def test_merge_dagdra():
token_list = pos_tok.tokenize("བཀྲ་ཤིས་-པ་")
token_list = [t for t in token_list if t.text != "-"
] # remove the "-" inserted to ensure we have two tokens
mp = MergeDagdra()
mp.merge(token_list)
assert len(token_list) == 1 and token_list[0].text == "བཀྲ་ཤིས་པ་"
token_list = pos_tok.tokenize("བཀྲ་ཤིས་-པའོ།")
token_list = [t for t in token_list if t.text != "-"
] # remove the "-" inserted to ensure we have two tokens
mp.merge(token_list)
assert len(token_list) == 3 and token_list[0].text == "བཀྲ་ཤིས་པ"
def test_tokenize():
profile = "empty"
main, custom = Config().get_tok_data_paths(profile)
tok = Tokenize(Trie(BoSyl, profile, main, custom))
tok.trie.inflect_n_modify_trie("བཀྲ་ཤིས་")
tok.trie.inflect_n_add_data("བཀྲ་ཤིས་\tNOUN\t\tབཀྲ་ཤིས་\t17500")
tok.trie.inflect_n_modify_trie("མཐའ་")
tok.trie.inflect_n_add_data("མཐའ་\tNOUN")
in_str = "མཐའི་བཀྲ་ཤིས། ཀཀ abc མཐའི་རྒྱ་མཚོ་"
preproc = TokChunks(in_str)
preproc.serve_syls_to_trie()
tokens = tok.tokenize(preproc)
expected = dedent(
"""\
text: "བཀྲ་ཤིས"
text_cleaned: "བཀྲ་ཤིས་"
text_unaffixed: "བཀྲ་ཤིས་"
syls: ["བཀྲ", "ཤིས"]
senses: | pos: NOUN, freq: 17500, sense: བཀྲ་ཤིས་, affixed: False |
char_types: |CONS|CONS|SUB_CONS|TSEK|CONS|VOW|CONS|
chunk_type: TEXT
syls_idx: [[0, 1, 2], [4, 5, 6]]
syls_start_end: [{'start': 0, 'end': 4}, {'start': 4, 'end': 7}]
start: 5
len: 7
"""
)
str(tokens[0])
assert str(tokens[1]) == expected
assert tokens[2].text == "། "
assert tokens[2].chunk_type == "PUNCT"
# add sense to བཀྲ་ཤིས་
pos_tok.tok.trie.inflect_n_add_data("བཀྲ་ཤིས་\tNOUN\t\tབཀྲ་ཤིས་\t17500")
tokens = pos_tok.tokenize(in_str)
expected = dedent(
"""\
text: "བཀྲ་ཤིས"
text_cleaned: "བཀྲ་ཤིས་"
text_unaffixed: "བཀྲ་ཤིས་"
syls: ["བཀྲ", "ཤིས"]
pos: NOUN
lemma: བཀྲ་ཤིས་
sense: བཀྲ་ཤིས་
senses: | pos: NOUN, freq: 17204, affixed: False, lemma: བཀྲ་ཤིས་ | pos: NOUN, freq: 17500, sense: བཀྲ་ཤིས་, affixed: False, lemma: བཀྲ་ཤིས་ |
char_types: |CONS|CONS|SUB_CONS|TSEK|CONS|VOW|CONS|
chunk_type: TEXT
freq: 17500
syls_idx: [[0, 1, 2], [4, 5, 6]]
syls_start_end: [{'start': 0, 'end': 4}, {'start': 4, 'end': 7}]
start: 5
len: 7
"""
)
assert str(tokens[2]) == expected
def test_num_lemmas_missing():
in_str = "སྟོང་ཕྲག་བརྒྱ་པ་སུམ་བརྒྱ་པ་བཅུ་པ་ལྔ་པ་"
tokens = pos_tok.tokenize(in_str)
assert [t.lemma for t in tokens] == [
"སྟོང་ཕྲག་",
"བརྒྱ་པ་",
"སུམ་བརྒྱ་པ་",
"བཅུ་པ་",
"ལྔ་པ་",
]
def test_missing_token26():
input_str = "སྲི་མོ་བཛྲ་ནོ་ཏི་སྟ་ཀཱི་"
tokens = pos_tok.tokenize(input_str, split_affixes=False)
assert [t.text for t in tokens] == [
"སྲི་",
"མོ་",
"བཛྲ་",
"ནོ་",
"ཏི་",
"སྟ་",
"ཀཱི་",
]
def test_adjust_tokens():
string = "ལ་ལ་ལ་ལ་ལ་བ་ཡོད།"
token_list = pos_tok.tokenize(string, split_affixes=False)
at = AdjustTokens(main=main, custom=custom)
adjusted = at.adjust(token_list)
assert token_list[0].text == "ལ་ལ་"
assert token_list[1].text == "ལ་ལ་"
assert adjusted[0].text == "ལ་"
assert adjusted[0].pos == "PART"
assert adjusted[1].text == "ལ་ལ་"
assert adjusted[1].pos == "PART"
assert adjusted[2].text == "ལ་"
assert adjusted[2].pos == "PART"
def test_missing_token13():
input_str = "།ཨོཾ་གི་རི་ཧི་རི་ཙི་རི། །ཨཱ་ཨཱ་ཤུ་མ་ཤ་"
tokens = pos_tok.tokenize(input_str, split_affixes=False)
assert [t.text for t in tokens] == [
"།",
"ཨོཾ་",
"གི་",
"རི་",
"ཧི་",
"རི་",
"ཙི་",
"རི",
"། །",
"ཨཱ་",
"ཨཱ་",
"ཤུ་",
"མ་",
"ཤ་",
]
def test_segmentation_bug():
tokens = pos_tok.tokenize("ལ་པོ་ལ་པོ་ལ་པོ་")
assert len(tokens) == 3
tokens = pos_tok.tokenize("ལ་མོ་ལ་མོ་ལ་མོ་")
assert len(tokens) == 3
tokens = pos_tok.tokenize("གྲོགས་པོ་གྲོགས་པོ་གྲོགས་པོ་")
assert len(tokens) == 3
tokens = pos_tok.tokenize("བདག་པོ་བདག་པོ་བདག་པོ་དང་")
assert len(tokens) == 4
tokens = pos_tok.tokenize("བདག་པོ་བདག་པོ་བདག་པོ་")
assert len(tokens) == 3
tokens = pos_tok.tokenize(
"བདག་པོ་བདག་པོ་བདག་པོ་བདག་པོ་བདག་པོ་བདག་པོ་བདག་པོ་བདག་པོ་བདག་པོ་")
assert len(tokens) == 9
def test_missing_token11():
input_str = "སྦོམ་ཞིང་ཆེ་"
tokens = pos_tok.tokenize(input_str, split_affixes=False)
assert [t.text for t in tokens] == ["སྦོམ་", "ཞིང་", "ཆེ་"]
def test_missing_token9():
input_str = "བསྐོལ། །རྡོ་རྗེ་"
tokens = pos_tok.tokenize(input_str, split_affixes=False)
assert [t.text for t in tokens] == ["བསྐོལ", "། །", "རྡོ་རྗེ་"]
def test_missing_token8():
input_str = "བི་སི་ནི་"
tokens = pos_tok.tokenize(input_str, split_affixes=False)
assert [t.text for t in tokens] == ["བི་", "སི་", "ནི་"]
def test_missing_token59():
input_str = "།་གླེན་ལྐུགས་"
tokens = pos_tok.tokenize(input_str, split_affixes=False)
assert [t.text for t in tokens] == ["།་", "གླེན་ལྐུགས་"]
def test_missing_token57():
input_str = "སྒལ་བརྒྱུད་ཞབས་"
tokens = pos_tok.tokenize(input_str, split_affixes=False)
assert [t.text for t in tokens] == ["སྒལ་", "བརྒྱུད་", "ཞབས་"]
def test_missing_token44():
input_str = "བཏུལ་མཚམས་བཅད་པ"
tokens = pos_tok.tokenize(input_str, split_affixes=False)
assert [t.text for t in tokens] == ["བཏུལ་", "མཚམས་", "བཅད་པ"]
def test_missing_token43():
input_str = "གཅོད་འཁོར་ལོ་"
tokens = pos_tok.tokenize(input_str, split_affixes=False)
assert [t.text for t in tokens] == ["གཅོད་", "འཁོར་ལོ་"]
def test_missing_token42():
input_str = "ཞིབ་བས་སྦལ།"
tokens = pos_tok.tokenize(input_str, split_affixes=False)
assert [t.text for t in tokens] == ["ཞིབ་", "བས་", "སྦལ", "།"]
def test_missing_token55():
input_str = "བྷ་གར་སྦྱོར་"
tokens = pos_tok.tokenize(input_str, split_affixes=False)
assert [t.text for t in tokens] == ["བྷ་", "གར་", "སྦྱོར་"]
def test_missing_token56():
input_str = "བརྒྱུད་སྐུ་གདུང་"
tokens = pos_tok.tokenize(input_str, split_affixes=False)
assert [t.text for t in tokens] == ["བརྒྱུད་", "སྐུ་གདུང་"]
def test_missing_token46():
input_str = "བསྐུར་ལས་ཀྱི་"
tokens = pos_tok.tokenize(input_str, split_affixes=False)
assert [t.text for t in tokens] == ["བསྐུར་", "ལས་", "ཀྱི་"]
def test_missing_token58():
input_str = "བརྩེགས་ཆེ་མཆོག་"
tokens = pos_tok.tokenize(input_str, split_affixes=False)
assert [t.text for t in tokens] == ["བརྩེགས་", "ཆེ་མཆོག་"]
def test_missing_token47():
input_str = "འཁོས་དུ། །ཆེ་"
tokens = pos_tok.tokenize(input_str, split_affixes=False)
assert [t.text for t in tokens] == ["འཁོས་", "དུ", "། །", "ཆེ་"]
def test_missing_token7():
input_str = "གདབ། །ཨོཾ་ན་"
tokens = pos_tok.tokenize(input_str, split_affixes=False)
assert [t.text for t in tokens] == ["གདབ", "། །", "ཨོཾ་", "ན་"]
def test_missing_token48():
input_str = "ནུ་ཧེ་རུ་"
tokens = pos_tok.tokenize(input_str, split_affixes=False)
assert [t.text for t in tokens] == ["ནུ་", "ཧེ་", "རུ་"]
def test_missing_token1():
input_str = "འཐུང་བུད་"
tokens = pos_tok.tokenize(input_str, split_affixes=False)
assert [t.text for t in tokens] == ["འཐུང་", "བུད་"]
def test_missing_token49():
input_str = "བརྩེགས་རྣམ་པར་"
tokens = pos_tok.tokenize(input_str, split_affixes=False)
assert [t.text for t in tokens] == ["བརྩེགས་", "རྣམ་པར་"]
def test_missing_token10():
input_str = "བསྐུས་ཤིང་མཉེས་"
tokens = pos_tok.tokenize(input_str, split_affixes=False)
assert [t.text for t in tokens] == ["བསྐུས་", "ཤིང་", "མཉེས་"]
def test_missing_token51():
input_str = "ནུ་ཡེ་ཤེས་"
tokens = pos_tok.tokenize(input_str, split_affixes=False)
assert [t.text for t in tokens] == ["ནུ་", "ཡེ་ཤེས་"]
def test_missing_token12():
input_str = "བྷ་ག་ཁ་ཆེ་"
tokens = pos_tok.tokenize(input_str, split_affixes=False)
assert [t.text for t in tokens] == ["བྷ་", "ག་", "ཁ་ཆེ་"]
def test_missing_token54():
input_str = "བྷ་གར་འཁྱིལ། །ཨོཾ་"
tokens = pos_tok.tokenize(input_str, split_affixes=False)
assert [t.text for t in tokens] == ["བྷ་", "གར་", "འཁྱིལ", "། །", "ཨོཾ་"]
def test_missing_token14():
input_str = "བཟླས་བྱས་"
tokens = pos_tok.tokenize(input_str, split_affixes=False)
assert [t.text for t in tokens] == ["བཟླས་", "བྱས་"]
def test_missing_token6():
input_str = "རབ་བསྐུས་ནས།"
tokens = pos_tok.tokenize(input_str, split_affixes=False)
assert [t.text for t in tokens] == ["རབ་", "བསྐུས་", "ནས", "།"]