ICU-6.2.4.tar.gz

[apple/icu.git] / icuSources / data / brkitr / sent.txt

diff --git a/icuSources/data/brkitr/sent.txt b/icuSources/data/brkitr/sent.txt
index 14b568baa4aa9c7e3700e5afd7fc9af592302295..c13b22d71ece76485c089f0003ff185ff5090060 100644 (file)
--- a/icuSources/data/brkitr/sent.txt
+++ b/icuSources/data/brkitr/sent.txt
@@ -1,41 +1,39 @@
 #
-#   Copyright (C) 2002-2003, International Business Machines Corporation and others.
+#   Copyright (C) 2002-2004, International Business Machines Corporation and others.
 #       All Rights Reserved.
 #
-#   file:  sent.txt   
+#   file:  sent.txt
 #
 #   ICU Sentence Break Rules
 #      See Unicode Standard Annex #29.
 #      These rules are based on TR 29 version 4.0.0
 #
-    
+
 
 #
 # Character categories as defined in TR 29
 #
 $Sep     = [\u000a \u000d \u0085 \u2028 \u2029];
-$Format  = [[:Format:]];
+$Format  = [[:Format:] - [:Grapheme_Extend:]];
 $Sp      = [[:Whitespace:] - $Sep];
 $Lower   = [[:Lowercase:]];
 $Upper   = [[:TitleCase_Letter:] [:Uppercase:]];
 $OLetter = [[:Alphabetic:] [:name = HEBREW PUNCTUATION GERESH:] - [$Lower $Upper]];
-$Numeric = [:LineBreak = Numeric:];
+$Numeric = [[:Nd:][:name = ARABIC DECIMAL SEPARATOR:][:name = ARABIC THOUSANDS SEPARATOR:]];
+
+$ATerm = [.];
 
-$ATerm = [.];  
+$Term  = [:STerm:];
 
-$Term  = [\u0021 \u003F \u0589 \u061F \u06D4 \u0700 \u0701 \u0702 \u0964 \u1362
-          \u1367 \u1368 \u104a \u104b \u166e \u1803 \u1809 \u203C \u203D \u2047 
-          \u2048 \u2049 \u3002 \uFE52 \uFE57 \uFF01 \uFF0E \uFF1F \uFF61];
-          
 $Close   = [[:Open_Punctuation:] [:Close_Punctuation:] [:Linebreak = Quotation:] -
            [[:name = HEBREW PUNCTUATION GERESH:] $ATerm $Term]];
-           
-           
+
+
 
 # Define extended forms of the character classes,
 #   incorporate grapheme cluster + format chars.
 
-$Extend     = [[:Grapheme_Extend = TRUE:]]; 
+$Extend     = [[:Grapheme_Extend = TRUE:]];
 $ATermEx    = $ATerm   $Extend* $Format*;
 $NumericEx  = $Numeric $Extend* $Format*;
 $UpperEx    = $Upper   $Extend* $Format*;
@@ -49,6 +47,9 @@ $SepSeq  = $Sep | \u000d\u000a;
 # $InteriorChars are those that never trigger a following break.
 $InteriorChars = [^$Term $ATerm $Sep];   #Note:  includes Extend and Format chars
 
+## -------------------------------------------------
+
+!!forward;
 
 # Rule 6.  Match an ATerm (.) that does not cause a break because a number immediately follows it.
 $NumberFollows = $InteriorChars* $ATermEx $NumericEx;
@@ -64,24 +65,86 @@ $LowerWordFollows  = $InteriorChars* $ATermEx $Close* $Sp* [^$OLetter $Upper $Lo
 # Rules 3, 9, 10, 11
 #                       Matches a simple sentence, or the trailing part of a complex sentence,
 #                       where a simple sentence contains no interior "."s.
-$EndSequence       = $InteriorChars* ($TermEx | $ATermEx) $Close* $Sp* $SepSeq? |
-                     $InteriorChars* $SepSeq?;
+$TermEndSequence   = $InteriorChars* ($TermEx | $ATermEx) $Close* $Sp* $SepSeq?;
+$EndSequence       = $InteriorChars* $SepSeq?;
 
+# Put them all together.
+($NumberFollows | $UppersSurround |  $LowerWordFollows)*  $TermEndSequence{0};   # status = UBRK_SENTENCE_TERM
+($NumberFollows | $UppersSurround |  $LowerWordFollows)*  $EndSequence{100};     # status = UBRK_SENTENCE_SEP
 
+## -------------------------------------------------
 
-# Put them all together.  
-($NumberFollows | $UppersSurround |  $LowerWordFollows)*  $EndSequence;
+!!reverse;
 
-     
-#
-#  Reverse Rules
-#
-$EndGorp                  = ($Term | $ATerm | $Sep | $Close | $Extend | $Format | $Sp);
-$RevEndSequence           = $EndGorp* $InteriorChars* $EndGorp* | $Sep [^$ATerm $Term]*;
-$ReverseLowerWordFollows  = $Lower [^$OLetter $Upper $Lower $Sep]* $ATerm $InteriorChars*;
-$ReverseUpperSurround     = $Upper $Format* $Extend* $ATerm $Format* $Extend* $Upper $InteriorChars*;
-$ReverseNumberFollows     = $Numeric $Format* $Extend* $ATerm $InteriorChars*;
-
-! $RevEndSequence ($ReverseLowerWordFollows | $ReverseUpperSurround | $ReverseNumberFollows)* .?;
-#! .*;
- 
+# rule 6
+
+$RULE6 = $Numeric $Format* $Extend* $ATerm;
+
+# rule 7
+
+$RULE7 = $Upper $Format* $Extend* $ATerm $Format* $Extend* $Upper;
+
+# rule 8
+
+$RULE8 = $Lower ($Format* $Extend* [^$OLetter $Upper $Lower $Sep])* 
+             ($Format* $Extend* $Sp)* ($Format* $Extend* $Close)*
+             $Format* $Extend* $ATerm;
+
+# rule 9, 10, 11
+
+# $CR $LF
+$End = $Sep | \u000a\u000d
+       | $Format* $Extend* $Sp* $Format* $Extend* $Close* $Format* 
+                $Extend* ($Term | $ATerm)
+          | $Sep $Format* $Extend* $Sp* $Format* $Extend* $Close* $Format* 
+                $Extend* ($Term | $ATerm);
+       
+# rule 12
+
+$RULE12 = [^$Sep $Term $ATerm];
+
+$Join = ($RULE6 | $RULE7 | $RULE8 | $RULE12)*;
+
+$End;
+
+$End? $Join [$RULE12 - $Sp - $Close];
+
+# forces a break at the beginning of text "$Sp blah blah blah"
+# remember the break iterators takes the longest match
+$End? $Join $Sp / [^$Term $ATerm $Sp $Close];
+
+# forces a break at the beginning of text "$Close blah blah blah"
+$End? $Join $Close / [^$Term $ATerm $Close];
+
+## -------------------------------------------------
+
+!!safe_reverse;
+
+# rule 4
+$Extend+ [^$Extend];
+
+# rule 7
+$Extend* $ATerm $Format* $Extend* $Upper;
+
+# rule 8
+($Extend* $Term)+ ($Extend* $Sp $Format*)* ($Extend* $Close $Format*)* $Extend* $ATerm;
+
+# rule 11
+($Extend* $Sp $Format*)* ($Extend* $Close $Format*)*;
+($Extend* $Sp $Format*)* ($Extend* $Close $Format*)* $Extend* ($Term | $ATerm);
+
+## -------------------------------------------------
+
+!!safe_forward;
+
+# rule 7
+
+$ATerm $Extend* $Format* $Upper;
+
+# rule 8
+
+$Lower .;
+
+# rule 11
+
+($Close $Extend* $Format*)* ($Sp $Extend* $Format*)*;