spaCy/spacy/tokens/underscore.py

78 lines
2.9 KiB
Python
Raw Normal View History

2017-10-27 15:39:09 +03:00
# coding: utf8
from __future__ import unicode_literals
import functools
from ..errors import Errors
2017-10-27 15:39:09 +03:00
2017-10-07 19:56:01 +03:00
class Underscore(object):
2017-10-07 19:00:14 +03:00
doc_extensions = {}
span_extensions = {}
token_extensions = {}
2017-10-07 19:56:01 +03:00
def __init__(self, extensions, obj, start=None, end=None):
object.__setattr__(self, '_extensions', extensions)
2017-10-07 19:00:14 +03:00
object.__setattr__(self, '_obj', obj)
# Assumption is that for doc values, _start and _end will both be None
# Span will set non-None values for _start and _end
# Token will have _start be non-None, _end be None
# This lets us key everything into the doc.user_data dictionary,
# (see _get_key), and lets us use a single Underscore class.
object.__setattr__(self, '_doc', obj.doc)
object.__setattr__(self, '_start', start)
2017-10-07 19:56:01 +03:00
object.__setattr__(self, '_end', end)
2017-10-07 19:00:14 +03:00
def __getattr__(self, name):
2017-10-07 19:56:01 +03:00
if name not in self._extensions:
raise AttributeError(Errors.E046.format(name=name))
2017-10-07 19:56:01 +03:00
default, method, getter, setter = self._extensions[name]
2017-10-07 19:00:14 +03:00
if getter is not None:
return getter(self._obj)
elif method is not None:
return functools.partial(method, self._obj)
2017-10-07 19:00:14 +03:00
else:
return self._doc.user_data.get(self._get_key(name), default)
def __setattr__(self, name, value):
2017-10-07 19:56:01 +03:00
if name not in self._extensions:
raise AttributeError(Errors.E047.format(name=name))
2017-10-07 19:56:01 +03:00
default, method, getter, setter = self._extensions[name]
2017-10-07 19:00:14 +03:00
if setter is not None:
return setter(self._obj, value)
else:
self._doc.user_data[self._get_key(name)] = value
def set(self, name, value):
return self.__setattr__(name, value)
def get(self, name):
return self.__getattr__(name)
def has(self, name):
return name in self._extensions
2017-10-07 19:00:14 +03:00
def _get_key(self, name):
return ('._.', name, self._start, self._end)
def get_ext_args(**kwargs):
"""Validate and convert arguments. Reused in Doc, Token and Span."""
default = kwargs.get('default')
getter = kwargs.get('getter')
setter = kwargs.get('setter')
method = kwargs.get('method')
if getter is None and setter is not None:
raise ValueError(Errors.E089)
valid_opts = ('default' in kwargs, method is not None, getter is not None)
nr_defined = sum(t is True for t in valid_opts)
if nr_defined != 1:
raise ValueError(Errors.E083.format(nr_defined=nr_defined))
if setter is not None and not hasattr(setter, '__call__'):
raise ValueError(Errors.E091.format(name='setter', value=repr(setter)))
if getter is not None and not hasattr(getter, '__call__'):
raise ValueError(Errors.E091.format(name='getter', value=repr(getter)))
if method is not None and not hasattr(method, '__call__'):
raise ValueError(Errors.E091.format(name='method', value=repr(method)))
return (default, method, getter, setter)