11625814262

Committed 01 Nov 2024 07:14AM UTC coverage: 20.782% (+20.8%) from 0.0%

Build # 11625814262

Build Type

Pull #952

github

Committed by

web-flow

Commit Message

Merge c8385dcae into 515fe7ced

Pull Request Pull Request #952: Specify a limited test suite

Run Details

45 of 80 new or added lines in 48 files covered. (56.25%)

1537 of 7396 relevant lines covered (20.78%)

0.21 hits per line

Source File
Press 'n' to go to next uncovered line, 'b' for previous

0.0

/pythainlp/spell/symspellpy.py

# -*- coding: utf-8 -*-
# SPDX-FileCopyrightText: 2016-2024 PyThaiNLP Project
# SPDX-License-Identifier: Apache-2.0
"""
symspellpy

symspellpy is a Python port of SymSpell v6.5.
We used unigram & bigram from Thai National Corpus (TNC).

:See Also:
    * \
        https://github.com/mammothb/symspellpy
"""
from typing import List

from symspellpy import SymSpell, Verbosity

from pythainlp.corpus import get_corpus_path, path_pythainlp_corpus

_UNIGRAM = "tnc_freq.txt"
_BIGRAM = "tnc_bigram_word_freqs"

sym_spell = SymSpell()
sym_spell.load_dictionary(
    path_pythainlp_corpus(_UNIGRAM), 0, 1, separator="\t", encoding="utf-8-sig"
)
sym_spell.load_bigram_dictionary(
    get_corpus_path(_BIGRAM), 0, 2, separator="\t", encoding="utf-8-sig"
)


def spell(text: str, max_edit_distance: int = 2) -> List[str]:
    return [
        str(i).split(",", maxsplit=1)[0]
        for i in list(
            sym_spell.lookup(
                text, Verbosity.CLOSEST, max_edit_distance=max_edit_distance
            )
        )
    ]


def correct(text: str, max_edit_distance: int = 1) -> str:
    return spell(text, max_edit_distance=max_edit_distance)[0]


def spell_sent(list_words: List[str], max_edit_distance: int = 2) -> List[str]:
    _temp = [
        str(i).split(",", maxsplit=1)[0].split(" ")
        for i in list(
            sym_spell.lookup_compound(
                " ".join(list_words),
                split_by_space=True,
                max_edit_distance=max_edit_distance,
            )
        )
    ]
    list_new = []
    for i in _temp:
        list_new.append(i)

    return list_new


def correct_sent(list_words: List[str], max_edit_distance=1) -> List[str]:
    return [
        i[0]
        for i in spell_sent(list_words, max_edit_distance=max_edit_distance)
    ]

1	# -- coding: utf-8 --
2	# SPDX-FileCopyrightText: 2016-2024 PyThaiNLP Project
3	# SPDX-License-Identifier: Apache-2.0
4	"""
5	symspellpy
6
7	symspellpy is a Python port of SymSpell v6.5.
8	We used unigram & bigram from Thai National Corpus (TNC).
9
10	:See Also:
11	* \
12	https://github.com/mammothb/symspellpy
13	"""
14	from typing import List	×
15
16	from symspellpy import SymSpell, Verbosity	×
17
NEW 18	from pythainlp.corpus import get_corpus_path, path_pythainlp_corpus	×
19
20	_UNIGRAM = "tnc_freq.txt"	×
21	_BIGRAM = "tnc_bigram_word_freqs"	×
22
23	sym_spell = SymSpell()	×
24	sym_spell.load_dictionary(	×
25	path_pythainlp_corpus(_UNIGRAM), 0, 1, separator="\t", encoding="utf-8-sig"
26	)
27	sym_spell.load_bigram_dictionary(	×
28	get_corpus_path(_BIGRAM), 0, 2, separator="\t", encoding="utf-8-sig"
29	)
30
31
32	def spell(text: str, max_edit_distance: int = 2) -> List[str]:	×
33	return [	×
34	str(i).split(",", maxsplit=1)[0]
35	for i in list(
36	sym_spell.lookup(
37	text, Verbosity.CLOSEST, max_edit_distance=max_edit_distance
38	)
39	)
40	]
41
42
43	def correct(text: str, max_edit_distance: int = 1) -> str:	×
44	return spell(text, max_edit_distance=max_edit_distance)[0]	×
45
46
47	def spell_sent(list_words: List[str], max_edit_distance: int = 2) -> List[str]:	×
48	_temp = [	×
49	str(i).split(",", maxsplit=1)[0].split(" ")
50	for i in list(
51	sym_spell.lookup_compound(
52	" ".join(list_words),
53	split_by_space=True,
54	max_edit_distance=max_edit_distance,
55	)
56	)
57	]
58	list_new = []	×
59	for i in _temp:	×
60	list_new.append(i)	×
61
62	return list_new	×
63
64
65	def correct_sent(list_words: List[str], max_edit_distance=1) -> List[str]:	×
66	return [	×
67	i[0]
68	for i in spell_sent(list_words, max_edit_distance=max_edit_distance)
69	]

PyThaiNLP / pythainlp / 11625814262

Source File Press 'n' to go to next uncovered line, 'b' for previous

Source File
Press 'n' to go to next uncovered line, 'b' for previous