# -*- coding: utf-8 -*-
"""
The implementation of tokenizer accorinding to Thai Character Clusters (TCCs)
rules purposed by `Theeramunkong et al. 2000. \
<http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.59.2548>`_
Credits:
* TCC: Jakkrit TeCho
* Grammar: Wittawat Jitkrittum (`link to the source file \
<https://github.com/wittawatj/jtcc/blob/master/TCC.g>`_)
* Python code: Korakot Chaovavanich
"""
import re
from typing import List, Set
_RE_TCC = (
"""\
เc็c
เcctาะ
เccีtยะ
เccีtย(?=[เ-ไก-ฮ]|$)
เcc็c
เcิc์c
เcิtc
เcีtยะ?
เcืtอะ?
เc[ิีุู]tย(?=[เ-ไก-ฮ]|$)
เctา?ะ?
cัtวะ
c[ัื]tc[ุิะ]?
c[ิุู]์
c[ะ-ู]t
c็
ct[ะาำ]?
แc็c
แcc์
แctะ
แcc็c
แccc์
โctะ
[เ-ไ]ct
""".replace(
"c", "[ก-ฮ]"
)
.replace("t", "[่-๋]?")
.split()
)
_PAT_TCC = re.compile("|".join(_RE_TCC))
[docs]def tcc(text: str) -> str:
"""
TCC generator, generates Thai Character Clusters
:param str text: text to be tokenized to character clusters
:return: subwords (character clusters)
:rtype: Iterator[str]
"""
if not text or not isinstance(text, str):
return ""
len_text = len(text)
p = 0
while p < len_text:
m = _PAT_TCC.match(text[p:])
if m:
n = m.span()[1]
else:
n = 1
yield text[p : p + n]
p += n
[docs]def tcc_pos(text: str) -> Set[int]:
"""
TCC positions
:param str text: text to be tokenized to character clusters
:return: list of the end position of subwords
:rtype: set[int]
"""
if not text or not isinstance(text, str):
return set()
p_set = set()
p = 0
for w in tcc(text):
p += len(w)
p_set.add(p)
return p_set
[docs]def segment(text: str) -> List[str]:
"""
Subword segmentation
:param str text: text to be tokenized to character clusters
:return: list of subwords (character clusters), tokenized from the text
:rtype: list[str]
"""
return list(tcc(text))