuktitutReader.iter_morphemes(morph_word))
desired_output = [('', '', '')]
self.assertEqual(actual_output, desired_output)
def test_iter_morphemes_empty_string(self):
"""Test iter_morphemes with an empty string."""
morph_word = ''
actual_output = list(InuktitutReader.iter_morphemes(morph_word))
desired_output = [('', '', '')]
self.assertEqual(actual_output, desired_output)
# Tests for the get_segments-method.
def test_get_segments(self):
"""Test get_segments with a standard segment word."""
xmor = 'VN|paaq^remove+VI|got^IMP_2sS'
actual_output = InuktitutReader.get_segments(xmor)
desired_output = ['paaq', 'got']
self.assertEqual(actual_output, desired_output)
# Tests for the get_glosses-method.
def test_get_glosses(self):
"""Test get_glosses with a standard gloss word."""
xmor = 'VN|paaq^remove+VI|got^IMP_2sS'
actual_output = InuktitutReader.get_glosses(xmor)
desired_output = ['remove', 'IMP_2sS']
self.assertEqual(actual_output, desired_output)
# Tests for the get_poses-method.
def test_get_poses(self):
"""Test get_poses with a standard pos word."""
xmor = 'VN|paaq^remove+VI|got^IMP_2sS'
actual_output = InuktitutReader.get_poses(xmor)
desired_output = ['VN', 'VI']
self.assertEqual(actual_output, desired_output)
# Tests for the get_morpheme_language-method.
def test_get_morpheme_language_inuktitut(self):
"""Test get_morpheme_language with a morpheme in Inuktitut."""
seg, gloss, pos = 'ba', '1sg', 'V'
actual_output = InuktitutReader.get_morpheme_language(seg, gloss, pos)
desired_output = 'Inuktitut'
self.assertEqual(actual_output, desired_output)
def test_get_morpheme_language_english(self):
"""Test get_morpheme_language with a morpheme in Inuktitut."""
seg, gloss, pos = 'ba@e', '1sg', 'V'
actual_output = InuktitutReader.get_morpheme_language(seg, gloss, pos)
desired_output = 'English'
self.assertEqual(actual_output, desired_output)
def test_get_morpheme_language_english(self):
"""Test get_morpheme_language with a morpheme in Inuktitut."""
seg, gloss, pos = 'ba@e', '1sg', 'V'
actual_output = InuktitutReader.get_morpheme_language(seg, gloss, pos)
desired_output = 'English'
self.assertEqual(actual_output, desired_output)
def test_get_poses(self):
"""Test get_poses with a standard pos word."""
xmor = 'VN|paaq^remove+VI|got^IMP_2sS'
actual_output = InuktitutReader.get_poses(xmor)
desired_output = ['VN', 'VI']
self.assertEqual(actual_output, desired_output)
def test_get_segments(self):
"""Test get_segments with a standard segment word."""
xmor = 'VN|paaq^remove+VI|got^IMP_2sS'
actual_output = InuktitutReader.get_segments(xmor)
desired_output = ['paaq', 'got']
self.assertEqual(actual_output, desired_output)
def test_iter_morphemes_empty_string(self):
"""Test iter_morphemes with an empty string."""
morph_word = ''
actual_output = list(InuktitutReader.iter_morphemes(morph_word))
desired_output = [('', '', '')]
self.assertEqual(actual_output, desired_output)
def test_iter_morphemes_no_match(self):
"""Test iter_morphemes with a morpheme that yields no match."""
morph_word = 'WH|nani|whereat'
actual_output = list(InuktitutReader.iter_morphemes(morph_word))
desired_output = [('', '', '')]
self.assertEqual(actual_output, desired_output)
def test_iter_morphemes_standard_case(self):
"""Test iter_morphemes with a morph in the expected format."""
morph_word = 'WH|nani^whereat'
actual_output = list(InuktitutReader.iter_morphemes(morph_word))
desired_output = [('WH', 'nani', 'whereat')]
self.assertEqual(actual_output, desired_output)
def test_get_actual_alternative_no_alternatives(self):
"""Test get_actual_alternative with 2 alternatives."""
utterance = 'nuutuinnaq hela nuutuinnaq .'
actual_output = InuktitutReader.get_actual_alternative(utterance)
desired_output = 'nuutuinnaq hela nuutuinnaq .'
self.assertEqual(actual_output, desired_output)
def get_reader(session_file):
return InuktitutReader(session_file)
def test_parse(self):
"""Test parse()."""
session_str = (
'@Participants:\tMAE Maemae Mother , RO Roro Target_Child\n'
'@ID:\tsme|Inuktitut|MAE||female|||Mother|||\n'
'@ID:\tsme|Inuktitut|RO||female|||Target_Child|||\n'
'*RO:\tAllaigutama .\n'
'%eng:\tGive me a comb .\n'
'%xmor:\tNR|ollaoguto^comb+NI|mim^MOD_SG .\n'
'%tim:\t00:01:32\n'
'%add:\tMAE\n'
'@End'
)
parser = InuktitutSessionParser(self.dummy_cha_path)
parser.reader = InuktitutReader(io.StringIO(session_str))
session = parser.parse()
utt = session.utterances[0]
utterance = [
utt.source_id == 'dummy_0',
utt.speaker.code == 'RO',
utt.addressee.code == 'MAE',
utt.childdirected is False,
utt.utterance_raw == 'Allaigutama .',
utt.utterance == 'Allaigutama',
utt.translation == 'Give me a comb .',
utt.morpheme_raw == 'NR|ollaoguto^comb+NI|mim^MOD_SG .',
utt.gloss_raw == 'NR|ollaoguto^comb+NI|mim^MOD_SG .',
utt.pos_raw == 'NR|ollaoguto^comb+NI|mim^MOD_SG .',
utt.sentence_type == 'default',
utt.start_raw == '00:01:32',
utt.end_raw == '',
utt.comment == '',
utt.warning == ''
]
w = utt.words[0]
words = [
w.word_language == '',
w.word == 'Allaigutama',
w.word_actual == 'Allaigutama',
w.word_target == 'Allaigutama',
w.warning == ''
]
m1 = utt.morphemes[0][0]
m2 = utt.morphemes[0][1]
morphemes = [
m1.morpheme_language == 'Inuktitut',
m1.morpheme == 'ollaoguto',
m1.gloss_raw == 'comb',
m1.pos_raw == 'NR',
m2.morpheme_language == 'Inuktitut',
m2.morpheme == 'mim',
m2.gloss_raw == 'MOD_SG',
m2.pos_raw == 'NI'
]
assert (False not in utterance
and False not in words
and False not in morphemes)