1/*
2 * Copyright (C) 2009 The Android Open Source Project
3 *
4 * Licensed under the Apache License, Version 2.0 (the "License");
5 * you may not use this file except in compliance with the License.
6 * You may obtain a copy of the License at
7 *
8 *      http://www.apache.org/licenses/LICENSE-2.0
9 *
10 * Unless required by applicable law or agreed to in writing, software
11 * distributed under the License is distributed on an "AS IS" BASIS,
12 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
13 * See the License for the specific language governing permissions and
14 * limitations under the License
15 */
16package com.android.providers.contacts;
17
18import java.util.Locale;
19import java.text.Collator;
20import java.text.CollationKey;
21import java.text.RuleBasedCollator;
22
23/**
24 * Converts a name to a normalized form by removing all non-letter characters and normalizing
25 * UNICODE according to http://unicode.org/unicode/reports/tr15
26 */
27public class NameNormalizer {
28
29    private static final RuleBasedCollator sCompressingCollator;
30    static {
31        sCompressingCollator = (RuleBasedCollator)Collator.getInstance(Locale.getDefault());
32        sCompressingCollator.setStrength(Collator.PRIMARY);
33        sCompressingCollator.setDecomposition(Collator.CANONICAL_DECOMPOSITION);
34    }
35
36    private static final RuleBasedCollator sComplexityCollator;
37    static {
38        sComplexityCollator = (RuleBasedCollator)Collator.getInstance(Locale.getDefault());
39        sComplexityCollator.setStrength(Collator.SECONDARY);
40    }
41
42    /**
43     * Converts the supplied name to a string that can be used to perform approximate matching
44     * of names.  It ignores non-letter, non-digit characters, and removes accents.
45     */
46    public static String normalize(String name) {
47        CollationKey key = sCompressingCollator.getCollationKey(lettersAndDigitsOnly(name));
48        return Hex.encodeHex(key.toByteArray(), true);
49    }
50
51    /**
52     * Compares "complexity" of two names, which is determined by the presence
53     * of mixed case characters, accents and, if all else is equal, length.
54     */
55    public static int compareComplexity(String name1, String name2) {
56        String clean1 = lettersAndDigitsOnly(name1);
57        String clean2 = lettersAndDigitsOnly(name2);
58        int diff = sComplexityCollator.compare(clean1, clean2);
59        if (diff != 0) {
60            return diff;
61        }
62        // compareTo sorts uppercase first. We know that there are no non-case
63        // differences from the above test, so we can negate here to get the
64        // lowercase-first comparison we really want...
65        diff = -clean1.compareTo(clean2);
66        if (diff != 0) {
67            return diff;
68        }
69        return name1.length() - name2.length();
70    }
71
72    /**
73     * Returns a string containing just the letters and digits from the original string.
74     */
75    private static String lettersAndDigitsOnly(String name) {
76        char[] letters = name.toCharArray();
77        int length = 0;
78        for (int i = 0; i < letters.length; i++) {
79            final char c = letters[i];
80            if (Character.isLetterOrDigit(c)) {
81                letters[length++] = c;
82            }
83        }
84
85        if (length != letters.length) {
86            return new String(letters, 0, length);
87        }
88
89        return name;
90    }
91}
92