explosion/spaCy
1
1
Fork 0
mirror of https://github.com/explosion/spaCy.git synced 2025-02-04 21:50:35 +03:00
Code Issues Packages Projects Releases Wiki Activity

* Preparations in place to handle hyphenation etc

Browse Source
This commit is contained in:
Matthew Honnibal 2014-10-10 20:23:23 +11:00
parent ff79dbac2e
commit 868e558037
2 changed files with 10 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
spacy/lang.pxd
View File

@ -52,6 +52,8 @@ cdef class Language:
cpdef Lexeme lookup(self, unicode text) cpdef Lexeme lookup(self, unicode text)
cdef int _tokenize(self, vector[LexemeC*] *tokens_v, String* string) except -1 cdef int _tokenize(self, vector[LexemeC*] *tokens_v, String* string) except -1
cdef int _split_body_token(self, vector[LexemeC*] *tokens, String* string) except -1
cdef int _find_prefix(self, Py_UNICODE* characters, size_t length) cdef int _find_prefix(self, Py_UNICODE* characters, size_t length)
cdef int _find_suffix(self, Py_UNICODE* characters, size_t length) cdef int _find_suffix(self, Py_UNICODE* characters, size_t length)

13
spacy/lang.pyx
View File

@ -155,16 +155,14 @@ cdef class Language:
cdef LexemeC* lexeme cdef LexemeC* lexeme
for lexeme in deref(prefixes): for lexeme in deref(prefixes):
tokens.push_back(lexeme) tokens.push_back(lexeme)
if string.n != 0: if not _extend_from_map(tokens, string, self.specials):
if not _extend_from_map(tokens, string, self.specials): self._split_body_token(tokens, string)
tokens.push_back(self.lexicon.get(string))
cdef vector[LexemeC*].reverse_iterator it = suffixes.rbegin() cdef vector[LexemeC*].reverse_iterator it = suffixes.rbegin()
while it != suffixes.rend(): while it != suffixes.rend():
tokens.push_back(deref(it)) tokens.push_back(deref(it))
preinc(it) preinc(it)
cdef int _save_cached(self, vector[LexemeC*] *tokens, cdef int _save_cached(self, vector[LexemeC*] *tokens, uint64_t key, size_t n) except -1:
uint64_t key, size_t n) except -1:
assert tokens.size() > n assert tokens.size() > n
lexemes = <LexemeC**>self._mem.alloc((tokens.size() - n) + 1, sizeof(LexemeC**)) lexemes = <LexemeC**>self._mem.alloc((tokens.size() - n) + 1, sizeof(LexemeC**))
cdef size_t i, j cdef size_t i, j
@ -172,6 +170,9 @@ cdef class Language:
lexemes[i] = tokens.at(j) lexemes[i] = tokens.at(j)
lexemes[i + 1] = NULL lexemes[i + 1] = NULL
self.cache.set(key, lexemes) self.cache.set(key, lexemes)
cdef int _split_body_token(self, vector[LexemeC*] *tokens, String* string) except -1:
tokens.push_back(self.lexicon.get(string))
cdef int _find_prefix(self, Py_UNICODE* chars, size_t length) except -1: cdef int _find_prefix(self, Py_UNICODE* chars, size_t length) except -1:
cdef unicode string = chars[:length] cdef unicode string = chars[:length]
@ -255,6 +256,8 @@ cdef class Lexicon:
cdef int _extend_from_map(vector[LexemeC*] *tokens, String* string, PreshMap map_) except -1: cdef int _extend_from_map(vector[LexemeC*] *tokens, String* string, PreshMap map_) except -1:
if string.n == 0:
return 1
lexemes = <LexemeC**>map_.get(string.key) lexemes = <LexemeC**>map_.get(string.key)
if lexemes == NULL: if lexemes == NULL:
return 0 return 0