1# Copyright (C) 2016 and later: Unicode, Inc. and others. 2# License & terms of use: http://www.unicode.org/copyright.html 3# 4# 5# Copyright (C) 2002-2015, International Business Machines Corporation and others. 6# All Rights Reserved. 7# 8# file: sent_el.txt 9# 10# ICU Sentence Break Rules 11# See Unicode Standard Annex #29. 12# These rules are based on UAX #29 Revision 26 for Unicode Version 8.0 13# 14 15!!quoted_literals_only; 16 17# 18# Character categories as defined in TR 29 19# 20$CR = [\p{Sentence_Break = CR}]; 21$LF = [\p{Sentence_Break = LF}]; 22$Extend = [\p{Sentence_Break = Extend}]; 23$Sep = [\p{Sentence_Break = Sep}]; 24$Format = [\p{Sentence_Break = Format}]; 25$Sp = [\p{Sentence_Break = Sp}]; 26$Lower = [\p{Sentence_Break = Lower}]; 27$Upper = [\p{Sentence_Break = Upper}]; 28$OLetter = [\p{Sentence_Break = OLetter}]; 29$Numeric = [\p{Sentence_Break = Numeric}]; 30$ATerm = [\p{Sentence_Break = ATerm}]; 31$SContinue = [\p{Sentence_Break = SContinue}]; 32$STerm = [\p{Sentence_Break = STerm} [\u003B \u037E]]; 33$Close = [\p{Sentence_Break = Close}]; 34 35# 36# Define extended forms of the character classes, 37# incorporate trailing Extend or Format chars. 38# Rules 4 and 5. 39 40$SpEx = $Sp ($Extend | $Format)*; 41$LowerEx = $Lower ($Extend | $Format)*; 42$UpperEx = $Upper ($Extend | $Format)*; 43$OLetterEx = $OLetter ($Extend | $Format)*; 44$NumericEx = $Numeric ($Extend | $Format)*; 45$ATermEx = $ATerm ($Extend | $Format)*; 46$SContinueEx= $SContinue ($Extend | $Format)*; 47$STermEx = $STerm ($Extend | $Format)*; 48$CloseEx = $Close ($Extend | $Format)*; 49 50 51## ------------------------------------------------- 52 53!!chain; 54!!forward; 55 56# Rule 3 - break after separators. Keep CR/LF together. 57# 58$CR $LF; 59 60 61# Rule 4 - Break after $Sep. 62# Rule 5 - Ignore $Format and $Extend 63# 64[^$Sep $CR $LF]? ($Extend | $Format)*; 65 66 67# Rule 6 68$ATermEx $NumericEx; 69 70# Rule 7 71($UpperEx | $LowerEx) $ATermEx $UpperEx; 72 73#Rule 8 74$NotLettersEx = [^$OLetter $Upper $Lower $Sep $CR $LF $ATerm $STerm] ($Extend | $Format)*; 75$ATermEx $CloseEx* $SpEx* $NotLettersEx* $Lower; 76 77# Rule 8a 78($STermEx | $ATermEx) $CloseEx* $SpEx* ($SContinueEx | $STermEx | $ATermEx); 79 80#Rule 9, 10, 11 81($STermEx | $ATermEx) $CloseEx* $SpEx* ($Sep | $CR | $LF)?; 82 83#Rule 12 84[[^$STerm $ATerm $Close $Sp $Sep $LF $CR $Format $Extend]{bof}] ($Extend | $Format | $Close | $Sp)* .; 85[[^$STerm $ATerm $Close $Sp $Sep $LF $CR $Format $Extend]{bof}] ($Extend | $Format | $Close | $Sp)* ([$Sep $LF $CR {eof}] | $CR $LF){100}; 86 87## ------------------------------------------------- 88 89!!safe_reverse; 90 91$SpEx_R = ($Extend | $Format)* $Sp; 92$ATermEx_R = ($Extend | $Format)* $ATerm; 93$STermEx_R = ($Extend | $Format)* $STerm; 94$CloseEx_R = ($Extend | $Format)* $Close; 95 96# 97# Reverse rules. 98# For now, use the old style inexact reverse rules, which are easier 99# to write, but less efficient. 100# TODO: exact reverse rules. It appears that exact reverse rules 101# may require improving support for look-ahead breaks in the 102# builder. Needs more investigation. 103# 104 105[{bof}] (.? | $LF $CR) [^$Sep $CR $LF]* [$Sep $CR $LF {eof}] ($SpEx_R* $CloseEx_R* ($STermEx_R | $ATermEx_R))*; 106 107# Explanation for this rule: 108# 109# It needs to back over 110# The $Sep at which we probably begin 111# All of the non $Sep chars leading to the preceding $Sep 112# The preceding $Sep, which will be the second one that the rule matches. 113# Any immediately preceding STerm or ATerm sequences. We need to see these 114# to get the correct rule status when moving forwards again. 115# 116# [{bof}] inhibit rule chaining. Without this, rule would loop on itself and match 117# the entire string. 118# 119# (.? | $LF $CR) Match one $Sep instance. Use .? rather than $Sep because position might be 120# at the beginning of the string at this point, and we don't want to fail. 121# Can only use {eof} once, and it is used later. 122# 123