Rooba
/
spaCy
mirror of https://github.com/explosion/spaCy.git
1
1
Fork 0
Code Issues Packages Projects Releases Wiki Activity

* Fix negative indices in tokens

This commit is contained in:
Matthew Honnibal 2015-01-20 01:16:29 +11:00
parent b65b0c07bf
commit f149259bf5
1 changed files with 17 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

17
spacy/tokens.pyx
View File

@ -81,12 +81,29 @@ cdef class Tokens:
self.max_length = size self.max_length = size
self.length = 0 self.length = 0
def sentences(self):
cdef int i
sentences = []
sent = Tokens(self.vocab)
cdef attr_t period = self.vocab.strings['.']
cdef attr_t question = self.vocab.strings['?']
cdef attr_t exclamation = self.vocab.strings['!']
for i in range(self.length):
idx = sent.push_back(idx, &self.data[i])
if self.data[i].lex.sic == period or self.data[i].lex.sic == exclamation or \
self.data[i].lex.sic == question:
sentences.append(sent)
sent = Tokens(self.vocab)
return sentences
def __getitem__(self, i): def __getitem__(self, i):
"""Retrieve a token. """Retrieve a token.
Returns: Returns:
token (Token): token (Token):
""" """
if i < 0:
i = self.length - i
bounds_check(i, self.length, PADDING) bounds_check(i, self.length, PADDING)
return Token(self, i) return Token(self, i)