"""
Defines the Sentence type and the associated parsing and output logic.
"""
import operator
import re
from pyconll.conllable import Conllable
from pyconll.tree._treebuilder import TreeBuilder
from pyconll.unit import Token
[docs]class Sentence(Conllable):
"""
A sentence in a CoNLL-U file. A sentence consists of several components.
First, are comments. Each sentence must have two comments per UD v2
guidelines, which are sent_id and text. Comments are stored as a dict in
the meta field. For singleton comments with no key-value structure, the
value in the dict has a value of None.
Note the sent_id field is also assigned to the id property, and the text
field is assigned to the text property for usability, and their importance
as comments. The text property is read only along with the paragraph and
document id. This is because the paragraph and document id are not defined
per Sentence but across multiple sentences. Instead, these fields can be
changed through changing the metadata of the Sentences.
Then comes the token annotations. Each sentence is made up of many token
lines that provide annotation to the text provided. While a sentence usually
means a collection of tokens, in this CoNLL-U sense, it is more useful to
think of it as a collection of annotations with some associated metadata.
Therefore the text of the sentence cannot be changed with this class, only
the associated annotations can be changed.
"""
__slots__ = ['_meta', '_tokens', '_ids_to_indexes']
COMMENT_MARKER = '#'
KEY_VALUE_COMMENT_PATTERN = COMMENT_MARKER + r'\s*([^=]+?)\s*=\s*(.+)'
SINGLETON_COMMENT_PATTERN = COMMENT_MARKER + r'\s*(\S.*?)\s*$'
SENTENCE_ID_KEY = 'sent_id'
TEXT_KEY = 'text'
[docs] def __init__(self, source):
"""
Construct a Sentence object from the provided CoNLL-U string.
Args:
source: The raw CoNLL-U string to parse. Comments must precede token
lines.
Raises:
ParseError: If there is any token that was not valid.
"""
lines = source.split('\n')
self._meta = {}
self._tokens = []
self._ids_to_indexes = {}
for line in lines:
if line:
if line[0] == Sentence.COMMENT_MARKER:
kv_match = re.match(Sentence.KEY_VALUE_COMMENT_PATTERN,
line)
singleton_match = re.match(
Sentence.SINGLETON_COMMENT_PATTERN, line)
if kv_match:
k = kv_match.group(1)
v = kv_match.group(2)
self._meta[k] = v
elif singleton_match:
k = singleton_match.group(1)
self._meta[k] = None
else:
token = Token(line)
self._tokens.append(token)
if token.id is not None:
self._ids_to_indexes[token.id] = len(self._tokens) - 1
@property
def id(self):
"""
Get the sentence id.
Returns:
The sentence id. If there is none, then returns None.
"""
try:
return self._meta[Sentence.SENTENCE_ID_KEY]
except KeyError:
return None
@id.setter
def id(self, new_id):
"""
Set the sentence id.
Args:
new_id: The new id of this sentence.
"""
self._meta[Sentence.SENTENCE_ID_KEY] = new_id
@property
def text(self):
"""
Get the continuous text for this sentence. Read-only.
Returns:
The continuous text of this sentence. If none is provided in
comments, then None is returned.
"""
try:
return self._meta[Sentence.TEXT_KEY]
except KeyError:
return None
[docs] def to_tree(self):
"""
Creates a Tree data structure from the current sentence.
An empty sentence will create a Tree with no data and no children.
Returns:
A constructed Tree that represents the dependency graph of the
sentence.
"""
children_tokens = {}
root_token = None
for token in self:
parent_key = token.head
try:
children_tokens[parent_key].append(token)
except KeyError:
children_tokens[parent_key] = [token]
if token.head == '0':
root_token = token
builder = TreeBuilder()
builder.create_root(root_token)
if root_token:
Sentence._create_tree_helper(builder, self, root_token,
children_tokens)
root = builder.build()
return root
@staticmethod
def _create_tree_helper(builder, sentence, root, children_tokens):
"""
Method to create a tree from a sentence given the root token.
Args:
builder: The TreeBuilder currently being used to create the Tree.
sentence: The sentence to construct the tree from.
root: The current token we are constructing the tree at.
children_tokens: A dictionary from token id to children tokens.
Returns:
A Tree constructed given the sentence structure.
"""
try:
tokens = children_tokens[root.id]
except KeyError:
tokens = []
for token in tokens:
builder.add_child(data=token, move=True)
Sentence._create_tree_helper(builder, sentence, token,
children_tokens)
builder.move_to_parent()
[docs] def conll(self):
"""
Convert the sentence to a CoNLL-U representation.
Returns:
A string representing the Sentence in CoNLL-U format.
"""
lines = []
sorted_meta = sorted(self._meta.items(), key=operator.itemgetter(0))
for meta in sorted_meta:
if meta[1] is not None:
line = '{} {} = {}'.format(Sentence.COMMENT_MARKER, meta[0],
meta[1])
else:
line = '{} {}'.format(Sentence.COMMENT_MARKER, meta[0])
lines.append(line)
for token in self._tokens:
lines.append(token.conll())
return '\n'.join(lines)
[docs] def __iter__(self):
"""
Iterate through all the tokens in the Sentence including multiword
tokens.
"""
for token in self._tokens:
yield token
[docs] def __getitem__(self, key):
"""
Return the desired tokens from the Sentence.
Args:
key: The indicator for the tokens to return. Can either be an
integer, a string, or a slice. For an integer, the numeric
indexes of Tokens are used. For a string, the id of the Token is
used. And for a slice the start and end must be the same data
types, and can be both string and integer.
Returns:
If the key is a string then the appropriate Token. The key can also
be a slice in which case a list of tokens is provided.
"""
if isinstance(key, str):
idx = self._ids_to_indexes[key]
return self._tokens[idx]
if isinstance(key, int):
return self._tokens[key]
if isinstance(key, slice):
if isinstance(key.start, str):
start_idx = self._ids_to_indexes[key.start]
else:
start_idx = key.start
if isinstance(key.stop, str):
end_idx = self._ids_to_indexes[key.stop]
else:
end_idx = key.stop
return self._tokens[start_idx:end_idx:key.step]
raise ValueError('The key must be a str, int, or slice.')
[docs] def __len__(self):
"""
Get the length of this sentence.
Returns:
The amount of tokens in this sentence. In the CoNLL-U sense, this
includes both all the multiword tokens and their decompositions.
"""
return len(self._tokens)