知微
fa45d8aa5f
- health_checklist.json: 192.168.1.122→node122
- ocr_client.py: docstring IP→node122
- docs/market-data-requirements.md: IP→node122
- 所有API调用通过ProxyHandler({})绕过系统代理
Privoxy对node122:18003返回500，直连正常
85 lines
2.2 KiB
Python
85 lines
2.2 KiB
Python
# Natural Language Toolkit: Tokenizer Interface
|
|
#
|
|
# Copyright (C) 2001-2026 NLTK Project
|
|
# Author: Edward Loper <edloper@gmail.com>
|
|
# Steven Bird <stevenbird1@gmail.com>
|
|
# URL: <https://www.nltk.org/>
|
|
# For license information, see LICENSE.TXT
|
|
|
|
"""
|
|
Tokenizer Interface
|
|
"""
|
|
|
|
from abc import ABC, abstractmethod
|
|
from collections.abc import Iterator
|
|
from typing import List, Tuple
|
|
|
|
from nltk.internals import overridden
|
|
from nltk.tokenize.util import string_span_tokenize
|
|
|
|
|
|
class TokenizerI(ABC):
|
|
"""
|
|
A processing interface for tokenizing a string.
|
|
Subclasses must define ``tokenize()`` or ``tokenize_sents()`` (or both).
|
|
"""
|
|
|
|
@abstractmethod
|
|
def tokenize(self, s: str) -> list[str]:
|
|
"""
|
|
Return a tokenized copy of *s*.
|
|
|
|
:rtype: List[str]
|
|
"""
|
|
if overridden(self.tokenize_sents):
|
|
return self.tokenize_sents([s])[0]
|
|
|
|
def span_tokenize(self, s: str) -> Iterator[tuple[int, int]]:
|
|
"""
|
|
Identify the tokens using integer offsets ``(start_i, end_i)``,
|
|
where ``s[start_i:end_i]`` is the corresponding token.
|
|
|
|
:rtype: Iterator[Tuple[int, int]]
|
|
"""
|
|
raise NotImplementedError()
|
|
|
|
def tokenize_sents(self, strings: list[str]) -> list[list[str]]:
|
|
"""
|
|
Apply ``self.tokenize()`` to each element of ``strings``. I.e.:
|
|
|
|
return [self.tokenize(s) for s in strings]
|
|
|
|
:rtype: List[List[str]]
|
|
"""
|
|
return [self.tokenize(s) for s in strings]
|
|
|
|
def span_tokenize_sents(
|
|
self, strings: list[str]
|
|
) -> Iterator[list[tuple[int, int]]]:
|
|
"""
|
|
Apply ``self.span_tokenize()`` to each element of ``strings``. I.e.:
|
|
|
|
return [self.span_tokenize(s) for s in strings]
|
|
|
|
:yield: List[Tuple[int, int]]
|
|
"""
|
|
for s in strings:
|
|
yield list(self.span_tokenize(s))
|
|
|
|
|
|
class StringTokenizer(TokenizerI):
|
|
"""A tokenizer that divides a string into substrings by splitting
|
|
on the specified string (defined in subclasses).
|
|
"""
|
|
|
|
@property
|
|
@abstractmethod
|
|
def _string(self):
|
|
raise NotImplementedError
|
|
|
|
def tokenize(self, s):
|
|
return s.split(self._string)
|
|
|
|
def span_tokenize(self, s):
|
|
yield from string_span_tokenize(s, self._string)
|