organicmaps/icu
13
Fork 0
mirror of https://github.com/unicode-org/icu.git synced 2025-04-17 02:37:25 +00:00
Code Issues 1 Projects Releases Wiki Activity Actions 8

ICU-872 Fix another round-trip problem

Browse source 
X-SVN-Rev: 3977
This commit is contained in:
Alan Liu 2001-03-08 21:20:10 +00:00
parent dbdfc915ea
commit 4187ad0621
1 changed files with 16 additions and 13 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

29
icu4c/data/ljamo.txt
View file

@ -5,11 +5,11 @@
//--------------------------------------------------------------------
// THIS IS A MACHINE-GENERATED FILE
// Tool: dumpICUrules.bat
// Source: /icu4j/src/com/ibm/text/resources/Transliterator_Latin_Jamo.utf8.txt
// Date: Wed Feb 28 14:45:49 2001
// Source: \icu4j\src\com\ibm\text\resources/Transliterator_Latin_Jamo.utf8.txt
// Date: Thu Mar 8 12:27:24 2001
//--------------------------------------------------------------------
// Latin-Jamo
// Latin_Jamo
ljamo {
Rule {
@ -209,20 +209,23 @@ ljamo {
// "a" by itself and "e" by itself are vowels, then we want to map A E
// to "a-e" so as not to round trip to AE. However, in the text Ki EO
// IEUNG E we don't need to map to "keo-e". "keoe" suffices. For
// vowels of the form "aei", in theory both "ae" + "i" and "a" + "ei"
// must be tested, but in practice only the former occurs.
// vowels of the form "aei", both "ae" + "i" and "a" + "ei" must be
// tested. NOTE: These rules used to have a left context of
// $latinInitial instead of [^$latinMedial]. The problem with this is
// sequences where an initial IEUNG is transliterated away:
// (IEUNG)(A)(IEUNG)(EO) => aeo => (IEUNG)(AE)(IEUNG)(O)
"'-' < $latinInitial [y w] e {} [$O $OE];"
"'-' < $latinInitial e {} [$O $OE $U];"
"'-' < $latinInitial [o a] {} [$E $EO $EU];"
"'-' < $latinInitial [w y] a {} [$E $EO $EU];"
"'-' < [^$latinMedial] [y w] e {} [$O $OE];"
"'-' < [^$latinMedial] e {} [$O $OE $U];"
"'-' < [^$latinMedial] [o a] {} [$E $EO $EU];"
"'-' < [^$latinMedial] [w y] a {} [$E $EO $EU];"
// Similar to the above, but with an intervening $IEUNG.
"'-' < $latinInitial [y w] e {} $IEUNG [$O $OE];"
"'-' < $latinInitial e {} $IEUNG [$O $OE $U];"
"'-' < $latinInitial [o a] {} $IEUNG [$E $EO $EU];"
"'-' < $latinInitial [w y] a {} $IEUNG [$E $EO $EU];"
"'-' < [^$latinMedial] [y w] e {} $IEUNG [$O $OE];"
"'-' < [^$latinMedial] e {} $IEUNG [$O $OE $U];"
"'-' < [^$latinMedial] [o a] {} $IEUNG [$E $EO $EU];"
"'-' < [^$latinMedial] [w y] a {} $IEUNG [$E $EO $EU];"
// Single finals followed by IEUNG. The jamo sequence A Xf IEUNG E,
// where Xi also exists, must be transliterated as "ax-e" to prevent