X-Git-Url: https://git.saurik.com/apple/icu.git/blobdiff_plain/48b980fed3435926e0b3a8d72ecb58be703a1c7a..729e4ab9bc6618bc3d8a898e575df7f4019e29ca:/icuSources/data/mappings/ibm-964_P110-1999.ucm?ds=sidebyside diff --git a/icuSources/data/mappings/ibm-964_P110-1999.ucm b/icuSources/data/mappings/ibm-964_P110-1999.ucm index 16476fd0..93530b82 100644 --- a/icuSources/data/mappings/ibm-964_P110-1999.ucm +++ b/icuSources/data/mappings/ibm-964_P110-1999.ucm @@ -1,6 +1,6 @@ # *************************************************************************** # * -# * Copyright (C) 1995-2007, International Business Machines +# * Copyright (C) 1995-2009, International Business Machines # * Corporation and others. All Rights Reserved. # * # *************************************************************************** @@ -18,15 +18,30 @@ "ASCII" "ibm-964_VPUA" - 0-8d, 8e:2, 90-9f, a1-fe:1, aa-c1:5, c3:5, fe:5 +# The fourth line is commented out (and does not count) +# because the state table is hand-optimized and does not use what would be +# the natural path for the encoding scheme. +# The third used to start with "a1-b0:3" but overrode every one +# of these byte values with a different state transition. + +# 0: Initial state, single bytes and lead bytes + 0-8d, 8e:2, 90-9f, a1-fe:1, aa-c1:4, c3:4, fe:4 +# 1: Trail byte state with mappings a1-fe - a1-b0:3, a1:4, a2:8, a3-ab:4, ac:7, ad:6, ae-b0:4 - a1-fe:1 - a1-fe:5 +# 2: Second of four bytes, follows lead byte 8e + a1:3, a2:7, a3-ab:3, ac:6, ad:5, ae-b0:3 +# (unreachable/optimized away) +# a1-fe:1 +# 3: Third of four bytes, 8e xx .. .. for most xx in a1-b0; all-unassigned + a1-fe:4 +# 4: All-unassigned trail byte state a1-fe.u - a1-a4:1, a5-fe:5 - a1-e2:1, e3-fe:5 - a1-f2:1, f3-fe:5 +# 5: 8e ad .. .. with some mappings + a1-a4:1, a5-fe:4 +# 6: 8e ac .. .. with some mappings + a1-e2:1, e3-fe:4 +# 7: 8e a2 .. .. with some mappings + a1-f2:1, f3-fe:4 CHARMAP \x00 |0