# -*- coding: utf-8 -*-
# Copyright 2014-2018 by Christopher C. Little.
# This file is part of Abydos.
#
# Abydos is free software: you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation, either version 3 of the License, or
# (at your option) any later version.
#
# Abydos is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with Abydos. If not, see <http://www.gnu.org/licenses/>.
"""abydos.phones.
The phones module implements ...
"""
from __future__ import division, unicode_literals
from unicodedata import normalize
from six import text_type
from six.moves import range
__all__ = ['FEATURE_MASK', 'PHONETIC_FEATURES', 'cmp_features', 'get_feature',
'ipa_to_features']
PHONETIC_FEATURES = {'t': 2783230754502126250,
't͇': 2783230754502126250,
't̪': 2783230479624219306,
't̠': 2783231579135847082,
'd': 2783230754501864106,
'd͇': 2783230754501864106,
'd̪': 2783230479623957162,
'd̠': 2783231579135584938,
's': 2783230754502125978,
's͇': 2783230754502125978,
's̪': 2783230479624219034,
's̟': 2783230479624219034,
'z': 2783230754501863834,
'z͇': 2783230754501863834,
'z̪': 2783230479623956890,
'z̟': 2783230479623956890,
'θ̱': 2783230754502125994,
'θ͇': 2783230754502125994,
'ɹ̝̊': 2783230754502125994,
'ð̠': 2783230754501863850,
'ð͇': 2783230754501863850,
'ɹ̝': 2783230754501863850,
'ɬ': 2783230754502060454,
'ɮ': 2783230754501863846,
'θ': 2783230479624219050,
's̄': 2783230479624219034,
'ð': 2783230479623956906,
'ð̞': 2693158487076546986,
'ʃ': 2783231556184615322,
'ʃʲ': 2783231556184615322,
'š': 2783231556184615322,
'ɹ̠̊˔': 2783231579135846826,
'ʒ': 2783231556184353178,
'ʒʲ': 2783231556184353178,
'ž': 2783231556184353178,
'ɹ̠˔': 2783231579135584682,
'c': 2783233463150095018,
'ɟ': 2783233463149832874,
'ç': 2783233463150094762,
'ʝ': 2783233463149832618,
'p': 2781720575281375914,
'p̪': 2781702983095331498,
'p͆': 2781702983095331498,
'b': 2781720575281113770,
'b̪': 2781702983095069354,
'b͆': 2781702983095069354,
'f': 2781702983095331242,
'v': 2781702983095069098,
'ɸ': 2781720575281375658,
'ɸ̞': 2691648582733965738,
'β': 2781720575281113514,
'β̞': 2691648582733703594,
'k': 2783233462881659562,
'ɡ': 2783233462881397418,
'g': 2783233462881397418,
'x': 2783233462881659306,
'x̞': 2981391846485961130,
'ɣ': 2783233462881397162,
'ɰ': 2981391846485698986,
'ɣ̞': 2981391846485698986,
'ɣ˕': 2981391846485698986,
'q': 2783233480061528746,
'ɢ': 2783233480061266602,
'χ': 2783233480061528490,
'x̣': 2783233480061528490,
'χ̞': 2693161487514118570,
'ʁ': 2783233480061266346,
'ʁ̝': 2783233480061266346,
'ʁ̞': 2693161487513856426,
'ħ': 2783233503273855402,
'ħ̞': 2693161510726445482,
'ʕ': 2783233503273593258,
'ʕ̝': 2783233503273593258,
'ʕ̞': 2693161510726183338,
'ɑ̯': 2693161510726183338,
'h': 2783233503281129898,
'h̃': 2693161510733719914,
'ɦ': 2783233503280867754,
'ʔ': 2783233503281179306,
'm': 2709662981243185770,
'm̥': 2709662981243447914,
'ɱ': 2709645389057141354,
'm̪': 2709645389057141354,
'n': 2711173160463936106,
'n̪': 2711172885586029162,
'n̥': 2711173160464198250,
'n̊': 2711173160464198250,
'n̪̊': 2711172885586291306,
'ɳ': 2711174259975563882,
'ɳ̊': 2711174259975826026,
'ɳ̥': 2711174259975826026,
'ɲ': 2711175869111904874,
'ñ': 2711175869111904874,
'n̠ʲ': 2711173944966556266,
'ɲ̟': 2711173944966556266,
'ɲ̊': 2711175869112167018,
'ɲ̥': 2711175869112167018,
'ŋ': 2711175868843469418,
'ŋ̊': 2711175868843731562,
'ɴ': 2711175886023338602,
'l': 2693158761954453926,
'l̠': 2693159586588174758,
'l̪': 2693158487076546982,
'ɫ': 2693158721554917798,
'lˠ': 2693158721554917798,
'lˤ': 2693158721554917798,
'l̴': 2693158721554917798,
'ɫ̪': 2693158446677010854,
'l̪ˠ': 2693158446677010854,
'l̪ˤ': 2693158446677010854,
'l̴̪': 2693158446677010854,
'l̥': 2693158761954716070,
'ʎ': 2693161470602422694,
'l̠ʲ': 2693159546457074086,
'ʎ̟': 2693159546457074086,
'ʟ': 2693161470333987238,
'r': 2711173160463936170,
'r̥': 2711173160464198314,
'ɹ': 2693158761954453930,
'ð̠˕': 2693158761954453930,
'ʀ': 2711175886023338666,
'ʀ̥': 2711175886023600810,
'j': 2981391846754134442,
'i̯': 2981391846754134442,
'j̊': 2981391846754396586,
'w': 2978753018579036586,
'u̯': 2978753018579036586,
'ɰʷ': 2978753018579036586,
'ʍ': 2978753018579298730,
'w̥': 2978753018579298730,
'ɰᵝ': 2978682649834858922,
'wᵝ': 2978682649834858922,
'ɥ': 2978753018847472042,
'jʷ': 2978753018847472042,
'ʡ': 2783233503274904234,
'ʜ': 2783233503274903978,
'ʢ': 2783233503274641834,
'ʢ̝': 2783233503274641834,
'ʢ̞': 2693161510727231914,
'ᴙ': 2711175909236714154,
'ʙ': 2709662981243185834,
'ʋ': 2691630990547659178,
'ɕ': 2783231539004746138,
'ʑ': 2783231539004483994,
'ɾ': 2711173160463935914,
'ᴅ': 2711173160463935914,
'ɾ̥': 2711173160464198058,
'ɾ̊': 2711173160464198058,
'ᴅ̥': 2711173160464198058,
'ᴅ̊': 2711173160464198058,
'ɾ̃': 2711173160463935850,
'n̆': 2711173160463935850,
'ⱱ': 2709645389057141162,
'ⱱ̟': 2709662981243185578,
'w̆': 2709662981243185578,
'b̆': 2709662981243185578,
'ɢ̆': 2711175886023338410,
'ʀ̆': 2711175886023338410,
'ʡ̯': 2711175909236713898,
'ɺ': 2711173160463935910,
'ʎ̯': 2711175869111904678,
'ʟ̆': 2711175868843469222,
'ʈ': 2783231854013754026,
'ɖ': 2783231854013491882,
'ʂ': 2783231854013753754,
'ʐ': 2783231854013491610,
'ɻ': 2693159861466081706,
'ɽ': 2711174259975563690,
'ɽ͡r': 2711174259975563946,
'ɭ': 2693159861466081702,
'ɭ̆': 2711174259975563686,
'ɺ˞': 2711174259975563686,
'ɺ̢': 2711174259975563686,
'ꞎ': 2783231854013688230,
'ʎ̝̊': 2783233463150094758,
'ʟ̝̊': 2783233462881659302,
'ʟ̝': 2783233462881397158,
# affricates & co-articulated
't͡ʃ': 2783231556184615833,
'ʧ': 2783231556184615833,
't͜ʃ': 2783231556184615833,
't̠ʲʃ': 2783231556184615833,
'č': 2783231556184615833,
'd͡ʒ': 2783231556184353689,
'ʤ': 2783231556184353689,
'd͜ʒ': 2783231556184353689,
'd̠ʲʒ': 2783231556184353689,
'ǯ': 2783231556184353689,
't͡s': 2783230754502126489,
'ʦ': 2783230754502126489,
't͜s': 2783230754502126489,
't̪͡s̪': 2783230479624219545,
't͡s̪': 2783230479624219545,
't̟͡s̟': 2783230479624219545,
't͡s̟': 2783230479624219545,
'ʦ̪': 2783230479624219545,
'ʦ̟': 2783230479624219545,
't͡θ̠': 2783230754502126505,
't͡θ͇': 2783230754502126505,
't͡θ': 2783230479624219561,
't͜θ': 2783230479624219561,
't̪͡θ': 2783230479624219561,
't̟͡θ': 2783230479624219561,
'd͡z': 2783230754501864345,
'ʣ': 2783230754501864345,
'd͜z': 2783230754501864345,
'd̪͡z̪': 2783230479623957401,
'd͡z̪': 2783230479623957401,
'd̟͡z̟': 2783230479623957401,
'd͡z̟': 2783230479623957401,
'ʣ̪': 2783230479623957401,
'ʣ̟': 2783230479623957401,
'd͡ð̠': 2783230754501864361,
'd͡ð̳': 2783230754501864361,
'd͡ð': 2783230479623957417,
'd͜ð': 2783230479623957417,
'd̪͡ð': 2783230479623957417,
'd̟͡ð': 2783230479623957417,
'k͡x': 2783233462881659817,
'ɡ͡ɣ': 2783233462881397673,
'g͡ɣ': 2783233462881397673,
'p͡f': 2781738167467420585,
'p̪͡f': 2781702983095331753,
'b͡v': 2781738167467158441,
'b̪͡v': 2781702983095069609,
'b̪͜v': 2781702983095069609,
't͡ɕ': 2783231539004746649,
't͜ɕ': 2783231539004746649,
'ʨ': 2783231539004746649,
'd͡ʑ': 2783231539004484505,
'd͜ʑ': 2783231539004484505,
'ʥ': 2783231539004484505,
'ʈ͡ʂ': 2783231854013754265,
't͡ʂ': 2783231854013754265,
'ɖ͡ʐ': 2783231854013492121,
'd͡ʐ': 2783231854013492121,
't͡ɬ': 2783230754502060965,
't͜ɬ': 2783230754502060965,
'ƛ': 2783230754502060965,
'd͡ɮ': 2783230754501864357,
'c͡ç': 2783233463150095273,
'c͜ç': 2783233463150095273,
'ɟ͡ʝ': 2783233463149833129,
'c͡ʎ̝̥': 2783233463150095269,
'k͡ʟ̝̊': 2783233462881659813,
'ɡ͡ʟ̝': 2783233462881397669,
'q͡χ': 2783233480061529001,
'ɢ͡ʁ': 2783233480061266857,
'ɢ͜ʁ': 2783233480061266857,
'ɧ': 2783231813614217626,
'k͡p': 2781720534881839786,
'k͜p': 2781720534881839786,
'ɡ͡b': 2781720534881577642,
'kʷ': 2780594634974997162,
'k͡w': 2780594634974997162,
'k͜w': 2780594634974997162,
'kʷh': 2780594634974931626,
'gʷ': 2780594634974735018,
'g͡w': 2780594634974735018,
'gʷh': 2780594634974669482,
# implosives
'ɓ̥': 2781720575281355434,
'ƥ': 2781720575281355434,
'pʼ↓': 2781720575281355434,
'ɓ': 2781720575281093290,
'ɗ̥': 2783230754502105770,
'ƭ': 2783230754502105770,
'tʼ↓': 2783230754502105770,
'ɗ': 2783230754501843626,
'ᶑ': 2783231854013471402,
'ʄ̊': 2783233463150074538,
'ƈ': 2783233463150074538,
'cʼ↓': 2783233463150074538,
'ʄ': 2783233463149812394,
'ɠ': 2783233462881376938,
'ʛ': 2783233480061246122,
# clicks
'ʘ': 2781720575281374890,
'ʘʰ': 2781720575281309354,
'ʘ̬': 2781720575281112746,
'ᶢʘ': 2781720575281112746,
'ʘ̃': 2781720575281112682,
'ᵑʘ': 2781720575281112682,
'ᵐʘ': 2781720575281112682,
'ʘ̥̃ʰ': 2781720575281309290,
'ᵑ̊ʘʰ': 2781720575281309290,
'ʘ̃ˀ': 2781720575281358442,
'ᵑʘˀ': 2781720575281358442,
'ʘ̃͜ʔ': 2781720575281358442,
'ᵑ̊ʘˀ': 2781720575281358442,
'ǀ': 2783230479624218282,
'ʇ': 2783230479624218282,
'ǀ̬': 2783230479623956138,
'ʇ̬': 2783230479623956138,
'ᶢǀ': 2783230479623956138,
'ᶢʇ': 2783230479623956138,
'ǀ̃': 2783230479623956074,
'ʇ̃': 2783230479623956074,
'ᵑǀ': 2783230479623956074,
'ⁿǀ': 2783230479623956074,
'ᵑʇ': 2783230479623956074,
'ǀ̥̃ʰ': 2783230479624152682,
'ʇ̥̃ʰ': 2783230479624152682,
'ᵑ̊ǀʰ': 2783230479624152682,
'ᵑ̊ʇʰ': 2783230479624152682,
'ǀ̃ˀ': 2783230479624201834,
'ʇ̃ˀ': 2783230479624201834,
'ᵑǀˀ': 2783230479624201834,
'ᵑʇˀ': 2783230479624201834,
'ᵑ̊ǀˀ': 2783230479624201834,
'ᵑ̊ʇˀ': 2783230479624201834,
'ǃ': 2783230754502125226,
'ʗ': 2783230754502125226,
'ǃʰ': 2783230754502059690,
'ʗʰ': 2783230754502059690,
'ǃ̬': 2783230754501863082,
'ʗ̬': 2783230754501863082,
'ᶢǃ': 2783230754501863082,
'ᶢʗ': 2783230754501863082,
'ǃ̃': 2783230754501863018,
'ᵑǃ': 2783230754501863018,
'ʗ̃': 2783230754501863018,
'ᵑʗ': 2783230754501863018,
'ǃ̥̃ʰ': 2783230754502059626,
'ʗ̃̊ʰ': 2783230754502059626,
'ᵑ̊ǃʰ': 2783230754502059626,
'ᵑ̊ʗʰ': 2783230754502059626,
'ǃ̃ˀ': 2783230754502108778,
'ʗ̃ˀ': 2783230754502108778,
'ᵑǃˀ': 2783230754502108778,
'ᵑʗˀ': 2783230754502108778,
'ǂ': 2783233463150093994,
'ʄ̵': 2783233463150093994,
'⨎': 2783233463150093994,
'ǂʰ': 2783233463150028458,
'ǂ̬': 2783233463149831850,
'ᶢǂ': 2783233463149831850,
'ǂ̃': 2783233463149831786,
'ᵑǂ': 2783233463149831786,
'ǂ̥̃ʰ': 2783233463150028394,
'ᵑ̊ǂʰ': 2783233463150028394,
'ǂ̃ˀ': 2783233463150077546,
'ᵑǂˀ': 2783233463150077546,
'ᵑǂ͡ʔ': 2783233463150077546,
'ᵑ̊ǂˀ': 2783233463150077546,
'ǃ͡s': 2783233463150093978,
'ǂᶴ': 2783233463150093978,
'ǁ': 2783230754502125222,
'ʖ': 2783230754502125222,
'ǁʰ': 2783230754502059686,
'ʖʰ': 2783230754502059686,
'ǁ̬': 2783230754501863078,
'ʖ̬': 2783230754501863078,
'ᶢǁ': 2783230754501863078,
'ᶢʖ': 2783230754501863078,
'ǁ̃': 2783230754501863014,
'ʖ̃': 2783230754501863014,
'ᵑǁ': 2783230754501863014,
'ᵑʖ': 2783230754501863014,
'ǁ̥̃ʰ': 2783230754502059622,
'ʖ̥̃ʰ': 2783230754502059622,
'ᵑ̊ǁʰ': 2783230754502059622,
'ᵑ̊ʖʰ': 2783230754502059622,
'ǁ̃ˀ': 2783230754502108774,
'ʖ̃ˀ': 2783230754502108774,
'ᵑǁˀ': 2783230754502108774,
'ᵑʖˀ': 2783230754502108774,
'ᵑǁ͡ʔ': 2783230754502108774,
'ʖ̃͜ʔ': 2783230754502108774,
'ǃ˞': 2783231854013753002,
'ǃǁ': 2783231854013753002,
'‼': 2783231854013753002,
'!!': 2783231854013753002,
'ǃ˞ʰ': 2783231854013687466,
'‼ʰ': 2783231854013687466,
'!!ʰ': 2783231854013687466,
'ǃ̬˞': 2783231854013490858,
'ᶢǃ˞': 2783231854013490858,
'‼̬': 2783231854013490858,
'!!̬': 2783231854013490858,
'ᶢ‼': 2783231854013490858,
'ᶢ!!': 2783231854013490858,
'ǃ̃˞': 2783231854013490794,
'ᵑǃ˞': 2783231854013490794,
'‼̃': 2783231854013490794,
'!!̃': 2783231854013490794,
'ᵑ‼': 2783231854013490794,
'ᵑ!!': 2783231854013490794,
'ǃ̥̃˞ʰ': 2783231854013687402,
'ᵑ̊ǃ˞ʰ': 2783231854013687402,
'‼̥̃ʰ': 2783231854013687402,
'!!̥̃ʰ': 2783231854013687402,
'ᵑ̊‼ʰ': 2783231854013687402,
'ᵑ̊!!ʰ': 2783231854013687402,
'ǃ̃˞ˀ': 2783231854013736554,
'ᵑǃ˞͜ʔ': 2783231854013736554,
'ᵑ̊‼ˀ': 2783231854013736554,
'ᵑ‼ˀ': 2783231854013736554,
'‼̃ˀ': 2783231854013736554,
'ᵑ̊!!ˀ': 2783231854013736554,
'ᵑ!!ˀ': 2783231854013736554,
'!!̃ˀ': 2783231854013736554,
# vowels
'i': 1826957412996131242,
'ɪ': 1826957413067434410,
'ɩ': 1826957413067434410,
'u': 1825831513894594986,
'ʊ': 1825831513965898154,
'ɷ': 1825831513965898154,
'ᴜ': 1825831513965898154,
'ɯ̽': 1826957413872740778,
'ʊ̜': 2115187790024452522,
'e': 1826957430176000426,
'ɛ': 1826957430247303594,
'o': 1825831531074464170,
'ɤ': 1826957430981306794,
'ɔ': 1825831531145767338,
'ʌ': 1826957431052609962,
'a': 1826957425952336298,
'a̟': 1826957425952336298,
'æ̞': 1826957425952336298,
'ɶ': 1825831526045493674,
'æ': 1826957425885227434,
'y': 1825831513089288618,
'ʏ': 1825831513160591786,
'ø': 1825831530269157802,
'œ': 1825831530340460970,
'ə': 1828083331160779178,
'ɵ̞': 1825831531347093930,
'ə̹': 1825831531347093930,
'ɞ̝': 1825831531347093930,
'ɘ̞': 1826957431253936554,
'ɯ': 1826957413805631914,
'ɒ': 1825831526850800042,
'ɑ': 1826957426757642666,
'ɨ': 1826957414069873066,
'ï': 1826957414069873066,
'ʉ': 1825831514163030442,
'ü': 1825831514163030442,
'ɘ': 1826957431249742250,
'ë': 1826957431249742250,
'ɵ': 1825831531342899626,
'ö': 1825831531342899626,
'ɜ': 1826957431316851114,
'ɛ̈': 1826957431316851114,
'ɞ': 1825831531410008490,
'ɔ̈': 1825831531410008490,
'ä': 1826957427026078122,
'a̠': 1826957427026078122,
'ɑ̈': 1826957427026078122,
'ɐ̞': 1826957427026078122,
'ɐ': 1828083326865811882,
'ɜ̞': 1826957426958969258,
'ɞ̞': 1825831527052126634,
'ɪ̟': 1826957413063240106,
'ʏ̟': 1825831513156397482,
'ʏ̫': 1825550038183881130,
'ʏʷ': 1825550038183881130,
'ɪʷ': 1825550038183881130,
'y̫': 1825550038112577962,
'yʷ': 1825550038112577962,
'iʷ': 1825550038112577962,
'u͍': 1825761145150417322,
'ɯᵝ': 1825761145150417322,
'ɯ͡β̞': 1825761145150417322,
'ʉ͍': 1825761145418852778,
'ɨᵝ': 1825761145418852778,
'ɨ͡β̞': 1825761145418852778,
'ɪ̈': 1826957414136981930,
'ɨ̞': 1826957414136981930,
'ɘ̝': 1826957414136981930,
'ʊ̈': 1825831514230139306,
'ʉ̞': 1825831514230139306,
'ø̫': 1825550055292447146,
'øʷ': 1825550055292447146,
'eʷ': 1825550055292447146,
'e̞': 1826957430180194730,
'ɛ̝': 1826957430180194730,
'ø̞': 1825831530273352106,
'œ̝': 1825831530273352106,
'o̞': 1825831531078658474,
'ɔ̝': 1825831531078658474,
'ɤ̞': 1826957430985501098,
'ʌ̝': 1826957430985501098,
# feature diacritics
'ʰ': 65536,
'ʱ': 327680,
'̤': 327680,
'ʼ': 16384,
'̚': 131072,
'̬': 262144,
'̌': 262144,
'̥': 524288,
'̊': 524288,
'ʷ': 6737807255011328,
'ʲ': 103012106240,
'̰': 278528,
'̩': 1152921504606846976,
'̯': 2305843009213693952,
'̃': 64,
'̨': 64,
'͊': 128,
'ˠ': 98180268032,
'̴': 98180268032,
'ˤ': 32505856,
'̘': 23068672,
'̙': 26214400,
'˭': 688256}
FEATURE_MASK = {'syllabic': 3458764513820540928,
'consonantal': 864691128455135232,
'sonorant': 216172782113783808,
'approximant': 54043195528445952,
'labial': 13510798882111488,
'round': 3377699720527872,
'protruded': 844424930131968,
'compressed': 211106232532992,
'labiodental': 52776558133248,
'coronal': 13194139533312,
'anterior': 3298534883328,
'distributed': 824633720832,
'dorsal': 206158430208,
'high': 51539607552,
'low': 12884901888,
'front': 3221225472,
'back': 805306368,
'tense': 201326592,
'pharyngeal': 50331648,
'atr': 12582912,
'rtr': 3145728,
'voice': 786432,
'spread_glottis': 196608,
'constricted_glottis': 49152,
'glottalic_suction': 12288,
'velaric_suction': 3072,
'continuant': 768,
'nasal': 192,
'strident': 48,
'lateral': 12,
'delayed_release': 3}
[docs]def ipa_to_features(ipa):
"""Convert IPA to features.
This translates an IPA string of one or more phones to a list of ints
representing the features of the string.
:param str ipa: the IPA representation of a phone or series of phones
:returns: a representation of the features of the input string
:rtype: [int]
>>> ipa_to_features('mut')
[2709662981243185770, 1825831513894594986, 2783230754502126250]
>>> ipa_to_features('fon')
[2781702983095331242, 1825831531074464170, 2711173160463936106]
>>> ipa_to_features('telz')
[2783230754502126250, 1826957430176000426, 2693158761954453926,
2783230754501863834]
"""
features = []
pos = 0
ipa = normalize('NFD', text_type(ipa.lower()))
maxsymlen = max(len(_) for _ in PHONETIC_FEATURES)
while pos < len(ipa):
found_match = False
for i in range(maxsymlen, 0, -1):
if pos+i-1 <= len(ipa) and ipa[pos:pos+i] in PHONETIC_FEATURES:
features.append(PHONETIC_FEATURES[ipa[pos:pos+i]])
pos += i
found_match = True
if not found_match:
features.append(-1)
pos += 1
return features
[docs]def get_feature(vector, feature):
"""Get a feature vector.
This returns a list of ints, equal in length to the vector input,
representing presence/absence/neutrality with respect to a particular
phonetic feature.
:param list vector: a tuple or list of ints representing the phonetic
features of a phone or series of phones
(such as is returned by the ipa_to_features function)
:param str feature: a feature name from the set:
'consonantal', 'sonorant', 'syllabic', 'labial', 'round', 'coronal',
'anterior', 'distributed', 'dorsal', 'high', 'low', 'back', 'tense',
'pharyngeal', 'ATR', 'voice', 'spread_glottis', 'constricted_glottis',
'continuant', 'strident', 'lateral', 'delayed_release', 'nasal'
:returns: a list indicating presence/absence/neutrality with respect to
the feature
:rtype: [int]
>>> tails = ipa_to_features('telz')
>>> get_feature(tails, 'consonantal')
[1, -1, 1, 1]
>>> get_feature(tails, 'sonorant')
[-1, 1, 1, -1]
>>> get_feature(tails, 'nasal')
[-1, -1, -1, -1]
>>> get_feature(tails, 'coronal')
[1, -1, 1, 1]
"""
# :param bool binary: if False, -1, 0, & 1 represent -, 0, & +
# if True, only binary oppositions are allowed:
# 0 & 1 represent - & + and 0s are mapped to -
if feature not in FEATURE_MASK:
raise AttributeError("feature must be one of: '" +
"', '".join(('consonantal', 'sonorant',
'syllabic', 'labial', 'round',
'coronal', 'anterior', 'distributed',
'dorsal', 'high', 'low', 'back',
'tense', 'pharyngeal', 'ATR',
'voice', 'spread_glottis',
'constricted_glottis', 'continuant',
'strident', 'lateral',
'delayed_release', 'nasal')) +
"'")
# each feature mask contains two bits, one each for - and +
mask = FEATURE_MASK[feature]
# the lower bit represents +
pos_mask = mask >> 1
retvec = []
for char in vector:
if char < 0:
retvec.append(float('NaN'))
else:
masked = char & mask
if masked == 0:
retvec.append(0) # 0
elif masked == mask:
retvec.append(2) # +/-
elif masked & pos_mask:
retvec.append(1) # +
else:
retvec.append(-1) # -
return retvec
[docs]def cmp_features(feat1, feat2):
"""Compare features.
This returns a number in the range [0, 1] representing a comparison of two
feature bundles.
If one of the bundles is negative, -1 is returned (for unknown values)
If the bundles are identical, 1 is returned.
If they are inverses of one another, 0 is returned.
Otherwise, a float representing their similarity is returned.
:param int feat1: a feature bundle
:param int feat2: a feature bundle
:returns: a comparison of the feature bundles
:rtype: float
>>> cmp_features(ipa_to_features('l')[0], ipa_to_features('l')[0])
1.0
>>> cmp_features(ipa_to_features('l')[0], ipa_to_features('n')[0])
0.8709677419354839
>>> cmp_features(ipa_to_features('l')[0], ipa_to_features('z')[0])
0.8709677419354839
>>> cmp_features(ipa_to_features('l')[0], ipa_to_features('i')[0])
0.564516129032258
"""
if feat1 < 0 or feat2 < 0:
return -1.0
if feat1 == feat2:
return 1.0
magnitude = len(FEATURE_MASK)
featxor = feat1 ^ feat2
diffbits = 0
# print(featxor)
while featxor:
if featxor & 0b1:
diffbits += 1
featxor >>= 1
# print(diffbits)
return 1 - (diffbits/(2*magnitude))
# TODO: finish implementation/testing/tuning
if __name__ == '__main__':
import doctest
doctest.testmod()