summaryrefslogtreecommitdiffstats
diff options
context:
space:
mode:
authorYuanQY <yuanqingyun@gmail.com>2013-09-09 16:32:56 +0800
committerSteve Kondik <shade@chemlab.org>2016-04-14 00:45:44 -0700
commita629c557c9990f32eb8b9dd00a68dc38dcdb9c70 (patch)
tree8a5ae63a4b92c8990bb987e94c4d64b1d939d65e
parentb3bdb841725c4fd1f3652b762156c7f813a22236 (diff)
downloadpackages_providers_ContactsProvider-a629c557c9990f32eb8b9dd00a68dc38dcdb9c70.tar.gz
packages_providers_ContactsProvider-a629c557c9990f32eb8b9dd00a68dc38dcdb9c70.tar.bz2
packages_providers_ContactsProvider-a629c557c9990f32eb8b9dd00a68dc38dcdb9c70.zip
ContactsProvider: Fix polyphones in Chinese.
Merge the resolution for polyphones in Chinese from https://github.com/ShenduOS/android_packages_providers_ContactsProvider/commit/d232513ba086fd0508def8ef0f7fb7d9cec5d171, thanks the Shendu OS team. Patch Set 2: Format the code style. Change-Id: I84d5921ec43502bcfd2a1aefd4cae9fd0600c9d1 (cherry picked from commit 61d7ec35b4c2fbbe6f8ed86e14e39900c301f00a)
-rw-r--r--src/com/android/providers/contacts/HanziToPinyin.java67
1 files changed, 65 insertions, 2 deletions
diff --git a/src/com/android/providers/contacts/HanziToPinyin.java b/src/com/android/providers/contacts/HanziToPinyin.java
index d140439b..a7d90c28 100644
--- a/src/com/android/providers/contacts/HanziToPinyin.java
+++ b/src/com/android/providers/contacts/HanziToPinyin.java
@@ -19,6 +19,7 @@ package com.android.providers.contacts;
import android.text.TextUtils;
import android.util.Log;
+import java.text.Collator;
import java.util.ArrayList;
import java.util.Locale;
@@ -37,6 +38,32 @@ public class HanziToPinyin {
private Transliterator mPinyinTransliterator;
private Transliterator mAsciiTransliterator;
+ private static final Collator COLLATOR = Collator.getInstance(Locale.CHINA);
+ private static final String[] MULTINAMES = {
+ "\u590f","\u77bf","\u66fe","\u77f3","\u89e3","\u85cf","\u7fdf","\u90fd",
+ "\u516d","\u8584","\u8d3e","\u5c45","\u67e5","\u76db","\u5854","\u548c",
+ "\u84dd","\u6bb7","\u4e7e","\u9646","\u4e5c","\u961a","\u53f6","\u5f3a",
+ "\u6c64","\u4e07","\u6c88","\u4ec7","\u5357","\u5355","\u535c","\u9e1f",
+ "\u601d","\u5bfb","\u65bc","\u4f59","\u6d45","\u6d63","\u65e0","\u4fe1",
+ "\u8a31","\u9f50","\u4fde","\u82e5", };
+
+ private static final byte[][] MULTIPYS = {
+ { 88, 73, 65, 0, 0, 0 }, { 81, 85, 0, 0, 0, 0 }, { 90, 69, 78, 71, 0, 0 },
+ { 83, 72, 73, 0, 0, 0 }, { 88, 73, 69, 0, 0, 0 }, { 90, 65, 78, 71, 0, 0 },
+ { 90, 72, 65, 73, 0, 0 }, { 68, 85, 0, 0, 0, 0 }, { 76, 85, 0, 0, 0, 0 },
+ { 66, 79, 0, 0, 0, 0 }, { 74, 73, 65, 0, 0, 0 }, { 74, 85, 0, 0, 0, 0 },
+ { 90, 72, 65, 0, 0, 0 }, { 83, 72, 69, 78, 71, 0 }, { 84, 65, 0, 0, 0, 0 },
+ { 72, 69, 0, 0, 0, 0 }, { 76, 65, 78, 0, 0, 0 }, { 89, 73, 78, 0, 0, 0 },
+ { 81, 73, 65, 78, 0, 0 }, { 76, 85, 0, 0, 0, 0 }, { 78, 73, 69, 0, 0, 0 },
+ { 75, 65, 78, 0, 0, 0 }, { 89, 69, 0, 0, 0, 0 }, { 81, 73, 65, 78, 71, 0 },
+ { 84, 65, 78, 71, 0, 0 }, { 87, 65, 78, 0, 0, 0 }, { 83, 72, 69, 78, 0, 0 },
+ { 81, 73, 85, 0, 0, 0 }, { 78, 65, 78, 0, 0, 0 }, { 83, 72, 65, 78, 0, 0 },
+ { 66, 85, 0, 0, 0, 0 }, { 78, 73, 65, 79, 0, 0 }, { 83, 73, 0, 0, 0, 0 },
+ { 88, 85, 78, 0, 0, 0 }, { 89, 85, 0, 0, 0, 0 }, { 89, 85, 0, 0, 0, 0 },
+ { 81, 73, 65, 78, 0, 0 }, { 87, 65, 78, 0, 0, 0 }, { 87, 85, 0, 0, 0, 0 },
+ { 88, 73, 78, 0, 0, 0 }, { 88, 85, 0, 0, 0, 0 }, { 81, 73, 0, 0, 0, 0 },
+ { 89, 85, 0, 0, 0, 0 }, { 82, 85, 79, 0, 0, 0 }, };
+
public static class Token {
/**
* Separator between target string for each source char
@@ -94,7 +121,34 @@ public class HanziToPinyin {
}
}
- private void tokenize(char character, Token token) {
+ /**
+ * Check if the first name is multi-pinyin
+ *
+ * @return right pinyin for this first name
+ */
+ private static String getMPinyin(final String firstName) {
+ int offset, cmp;
+ cmp = -1;
+
+ for (offset = 0; offset < MULTINAMES.length; offset++) {
+ cmp = COLLATOR.compare(firstName, MULTINAMES[offset]);
+ if (cmp == 0) {
+ break;
+ }
+ }
+
+ if (cmp != 0) {
+ return null;
+ } else {
+ StringBuilder pinyin = new StringBuilder();
+ for (int j = 0; j < MULTIPYS[offset].length && MULTIPYS[offset][j] != 0; j++) {
+ pinyin.append((char) MULTIPYS[offset][j]);
+ }
+ return pinyin.toString();
+ }
+ }
+
+ private void tokenize(char character, Token token, boolean isFirstName) {
token.source = Character.toString(character);
// ASCII
@@ -113,6 +167,14 @@ public class HanziToPinyin {
}
token.type = Token.PINYIN;
+
+ if (isFirstName) {
+ token.target = getMPinyin(Character.toString(character));
+ if (token.target != null) {
+ return;
+ }
+ }
+
token.target = mPinyinTransliterator.transliterate(token.source);
if (TextUtils.isEmpty(token.target) ||
TextUtils.equals(token.source, token.target)) {
@@ -144,6 +206,7 @@ public class HanziToPinyin {
final StringBuilder sb = new StringBuilder();
int tokenType = Token.LATIN;
Token token = new Token();
+ boolean firstname;
// Go through the input, create a new token when
// a. Token type changed
@@ -156,7 +219,7 @@ public class HanziToPinyin {
addToken(sb, tokens, tokenType);
}
} else {
- tokenize(character, token);
+ tokenize(character, token, i == 0);
if (token.type == Token.PINYIN) {
if (sb.length() > 0) {
addToken(sb, tokens, tokenType);