mirror of
https://github.com/explosion/spaCy.git
synced 2024-12-27 02:16:32 +03:00
34 lines
799 B
Cython
34 lines
799 B
Cython
from cymem.cymem cimport Pool
|
|
from preshed.maps cimport PreshMap
|
|
from murmurhash.mrmr cimport hash64
|
|
from .typedefs cimport attr_t
|
|
|
|
from libc.stdint cimport int64_t
|
|
|
|
from .typedefs cimport hash_t
|
|
|
|
DEF UINT64_MAX = 18446744073709551615
|
|
|
|
cpdef hash_t hash_string(unicode string) except 0
|
|
|
|
|
|
ctypedef union Utf8Str:
|
|
unsigned char[8] s
|
|
unsigned char* p
|
|
|
|
|
|
cdef class StringStore:
|
|
cdef Pool mem
|
|
cdef Utf8Str* c
|
|
cdef int64_t size
|
|
|
|
cdef public PreshMap _map
|
|
cdef int64_t _resize_at
|
|
cdef PreshMap oov_maps
|
|
|
|
cpdef int remove_oov_map(self, Pool mem) except -1
|
|
|
|
cpdef attr_t intern(self, basestring base_string, Pool mem=*) except -1
|
|
cdef const Utf8Str* _intern_utf8(self, const unsigned char* utf8_string,
|
|
int length) except NULL
|