/*
 *******************************************************************************
 * Copyright (C) 1996-2000, International Business Machines Corporation and    *
 * others. All Rights Reserved.                                                *
 *******************************************************************************
 *
 * $Source: /xsrl/Nsvn/icu/icu4j/src/com/ibm/text/resources/Attic/BreakIteratorRules.java,v $ 
 * $Date: 2000/03/10 04:07:26 $ 
 * $Revision: 1.2 $
 *
 *****************************************************************************************
 */
package com.ibm.text.resources;

import java.util.ListResourceBundle;

/**
 * Default break-iterator rules.  These rules are more or less general for
 * all locales, although there are probably a few we're missing.  The
 * behavior currently mimics the behavior of BreakIterator in JDK 1.2.
 * There are known deficiencies in this behavior, including the fact that
 * the logic for handling CJK characters works for Japanese but not for
 * Chinese, and that we don't currently have an appropriate locale for
 * Thai.  The resources will eventually be updated to fix these problems.
 */

 /* Modified for Hindi 3/1/99. */

public class BreakIteratorRules extends ListResourceBundle {
    public Object[][] getContents() {
        return contents;
    }

    static final Object[][] contents = {
        // BreakIteratorClasses lists the class names to instantiate for each
        // built-in type of BreakIterator
        { "BreakIteratorClasses",
            new String[] { "RuleBasedBreakIterator",     // character-break iterator class
                           "RuleBasedBreakIterator",     // word-break iterator class
                           "RuleBasedBreakIterator",     // line-break iterator class
                           "RuleBasedBreakIterator" }    // sentence-break iterator class
        },

        // rules describing how to break between logical characters
        { "CharacterBreakRules",
            // ignore non-spacing marks and enclosing marks (since we never
            // put a break before ignore characters, this keeps combining
            // accents with the base characters they modify)
            "$ignore=[[:Mn:][:Me:]];"

            // other category definitions
            + "choseong=[\u1100-\u115f];"
            + "jungseong=[\u1160-\u11a7];"
            + "jongseong=[\u11a8-\u11ff];"
            + "surr-hi=[\ud800-\udbff];"
            + "surr-lo=[\udc00-\udfff];"

            // break after every character, except as follows:
            + ".;"

            // keep CRLF sequences together
            + "\r\n;"

            // keep surrogate pairs together
            + "{surr-hi}{surr-lo};"

            // keep Hangul syllables spelled out using conjoining jamo together
            + "{choseong}*{jungseong}*{jongseong}*;"

            // various additions for Hindi support
            + "nukta=[\u093c];"
            + "danda=[\u0964\u0965];"
            + "virama=[\u094d];"
            + "devVowelSign=[\u093e-\u094c\u0962\u0963];"
            + "devConsonant=[\u0915-\u0939];"
            + "devNuktaConsonant=[\u0958-\u095f];"
            + "devCharEnd=[\u0902\u0903\u0951-\u0954];"
            + "zwj=[\u200d];"
            
            + "devCAMN=({devConsonant}{nukta}?);"
            + "devConsonant1=({devNuktaConsonant}|{devCAMN});"
            + "devConjunct=(({devConsonant1}{virama}{zwj}?)?{devConsonant1});"

            + "{devConjunct}{devVowelSign}?{devCharEnd}?;"
            + "{danda}{nukta};"
        },

        // default rules for finding word boundaries
        { "WordBreakRules",
            // ignore non-spacing marks, enclosing marks, and format characters,
            // all of which should not influence the algorithm
            "$ignore=[[:Mn:][:Me:][:Cf:]];"

            // Hindi phrase separator, kanji, katakana, hiragana, CJK diacriticals,
            // other letters, and digits
            + "danda=[\u0964\u0965];"
            + "kanji=[\u3005\u4e00-\u9fa5\uf900-\ufa2d];"
            + "kata=[\u3099-\u309c\u30a1-\u30fe];"
            + "hira=[\u3041-\u309e\u30fc];"
            + "let=[[[:L:][:Mc:]]-[{kanji}{kata}{hira}]];"
            + "dgt=[:N:];"

            // punctuation that can occur in the middle of a word: currently
            // dashes, apostrophes, quotation marks, and periods
            + "mid-word=[[:Pd:]\u00ad\u2027\\\"\\\'\\.];"

            // punctuation that can occur in the middle of a number: currently
            // apostrophes, qoutation marks, periods, commas, and the Arabic
            // decimal point
            + "mid-num=[\\\"\\\'\\,\u066b\\.];"

            // punctuation that can occur at the beginning of a number: currently
            // the period, the number sign, and all currency symbols except the cents sign
            + "pre-num=[[[:Sc:]-[\u00a2]]\\#\\.];"

            // punctuation that can occur at the end of a number: currently
            // the percent, per-thousand, per-ten-thousand, and Arabic percent
            // signs, the cents sign, and the ampersand
            + "post-num=[\\%\\&\u00a2\u066a\u2030\u2031];"

            // line separators: currently LF, FF, PS, and LS
            + "ls=[\n\u000c\u2028\u2029];"

            // whitespace: all space separators and the tab character
            + "ws=[[:Zs:]\t];"

            // a word is a sequence of letters that may contain internal
            // punctuation, as long as it begins and ends with a letter and
            // never contains two punctuation marks in a row
            + "word=({let}+({mid-word}{let}+)*{danda}?);"

            // a number is a sequence of digits that may contain internal
            // punctuation, as long as it begins and ends with a digit and
            // never contains two punctuation marks in a row.
            + "number=({dgt}+({mid-num}{dgt}+)*);"

            // break after every character, with the following exceptions
            // (this will cause punctuation marks that aren't considered
            // part of words or numbers to be treated as words unto themselves)
            + ".;"

            // keep together any sequence of contiguous words and numbers
            // (including just one of either), plus an optional trailing
            // number-suffix character
            + "{word}?({number}{word})*({number}{post-num}?)?;"

            // keep together and sequence of contiguous words and numbers
            // that starts with a number-prefix character and a number,
            // and may end with a number-suffix character
            + "{pre-num}({number}{word})*({number}{post-num}?)?;"

            // keep together runs of whitespace (optionally with a single trailing
            // line separator or CRLF sequence)
            + "{ws}*\r?{ls}?;"

            // keep together runs of Katakana
            + "{kata}*;"

            // keep together runs of Hiragana
            + "{hira}*;"

            // keep together runs of Kanji
            + "{kanji}*;"
        },

        // default rules for determining legal line-breaking positions
        { "LineBreakRules",
            // ignore non-spacing marks, enclosing marks, and format characters
            "$ignore=[[:Mn:][:Me:][:Cf:]];"

            // Hindi phrase separators
            + "danda=[\u0964\u0965];"

            // characters that always cause a break: ETX, tab, LF, FF, LS, and PS
            + "break=[\u0003\t\n\f\u2028\u2029];"

            // characters that always prevent a break: the non-breaking space
            // and similar characters
            + "nbsp=[\u00a0\u2007\u2011\ufeff];"

            // whitespace: space separators and control characters, except for
            // CR and the other characters mentioned above
            + "space=[[[:Zs:][:Cc:]]-[{nbsp}{break}\r]];"

            // dashes: dash punctuation and the discretionary hyphen, except for
            // non-breaking hyphens
            + "dash=[[[:Pd:]\u00ad]-[{nbsp}]];"

            // characters that stick to a word if they precede it: currency symbols
            // (except the cents sign) and starting punctuation
            + "pre-word=[[[:Sc:]-[\u00a2]][:Ps:]\\\"\\\'];"

            // characters that stick to a word if they follow it: ending punctuation,
            // other punctuation that usually occurs at the end of a sentence,
            // small Kana characters, some CJK diacritics, etc.
            + "post-word=[[:Pe:]\\!\\\"\\\'\\%\\.\\,\\:\\;\\?\u00a2\u00b0\u066a\u2030-\u2034"
                    + "\u2103\u2105\u2109\u3001\u3002\u3005\u3041\u3043\u3045\u3047\u3049\u3063"
                    + "\u3083\u3085\u3087\u308e\u3099-\u309e\u30a1\u30a3\u30a5\u30a7\u30a9"
                    + "\u30c3\u30e3\u30e5\u30e7\u30ee\u30f5\u30f6\u30fc-\u30fe\uff01\uff0c"
                    + "\uff0e\uff1f];"

            // Kanji: actually includes both Kanji and Kana, except for small Kana and
            // CJK diacritics
            + "kanji=[[\u4e00-\u9fa5\uf900-\ufa2d\u3041-\u3094\u30a1-\u30fa]-[{post-word}{$ignore}]];"

            // digits
            + "digit=[[:Nd:][:No:]];"

            // punctuation that can occur in the middle of a number: periods and commas
            + "mid-num=[\\.\\,];"

            // everything not mentioned above, plus the quote marks (which are both
            // <pre-word>, <post-word>, and <char>)
            + "char=[^{break}{space}{dash}{kanji}{nbsp}{$ignore}{pre-word}{post-word}{mid-num}{danda}\r\\\"\\\'];"

            // a "number" is a run of prefix characters and dashes, followed by one or
            // more digits with isolated number-punctuation characters interspersed
            + "number=([{pre-word}{dash}]*{digit}+({mid-num}{digit}+)*);"

            // the basic core of a word can be either a "number" as defined above, a single
            // "Kanji" character, or a run of any number of not-explicitly-mentioned
            // characters (this includes Latin letters)
            + "word-core=([{pre-word}{char}]*|{kanji}|{number});"

            // a word may end with an optional suffix that be either a run of one or
            // more dashes or a run of word-suffix characters, followed by an optional
            // run of whitespace
            + "word-suffix=(({dash}+|{post-word}*){space}*);"

            // a word, thus, is an optional run of word-prefix characters, followed by
            // a word core and a word suffix (the syntax of <word-core> and <word-suffix>
            // actually allows either of them to match the empty string, putting a break
            // between things like ")(" or "aaa(aaa"
            + "word=({pre-word}*{word-core}{word-suffix});"

            // finally, the rule that does the work: Keep together any run of words that
            // are joined by runs of one of more non-spacing mark.  Also keep a trailing
            // line-break character or CRLF combination with the word.  (line separators
            // "win" over nbsp's)
            + "{word}({nbsp}+{word})*\r?{break}?;"
        },

        // default rules for finding sentence boundaries
        { "SentenceBreakRules",
            // ignore non-spacing marks, enclosing marks, and format characters
            "$ignore=[[:Mn:][:Me:][:Cf:]];"

            // lowercase letters
            + "lc=[:Ll:];"
            
            // uppercase Latin letters
            + "ucLatin=[A-Z];"

            // whitespace (line separators are treated as whitespace)
            + "space=[\t\r\f\n\u2028[:Zs:]];"

            // punctuation which may occur at the beginning of a sentence: "starting
            // punctuation" and quotation marks
            + "start=[[:Ps:]\\\"\\\'];"

            // punctuation with may occur at the end of a sentence: "ending punctuation"
            // and quotation marks
            + "end=[[:Pe:]\\\"\\\'];"

            // digits
            + "digit=[:N:];"

            // characters that unambiguously signal the end of a sentence
            + "term=[\\!\\?\u3002\uff01\uff1f];"

            // periods, which MAY signal the end of a sentence
            + "period=[\\.\uff0e];"

            // characters that may occur at the beginning of a sentence: basically anything
            // not mentioned above (lowercase letters and digits are specifically excluded)
            + "sent-start=[^{lc}{ucLatin}{space}{start}{end}{digit}{term}{period}\u2029{$ignore}];"

            // Hindi phrase separator
            + "danda=[\u0964\u0965];"

            // always break sentences after paragraph separators
            + ".*?\u2029?;"

            // always break after a danda, if it's followed by whitespace
            + ".*?{danda}{space}*;"

            // if you see a period, skip over additional periods and ending punctuation
            // and if the next character is a paragraph separator, break after the
            // paragraph separator
            + ".*?{period}[{period}{end}]*{space}*\u2029;"

            // if you see a period, skip over additional periods and ending punctuation,
            // followed by optional whitespace, followed by optional starting punctuation,
            // and if the next character is something that can start a sentence
            // (basically, a capital letter), then put the sentence break between the
            // whitespace and the opening punctuation
            + ".*?{period}[{period}{end}]*{space}*/({start}*{sent-start}|{start}+{ucLatin});"
            
            // same as above, except that there's a sentence break before a Latin capital
            // letter only if there's at least one space after the period
            + ".*?{period}[{period}{end}]*{space}+/{ucLatin};"

            // if you see a sentence-terminating character, skip over any additional
            // terminators, periods, or ending punctuation, followed by any whitespace,
            // followed by a SINGLE optional paragraph separator, and put the break there
            + ".*?{term}[{term}{period}{end}]*{space}*\u2029?;"

            // The following rules are here to aid in backwards iteration.  The automatically
            // generated backwards state table will rewind to the beginning of the
            // paragraph all the time (or all the way to the beginning of the document
            // if the document doesn't use the Unicode PS character) because the only
            // unambiguous character pairs are those involving paragraph separators.
            // These specify a few more unambiguous breaking situations.

            // if you see a sentence-starting character, followed by starting punctuation
            // (remember, we're iterating backwards), followed by an optional run of
            // whitespace, followed by an optional run of ending punctuation, followed
            // by a period, this is a safe place to turn around
            + "![{sent-start}{ucLatin}]{start}*{space}+{end}*{period};"

            // if you see a letter or a digit, followed by an optional run of
            // starting punctuation, followed by an optional run of whitespace,
            // followed by an optional run of ending punctuation, followed by
            // a sentence terminator, this is a safe place to turn around
            + "![{sent-start}{lc}{digit}]{start}*{space}*{end}*{term};"
        }
    };
}