Home | History | Annotate | Download | only in contacts 
      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  */
     16 package com.android.providers.contacts;
     17 
     18 import com.android.providers.contacts.util.Hex;
     19 import com.google.common.annotations.VisibleForTesting;
     20 
     21 import java.text.CollationKey;
     22 import java.text.Collator;
     23 import java.text.RuleBasedCollator;
     24 import java.util.Locale;
     25 
     26 /**
     27  * Converts a name to a normalized form by removing all non-letter characters and normalizing
     28  * UNICODE according to http://unicode.org/unicode/reports/tr15
     29  */
     30 public class NameNormalizer {
     31 
     32     private static final Object sCollatorLock = new Object();
     33 
     34     private static Locale sCollatorLocale;
     35 
     36     private static RuleBasedCollator sCachedCompressingCollator;
     37     private static RuleBasedCollator sCachedComplexityCollator;
     38 
     39     /**
     40      * Ensure that the cached collators are for the current locale.
     41      */
     42     private static void ensureCollators() {
     43         final Locale locale = Locale.getDefault();
     44         if (locale.equals(sCollatorLocale)) {
     45             return;
     46         }
     47         sCollatorLocale = locale;
     48 
     49         sCachedCompressingCollator = (RuleBasedCollator) Collator.getInstance(locale);
     50         sCachedCompressingCollator.setStrength(Collator.PRIMARY);
     51         sCachedCompressingCollator.setDecomposition(Collator.CANONICAL_DECOMPOSITION);
     52 
     53         sCachedComplexityCollator = (RuleBasedCollator) Collator.getInstance(locale);
     54         sCachedComplexityCollator.setStrength(Collator.SECONDARY);
     55     }
     56 
     57     @VisibleForTesting
     58     static RuleBasedCollator getCompressingCollator() {
     59         synchronized (sCollatorLock) {
     60             ensureCollators();
     61             return sCachedCompressingCollator;
     62         }
     63     }
     64 
     65     @VisibleForTesting
     66     static RuleBasedCollator getComplexityCollator() {
     67         synchronized (sCollatorLock) {
     68             ensureCollators();
     69             return sCachedComplexityCollator;
     70         }
     71     }
     72 
     73     /**
     74      * Converts the supplied name to a string that can be used to perform approximate matching
     75      * of names.  It ignores non-letter, non-digit characters, and removes accents.
     76      */
     77     public static String normalize(String name) {
     78         CollationKey key = getCompressingCollator().getCollationKey(lettersAndDigitsOnly(name));
     79         return Hex.encodeHex(key.toByteArray(), true);
     80     }
     81 
     82     /**
     83      * Compares "complexity" of two names, which is determined by the presence
     84      * of mixed case characters, accents and, if all else is equal, length.
     85      */
     86     public static int compareComplexity(String name1, String name2) {
     87         String clean1 = lettersAndDigitsOnly(name1);
     88         String clean2 = lettersAndDigitsOnly(name2);
     89         int diff = getComplexityCollator().compare(clean1, clean2);
     90         if (diff != 0) {
     91             return diff;
     92         }
     93         // compareTo sorts uppercase first. We know that there are no non-case
     94         // differences from the above test, so we can negate here to get the
     95         // lowercase-first comparison we really want...
     96         diff = -clean1.compareTo(clean2);
     97         if (diff != 0) {
     98             return diff;
     99         }
    100         return name1.length() - name2.length();
    101     }
    102 
    103     /**
    104      * Returns a string containing just the letters and digits from the original string.
    105      */
    106     private static String lettersAndDigitsOnly(String name) {
    107         char[] letters = name.toCharArray();
    108         int length = 0;
    109         for (int i = 0; i < letters.length; i++) {
    110             final char c = letters[i];
    111             if (Character.isLetterOrDigit(c)) {
    112                 letters[length++] = c;
    113             }
    114         }
    115 
    116         if (length != letters.length) {
    117             return new String(letters, 0, length);
    118         }
    119 
    120         return name;
    121     }
    122 }
    123