icuSources/test/testdata/testnorm.txt

   1 # Copyright (C) 2016 and later: Unicode, Inc. and others.
   2 # License & terms of use: http://www.unicode.org/copyright.html
   3 #   Copyright (C) 2010, International Business Machines
   4 #   Corporation and others.  All Rights Reserved.
   5 #
   6 #   file name:  testnorm.txt
   7 #   encoding:   US-ASCII
   8 #   tab size:   8 (not used)
   9 #   indentation:4
  10 #
  11 #   created on: 2010feb15
  12 #   created by: Markus W. Scherer
  13 #
  14 # Normalization test data, for improving code coverage.
  15
  16 # Selection of Canonical_Combining_Class (ccc) values
  17 0300..0314:230
  18 0315:232
  19 0316..0319:220
  20 031A:232
  21 031B:216
  22 031C..0320:220
  23 0321..0322:202
  24 0323..0326:220
  25 0327..0328:202
  26 0329..0333:220
  27 0334..0338:1
  28 0339..033C:220
  29 033D..0344:230
  30 0345:240
  31 0346:230
  32 0347..0349:220
  33 034A..034C:230
  34 034D..034E:220
  35 0350..0352:230
  36 0353..0356:220
  37 0357:230
  38 0358:232
  39 0359..035A:220
  40 035B:230
  41 035C:233
  42 035D..035E:234
  43 035F:233
  44 0360..0361:234
  45 0362:233
  46 0363..036F:230
  47 # ICU 63 normalization with UCPTrie requires inert surrogate code points.
  48 # D802:2  # surrogates with non-zero combining classes
  49 # D803:3
  50 # D804:4
  51 110B9:9
  52 110BA:7
  53
  54 # Some interesting mappings
  55 00C0=0041 0300
  56 00C1=0041 0301
  57 00C2=0041 0302
  58 00C3=0041 0303
  59 00C4=0041 0308
  60 00C5=0041 030A
  61 00C7=0043 0327
  62 # ICU 63 normalization with UCPTrie requires inert surrogate code points.
  63 # D800>D7FF  # surrogates with mappings, and mappings to empty strings
  64 # D801>
  65 # DFFE>
  66 # DFFF>FFFF
  67 E000>
  68 E001=61 338  # composition with trail<=33FF and composite>7FFF
  69 E002=E001 308  # recursive mapping needs reordering
  70 E003>62 307 327 337  # mapping needs reordering
  71 E011=E010 F0011  # composition of BMP+supplementary, and F0011 is maybe & combines-fwd
  72 E111>1101  # mapping ends in Jamo L
  73 E112>1102 62  # mapping starts with Jamo L
  74 FFF3>FFF4
  75 FFF4>FFF5
  76 FFF5>FFF7
  77 FFF7>10037
  78 10036>FFF6
  79 10077>10037
  80 1109A=11099 110BA
  81 1109C=1109B 110BA
  82 110AB=110A5 110BA
  83 F0010=F0011 E012  # composition of supplementary+BMP