icuSources/test/testdata/break_rules/sentence.txt

   1 #
   2 # Copyright (c) 2016, International Business Machines Corporation and others. All Rights Reserved.
   3 # file: sentence.txt
   4
   5 type = sentence;      # one of grapheme | word | line | sentence
   6 locale = en;
   7
   8 CR        = [\p{Sentence_Break = CR}];
   9 LF        = [\p{Sentence_Break = LF}];
  10 Extend    = [\p{Sentence_Break = Extend}];
  11 Sep       = [\p{Sentence_Break = Sep}];
  12 Format    = [\p{Sentence_Break = Format}];
  13 Sp        = [\p{Sentence_Break = Sp}];
  14 Lower     = [\p{Sentence_Break = Lower}];
  15 Upper     = [\p{Sentence_Break = Upper}];
  16 OLetter   = [\p{Sentence_Break = OLetter}];
  17 Numeric   = [\p{Sentence_Break = Numeric}];
  18 ATerm     = [\p{Sentence_Break = ATerm}];
  19 SContinue = [\p{Sentence_Break = SContinue}];
  20 STerm     = [\p{Sentence_Break = STerm}];
  21 Close     = [\p{Sentence_Break = Close}];
  22
  23 ParaSep   = [Sep CR LF];
  24 SATerm    = [STerm ATerm];
  25 ExtFmt    = [Extend Format];
  26
  27 # SB2:  ÷  eot
  28 #       Conventional regular expression matching for '$' as end-of-text also matches
  29 #       at a line separator just preceding the physical end of text.
  30 #       Instead, use a look-ahead assertion that there is no following character.
  31 SB2:    . ÷ (?!.);
  32
  33 SB3:    CR LF;
  34 SB4:    ParaSep ÷;
  35
  36 # SB5: ignore Format and Extend characters.
  37
  38 SB6:    ATerm ExtFmt* Numeric;
  39 SB7:    (Upper | Lower) ExtFmt* ATerm ExtFmt* Upper;
  40 SB8:    ATerm ExtFmt* (Close ExtFmt*)* (Sp ExtFmt*)* ([^OLetter Upper Lower ParaSep SATerm ExtFmt] ExtFmt *)* Lower;
  41 SB8a:   SATerm ExtFmt* (Close ExtFmt*)* (Sp ExtFmt*)* (SContinue | SATerm);
  42
  43 SB9:    SATerm ExtFmt* (Close ExtFmt*)* (Sp ExtFmt*)* (CR LF | ParaSep)? ÷;
  44         # Also covers SB10, SB11.
  45
  46 SB12:   . ExtFmt* [^ExtFmt]?;
  47