1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
|
#!/usr/bin/env python
# -*- coding:utf-8 -*-
"""
A Unicode sample.
We assume that the source code is written in UTF-8 encoding (see the
encoding declaration in line 2). We can use 8-bit strings as they are
with SimString.
"""
import simstring
# Open a SimString database for writing with Unicode mode.
db = simstring.writer('sample_unicode.db', 3, False, True)
# Write a string, and close the database.
db.insert('スパゲティ')
db.close()
# Open the SimString database for reading.
db = simstring.reader('sample_unicode.db')
# Set a similarity measure and threshold.
db.measure = simstring.cosine
db.threshold = 0.6
# Use an 8-bit string encoded in UTF-8.
print ' '.join(db.retrieve('スパゲティー'))
# Convert a Unicode object into an UTF-8 query string.
print ' '.join(db.retrieve(u'スパゲティー'.encode('utf-8')))
|
