1*912701f9SAndroid Build Coastguard Worker<?xml version="1.0" encoding="UTF-8" ?> 2*912701f9SAndroid Build Coastguard Worker<!DOCTYPE supplementalData SYSTEM "../../common/dtd/ldmlSupplemental.dtd"> 3*912701f9SAndroid Build Coastguard Worker<!-- 4*912701f9SAndroid Build Coastguard WorkerCopyright © 1991-2013 Unicode, Inc. 5*912701f9SAndroid Build Coastguard WorkerCLDR data files are interpreted according to the LDML specification (http://unicode.org/reports/tr35/) 6*912701f9SAndroid Build Coastguard WorkerFor terms of use, see http://www.unicode.org/copyright.html 7*912701f9SAndroid Build Coastguard Worker--> 8*912701f9SAndroid Build Coastguard Worker<supplementalData> 9*912701f9SAndroid Build Coastguard Worker <version number="$Revision$"/> 10*912701f9SAndroid Build Coastguard Worker <transforms> 11*912701f9SAndroid Build Coastguard Worker <transform source="Latin" target="NumericPinyin" direction="both" alias="und-pinyin-t-d0-npinyin" backwardAlias="und-pinyin-t-s0-npinyin"> 12*912701f9SAndroid Build Coastguard Worker <tRule><![CDATA[ 13*912701f9SAndroid Build Coastguard Worker# According to the pinyin definitions I've been able to find: 14*912701f9SAndroid Build Coastguard Worker# 'a', 'e' are the preferred bases 15*912701f9SAndroid Build Coastguard Worker# otherwise 'o' 16*912701f9SAndroid Build Coastguard Worker# otherwise last vowel 17*912701f9SAndroid Build Coastguard Worker# The trailing form of syllables are the following: 18*912701f9SAndroid Build Coastguard Worker# "a", "ai", "ao", "an", "ang", 19*912701f9SAndroid Build Coastguard Worker# "o", "ou", "ong", 20*912701f9SAndroid Build Coastguard Worker# "e", "ei", "er", "en", "eng", 21*912701f9SAndroid Build Coastguard Worker# "i", "ia", "iao", "ie", "iu", "ian", "in", "iang", "ing", "iong", 22*912701f9SAndroid Build Coastguard Worker# "u", "ua", "uo", "uai", "ui", "uan", "un", "uang", "ueng", 23*912701f9SAndroid Build Coastguard Worker# "ü", "üe", "üan", "ün" 24*912701f9SAndroid Build Coastguard Worker# so the letters the tone will 'hop' are: 25*912701f9SAndroid Build Coastguard Worker::NFD (NFC); 26*912701f9SAndroid Build Coastguard Worker$tone = [̄́̌̀̆] ; 27*912701f9SAndroid Build Coastguard Worker# Move the tone to the end of a syllable, and convert to number 28*912701f9SAndroid Build Coastguard Workere {($tone) r} → r &Pinyin-NumericPinyin($1); 29*912701f9SAndroid Build Coastguard Worker($tone) ( [i o n u {o n} {n g}]) → $2 &Pinyin-NumericPinyin($1); 30*912701f9SAndroid Build Coastguard Worker($tone) → &Pinyin-NumericPinyin($1); 31*912701f9SAndroid Build Coastguard Worker# The following backs up until it finds the right vowel, then deposits the tone 32*912701f9SAndroid Build Coastguard Worker$vowel = [aAeEiIoOuU {ü} {Ü} vV]; 33*912701f9SAndroid Build Coastguard Worker$consonant = [[a-z A-Z] - [$vowel]]; 34*912701f9SAndroid Build Coastguard Worker$digit = [1-5]; 35*912701f9SAndroid Build Coastguard Worker$1 &NumericPinyin-Pinyin($3) $2 ← ([aAeE]) ($vowel* $consonant*) ($digit); 36*912701f9SAndroid Build Coastguard Worker$1 &NumericPinyin-Pinyin($3) $2 ← ([oO]) ([$vowel-[aeAE]]* $consonant*) ($digit); 37*912701f9SAndroid Build Coastguard Worker$1 &NumericPinyin-Pinyin($3) $2 ← ($vowel) ($consonant*) ($digit); 38*912701f9SAndroid Build Coastguard Worker&NumericPinyin-Pinyin($1) ← [:letter:] {($digit)}; 39*912701f9SAndroid Build Coastguard Worker::NFC (NFD); 40*912701f9SAndroid Build Coastguard Worker ]]></tRule> 41*912701f9SAndroid Build Coastguard Worker </transform> 42*912701f9SAndroid Build Coastguard Worker </transforms> 43*912701f9SAndroid Build Coastguard Worker</supplementalData> 44