textblob.base — TextBlob 0.19.0 documentation (original) (raw)

Source code for textblob.base

"""Abstract base classes for models (taggers, noun phrase extractors, etc.) which define the interface for descendant classes.

.. versionchanged:: 0.7.0 All base classes are defined in the same module, textblob.base. """

from future import annotations

from abc import ABCMeta, abstractmethod from typing import TYPE_CHECKING

import nltk

if TYPE_CHECKING: from typing import Any, AnyStr

POS TAGGERS

[docs] class BaseTagger(metaclass=ABCMeta): """Abstract tagger class from which all taggers inherit from. All descendants must implement a tag() method. """

[docs] @abstractmethod def tag(self, text: str, tokenize=True) -> list[tuple[str, str]]: """Return a list of tuples of the form (word, tag) for a given set of text or BaseBlob instance. """ ...

NOUN PHRASE EXTRACTORS

TOKENIZERS

[docs] class BaseTokenizer(nltk.tokenize.api.TokenizerI, metaclass=ABCMeta): # pyright: ignore """Abstract base class from which all Tokenizer classes inherit. Descendant classes must implement a tokenize(text) method that returns a list of noun phrases as strings. """

[docs] @abstractmethod def tokenize(self, text: str) -> list[str]: """Return a list of tokens (strings) for a body of text.

    :rtype: list
    """
    ...

[docs] def itokenize(self, text: str, *args, **kwargs): """Return a generator that generates tokens "on-demand".

    .. versionadded:: 0.6.0

    :rtype: generator
    """
    return (t for t in self.tokenize(text, *args, **kwargs))

SENTIMENT ANALYZERS

DISCRETE = "ds" CONTINUOUS = "co"

[docs] class BaseSentimentAnalyzer(metaclass=ABCMeta): """Abstract base class from which all sentiment analyzers inherit. Should implement an analyze(text) method which returns either the results of analysis. """

_trained: bool

kind = DISCRETE

def __init__(self):
    self._trained = False

def train(self):
    # Train me
    self._trained = True

[docs] @abstractmethod def analyze(self, text) -> Any: """Return the result of of analysis. Typically returns either a tuple, float, or dictionary. """ # Lazily train the classifier if not self._trained: self.train() # Analyze text return None

PARSERS

[docs] class BaseParser(metaclass=ABCMeta): """Abstract parser class from which all parsers inherit from. All descendants must implement a parse() method. """

[docs] @abstractmethod def parse(self, text: AnyStr): """Parses the text.""" ...