explosion/spaCy
1
1
Fork 0
mirror of https://github.com/explosion/spaCy.git synced 2024-09-21 03:19:13 +03:00
Code Issues Packages Projects Releases Wiki Activity

* Remove hashing modules, which are now taken over by external lib

Browse Source
This commit is contained in:
Matthew Honnibal 2014-09-26 18:39:40 +02:00
parent bfab6403bc
commit 11a346fd5e
2 changed files with 0 additions and 88 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

21
spacy/_hashing.pxd
View File

@ -1,21 +0,0 @@
from libc.stdint cimport uint64_t
from cymem.cymem cimport Address
ctypedef uint64_t key_t
ctypedef void* val_t
cdef struct Cell:
key_t key
val_t value
cdef class PointerHash:
cdef size_t size
cdef size_t filled
cdef Cell* cells
cdef Address _mem
cdef val_t get(self, key_t key) nogil
cdef void set(self, key_t key, val_t value) except *
cdef void resize(self, size_t new_size) except *

67
spacy/_hashing.pyx
View File

@ -1,67 +0,0 @@
# cython: profile=True
from cymem.cymem cimport Address
cimport cython
cdef class PointerHash:
def __cinit__(self, size_t initial_size=8):
self.size = initial_size
self.filled = 0
# Size must be power of two
assert self.size != 0
assert self.size & (self.size - 1) == 0
self._mem = Address(self.size, sizeof(Cell))
self.cells = <Cell*>self._mem.addr
def __getitem__(self, key_t key):
assert key != 0
cdef val_t value = self.get(key)
return <size_t>value if value != NULL else None
def __setitem__(self, key_t key, size_t value):
assert key != 0 and value != 0
self.set(key, <val_t>value)
cdef val_t get(self, key_t key) nogil:
cdef Cell* cell = _find_cell(self.cells, self.size, key)
return cell.value
cdef void set(self, key_t key, val_t value) except *:
cdef Cell* cell
cell = _find_cell(self.cells, self.size, key)
if cell.key == 0:
cell.key = key
self.filled += 1
cell.value = value
if (self.filled + 1) * 4 >= (self.size * 3):
self.resize(self.size * 2)
cdef void resize(self, size_t new_size) except *:
assert new_size != 0
assert (new_size & (new_size - 1)) == 0 # Must be a power of 2
assert self.filled * 4 <= new_size * 3
cdef Cell* old_cells = self.cells
cdef size_t old_size = self.size
self.size = new_size
cdef Address new_mem = Address(new_size, sizeof(Cell))
self.cells = <Cell*>new_mem.addr
self.filled = 0
cdef size_t i
cdef size_t slot
for i in range(old_size):
if old_cells[i].key != 0:
assert old_cells[i].value != NULL, i
self.set(old_cells[i].key, old_cells[i].value)
self._mem = new_mem
@cython.cdivision
cdef inline Cell* _find_cell(Cell* cells, size_t size, key_t key) nogil:
# Modulo for powers-of-two via bitwise &
cdef size_t i = (key & (size - 1))
while cells[i].key != 0 and cells[i].key != key:
i = (i + 1) & (size - 1)
return &cells[i]