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

* Fix bug in tokenizer that caused new tokens to be added for affixes

This commit is contained in:
Matthew Honnibal 2016-02-21 23:17:47 +00:00
parent 9d8966a2c0
commit 141639ea3a
2 changed files with 9 additions and 9 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
spacy/tokenizer.pxd
View File

@ -25,7 +25,7 @@ cdef class Tokenizer:
cdef int _try_cache(self, hash_t key, Doc tokens) except -1 cdef int _try_cache(self, hash_t key, Doc tokens) except -1
cdef int _tokenize(self, Doc tokens, unicode span, hash_t key) except -1 cdef int _tokenize(self, Doc tokens, unicode span, hash_t key) except -1
cdef unicode _split_affixes(self, unicode string, vector[LexemeC*] *prefixes, cdef unicode _split_affixes(self, Pool mem, unicode string, vector[LexemeC*] *prefixes,
vector[LexemeC*] *suffixes) vector[LexemeC*] *suffixes)
cdef int _attach_tokens(self, Doc tokens, unicode string, cdef int _attach_tokens(self, Doc tokens, unicode string,
vector[LexemeC*] *prefixes, vector[LexemeC*] *suffixes) except -1 vector[LexemeC*] *prefixes, vector[LexemeC*] *suffixes) except -1

16
spacy/tokenizer.pyx
View File

@ -155,11 +155,11 @@ cdef class Tokenizer:
cdef vector[LexemeC*] suffixes cdef vector[LexemeC*] suffixes
cdef int orig_size cdef int orig_size
orig_size = tokens.length orig_size = tokens.length
span = self._split_affixes(span, &prefixes, &suffixes) span = self._split_affixes(tokens.mem, span, &prefixes, &suffixes)
self._attach_tokens(tokens, span, &prefixes, &suffixes) self._attach_tokens(tokens, span, &prefixes, &suffixes)
self._save_cached(&tokens.c[orig_size], orig_key, tokens.length - orig_size) self._save_cached(&tokens.c[orig_size], orig_key, tokens.length - orig_size)
cdef unicode _split_affixes(self, unicode string, vector[const LexemeC*] *prefixes, cdef unicode _split_affixes(self, Pool mem, unicode string, vector[const LexemeC*] *prefixes,
vector[const LexemeC*] *suffixes): vector[const LexemeC*] *suffixes):
cdef size_t i cdef size_t i
cdef unicode prefix cdef unicode prefix
@ -176,7 +176,7 @@ cdef class Tokenizer:
# Check whether we've hit a special-case # Check whether we've hit a special-case
if minus_pre and self._specials.get(hash_string(minus_pre)) != NULL: if minus_pre and self._specials.get(hash_string(minus_pre)) != NULL:
string = minus_pre string = minus_pre
prefixes.push_back(self.vocab.get(self.vocab.mem, prefix)) prefixes.push_back(self.vocab.get(mem, prefix))
break break
suf_len = self.find_suffix(string) suf_len = self.find_suffix(string)
if suf_len != 0: if suf_len != 0:
@ -185,18 +185,18 @@ cdef class Tokenizer:
# Check whether we've hit a special-case # Check whether we've hit a special-case
if minus_suf and (self._specials.get(hash_string(minus_suf)) != NULL): if minus_suf and (self._specials.get(hash_string(minus_suf)) != NULL):
string = minus_suf string = minus_suf
suffixes.push_back(self.vocab.get(self.vocab.mem, suffix)) suffixes.push_back(self.vocab.get(mem, suffix))
break break
if pre_len and suf_len and (pre_len + suf_len) <= len(string): if pre_len and suf_len and (pre_len + suf_len) <= len(string):
string = string[pre_len:-suf_len] string = string[pre_len:-suf_len]
prefixes.push_back(self.vocab.get(self.vocab.mem, prefix)) prefixes.push_back(self.vocab.get(mem, prefix))
suffixes.push_back(self.vocab.get(self.vocab.mem, suffix)) suffixes.push_back(self.vocab.get(mem, suffix))
elif pre_len: elif pre_len:
string = minus_pre string = minus_pre
prefixes.push_back(self.vocab.get(self.vocab.mem, prefix)) prefixes.push_back(self.vocab.get(mem, prefix))
elif suf_len: elif suf_len:
string = minus_suf string = minus_suf
suffixes.push_back(self.vocab.get(self.vocab.mem, suffix)) suffixes.push_back(self.vocab.get(mem, suffix))
if string and (self._specials.get(hash_string(string)) != NULL): if string and (self._specials.get(hash_string(string)) != NULL):
break break
return string return string