001: /*
002: * Portions Copyright 1999-2007 Sun Microsystems, Inc. All Rights Reserved.
003: * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
004: *
005: * This code is free software; you can redistribute it and/or modify it
006: * under the terms of the GNU General Public License version 2 only, as
007: * published by the Free Software Foundation. Sun designates this
008: * particular file as subject to the "Classpath" exception as provided
009: * by Sun in the LICENSE file that accompanied this code.
010: *
011: * This code is distributed in the hope that it will be useful, but WITHOUT
012: * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
013: * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License
014: * version 2 for more details (a copy is included in the LICENSE file that
015: * accompanied this code).
016: *
017: * You should have received a copy of the GNU General Public License version
018: * 2 along with this work; if not, write to the Free Software Foundation,
019: * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA.
020: *
021: * Please contact Sun Microsystems, Inc., 4150 Network Circle, Santa Clara,
022: * CA 95054 USA or visit www.sun.com if you need additional information or
023: * have any questions.
024: */
025:
026: /*
027: * @(#)BreakIteratorRules.java 1.34 07/05/05
028: */
029:
030: /*
031: * Licensed Materials - Property of IBM
032: *
033: * (C) Copyright IBM Corp. 1999 All Rights Reserved.
034: * (C) IBM Corp. 1997-1998. All Rights Reserved.
035: *
036: * The program is provided "as is" without any warranty express or
037: * implied, including the warranty of non-infringement and the implied
038: * warranties of merchantibility and fitness for a particular purpose.
039: * IBM will not be liable for any damages suffered by you as a result
040: * of using the Program. In no event will IBM be liable for any
041: * special, indirect or consequential damages or lost profits even if
042: * IBM has been advised of the possibility of their occurrence. IBM
043: * will not be liable for any third party claims against you.
044: */
045:
046: package sun.text.resources;
047:
048: import java.util.ListResourceBundle;
049:
050: /**
051: * Default break-iterator rules. These rules are more or less general for
052: * all locales, although there are probably a few we're missing. The
053: * behavior currently mimics the behavior of BreakIterator in JDK 1.2.
054: * There are known deficiencies in this behavior, including the fact that
055: * the logic for handling CJK characters works for Japanese but not for
056: * Chinese, and that we don't currently have an appropriate locale for
057: * Thai. The resources will eventually be updated to fix these problems.
058: */
059:
060: /* Modified for Hindi 3/1/99. */
061:
062: /*
063: * Since JDK 1.5.0, this file no longer goes to runtime and is used at J2SE
064: * build phase in order to create [Character|Word|Line|Sentence]BreakIteratorData
065: * files which are used on runtime instead.
066: */
067:
068: public class BreakIteratorRules extends ListResourceBundle {
069: protected final Object[][] getContents() {
070: return new Object[][] {
071: // rules describing how to break between logical characters
072: { "CharacterBreakRules",
073:
074: // ignore non-spacing marks and enclosing marks (since we never
075: // put a break before ignore characters, this keeps combining
076: // accents with the base characters they modify)
077: "<enclosing>=[:Mn::Me:];"
078:
079: // other category definitions
080: + "<choseong>=[\u1100-\u115f];"
081: + "<jungseong>=[\u1160-\u11a7];"
082: + "<jongseong>=[\u11a8-\u11ff];"
083: + "<surr-hi>=[\ud800-\udbff];"
084: + "<surr-lo>=[\udc00-\udfff];"
085:
086: // break after every character, except as follows:
087: + ".;"
088:
089: // keep base and combining characters togethers
090: + "<base>=[^<enclosing>^[:Cc::Cf::Zl::Zp:]];"
091: + "<base><enclosing><enclosing>*;"
092:
093: // keep CRLF sequences together
094: + "\r\n;"
095:
096: // keep surrogate pairs together
097: + "<surr-hi><surr-lo>;"
098:
099: // keep Hangul syllables spelled out using conjoining jamo together
100: + "<choseong>*<jungseong>*<jongseong>*;"
101:
102: // various additions for Hindi support
103: + "<nukta>=[\u093c];"
104: + "<danda>=[\u0964\u0965];"
105: + "<virama>=[\u094d];"
106: + "<devVowelSign>=[\u093e-\u094c\u0962\u0963];"
107: + "<devConsonant>=[\u0915-\u0939];"
108: + "<devNuktaConsonant>=[\u0958-\u095f];"
109: + "<devCharEnd>=[\u0902\u0903\u0951-\u0954];"
110: + "<devCAMN>=(<devConsonant>{<nukta>});"
111: + "<devConsonant1>=(<devNuktaConsonant>|<devCAMN>);"
112: + "<zwj>=[\u200d];"
113: + "<devConjunct>=({<devConsonant1><virama>{<zwj>}}<devConsonant1>);"
114: + "<devConjunct>{<devVowelSign>}{<devCharEnd>};"
115: + "<danda><nukta>;" },
116:
117: // default rules for finding word boundaries
118: { "WordBreakRules",
119: // ignore non-spacing marks, enclosing marks, and format characters,
120: // all of which should not influence the algorithm
121: //"<ignore>=[:Mn::Me::Cf:];"
122: "<ignore>=[:Cf:];"
123:
124: + "<enclosing>=[:Mn::Me:];"
125:
126: // Hindi phrase separator, kanji, katakana, hiragana, CJK diacriticals,
127: // other letters, and digits
128: + "<danda>=[\u0964\u0965];"
129: + "<kanji>=[\u3005\u4e00-\u9fa5\uf900-\ufa2d];"
130: + "<kata>=[\u30a1-\u30fa\u30fd\u30fe];"
131: + "<hira>=[\u3041-\u3094\u309d\u309e];"
132: + "<cjk-diacrit>=[\u3099-\u309c\u30fb\u30fc];"
133: + "<letter-base>=[:L::Mc:^[<kanji><kata><hira><cjk-diacrit>]];"
134: + "<let>=(<letter-base><enclosing>*);"
135: + "<digit-base>=[:N:];"
136: + "<dgt>=(<digit-base><enclosing>*);"
137:
138: // punctuation that can occur in the middle of a word: currently
139: // dashes, apostrophes, quotation marks, and periods
140: + "<mid-word>=[:Pd::Pc:\u00ad\u2027\\\"\\\'\\.];"
141:
142: // punctuation that can occur in the middle of a number: currently
143: // apostrophes, qoutation marks, periods, commas, and the Arabic
144: // decimal point
145: + "<mid-num>=[\\\"\\\'\\,\u066b\\.];"
146:
147: // punctuation that can occur at the beginning of a number: currently
148: // the period, the number sign, and all currency symbols except the cents sign
149: + "<pre-num>=[:Sc:\\#\\.^\u00a2];"
150:
151: // punctuation that can occur at the end of a number: currently
152: // the percent, per-thousand, per-ten-thousand, and Arabic percent
153: // signs, the cents sign, and the ampersand
154: + "<post-num>=[\\%\\&\u00a2\u066a\u2030\u2031];"
155:
156: // line separators: currently LF, FF, PS, and LS
157: + "<ls>=[\n\u000c\u2028\u2029];"
158:
159: // whitespace: all space separators and the tab character
160: + "<ws-base>=[:Zs:\t];"
161: + "<ws>=(<ws-base><enclosing>*);"
162:
163: // a word is a sequence of letters that may contain internal
164: // punctuation, as long as it begins and ends with a letter and
165: // never contains two punctuation marks in a row
166: + "<word>=((<let><let>*(<mid-word><let><let>*)*){<danda>});"
167:
168: // a number is a sequence of digits that may contain internal
169: // punctuation, as long as it begins and ends with a digit and
170: // never contains two punctuation marks in a row.
171: + "<number>=(<dgt><dgt>*(<mid-num><dgt><dgt>*)*);"
172:
173: // break after every character, with the following exceptions
174: // (this will cause punctuation marks that aren't considered
175: // part of words or numbers to be treated as words unto themselves)
176: + ".;"
177:
178: // keep together any sequence of contiguous words and numbers
179: // (including just one of either), plus an optional trailing
180: // number-suffix character
181: + "{<word>}(<number><word>)*{<number>{<post-num>}};"
182:
183: // keep together and sequence of contiguous words and numbers
184: // that starts with a number-prefix character and a number,
185: // and may end with a number-suffix character
186: + "<pre-num>(<number><word>)*{<number>{<post-num>}};"
187:
188: // keep together runs of whitespace (optionally with a single trailing
189: // line separator or CRLF sequence)
190: + "<ws>*{\r}{<ls>};"
191:
192: // keep together runs of Katakana and CJK diacritical marks
193: + "[<kata><cjk-diacrit>]*;"
194:
195: // keep together runs of Hiragana and CJK diacritical marks
196: + "[<hira><cjk-diacrit>]*;"
197:
198: // keep together runs of Kanji
199: + "<kanji>*;"
200:
201: // keep together anything else and an enclosing mark
202: + "<base>=[^<enclosing>^[:Cc::Cf::Zl::Zp:]];"
203: + "<base><enclosing><enclosing>*;" },
204:
205: // default rules for determining legal line-breaking positions
206: {
207: "LineBreakRules",
208: // characters that always cause a break: ETX, tab, LF, FF, LS, and PS
209: "<break>=[\u0003\t\n\f\u2028\u2029];"
210:
211: // ignore format characters and control characters EXCEPT for breaking chars
212: + "<ignore>=[:Cf:[:Cc:^[<break>\r]]];"
213:
214: // enclosing marks
215: + "<enclosing>=[:Mn::Me:];"
216:
217: // Hindi phrase separators
218: + "<danda>=[\u0964\u0965];"
219:
220: // characters that always prevent a break: the non-breaking space
221: // and similar characters
222: + "<glue>=[\u00a0\u0f0c\u2007\u2011\u202f\ufeff];"
223:
224: // whitespace: space separators and control characters, except for
225: // CR and the other characters mentioned above
226: + "<space>=[:Zs::Cc:^[<glue><break>\r]];"
227:
228: // dashes: dash punctuation and the discretionary hyphen, except for
229: // non-breaking hyphens
230: + "<dash>=[:Pd:\u00ad^<glue>];"
231:
232: // characters that stick to a word if they precede it: currency symbols
233: // (except the cents sign) and starting punctuation
234: + "<pre-word>=[:Sc::Ps::Pi:^[\u00a2]\\\"\\\'];"
235:
236: // characters that stick to a word if they follow it: ending punctuation,
237: // other punctuation that usually occurs at the end of a sentence,
238: // small Kana characters, some CJK diacritics, etc.
239: + "<post-word>=[\\\":Pe::Pf:\\!\\%\\.\\,\\:\\;\\?\u00a2\u00b0\u066a\u2030-\u2034\u2103"
240: + "\u2105\u2109\u3001\u3002\u3005\u3041\u3043\u3045\u3047\u3049\u3063"
241: + "\u3083\u3085\u3087\u308e\u3099-\u309e\u30a1\u30a3\u30a5\u30a7\u30a9"
242: + "\u30c3\u30e3\u30e5\u30e7\u30ee\u30f5\u30f6\u30fc-\u30fe\uff01\uff05"
243: + "\uff0c\uff0e\uff1a\uff1b\uff1f];"
244:
245: // Kanji: actually includes Kanji,Kana and Hangul syllables,
246: // except for small Kana and CJK diacritics
247: + "<kanji>=[\u4e00-\u9fa5\uac00-\ud7a3\uf900-\ufa2d\ufa30-\ufa6a\u3041-\u3094\u30a1-\u30fa^[<post-word><ignore>]];"
248:
249: // digits
250: + "<digit>=[:Nd::No:];"
251:
252: // punctuation that can occur in the middle of a number: periods and commas
253: + "<mid-num>=[\\.\\,];"
254:
255: // everything not mentioned above
256: + "<char>=[^[<break><space><dash><kanji><glue><ignore><pre-word><post-word><mid-num>\r<danda>]];"
257:
258: // a "number" is a run of prefix characters and dashes, followed by one or
259: // more digits with isolated number-punctuation characters interspersed
260: + "<number>=([<pre-word><dash>]*<digit><digit>*(<mid-num><digit><digit>*)*);"
261:
262: // the basic core of a word can be either a "number" as defined above, a single
263: // "Kanji" character, or a run of any number of not-explicitly-mentioned
264: // characters (this includes Latin letters)
265: + "<word-core>=(<char>*|<kanji>|<number>);"
266:
267: // a word may end with an optional suffix that be either a run of one or
268: // more dashes or a run of word-suffix characters
269: + "<word-suffix>=((<dash><dash>*|<post-word>*));"
270:
271: // a word, thus, is an optional run of word-prefix characters, followed by
272: // a word core and a word suffix (the syntax of <word-core> and <word-suffix>
273: // actually allows either of them to match the empty string, putting a break
274: // between things like ")(" or "aaa(aaa"
275: + "<word>=(<pre-word>*<word-core><word-suffix>);"
276:
277: + "<hack1>=[\\(];"
278: + "<hack2>=[\\)];"
279: + "<hack3>=[\\$\\'];"
280:
281: // finally, the rule that does the work: Keep together any run of words that
282: // are joined by runs of one of more non-spacing mark. Also keep a trailing
283: // line-break character or CRLF combination with the word. (line separators
284: // "win" over nbsp's)
285: + "<word>(((<space>*<glue><glue>*{<space>})|<hack3>)<word>)*<space>*{<enclosing>*}{<hack1><hack2><post-word>*}{<enclosing>*}{\r}{<break>};"
286: + "\r<break>;" },
287:
288: // default rules for finding sentence boundaries
289: {
290: "SentenceBreakRules",
291: // ignore non-spacing marks, enclosing marks, and format characters
292: "<ignore>=[:Mn::Me::Cf:];"
293:
294: // letters
295: + "<letter>=[:L:];"
296:
297: // lowercase letters
298: + "<lc>=[:Ll:];"
299:
300: // uppercase letters
301: + "<uc>=[:Lu:];"
302:
303: // NOT lowercase letters
304: + "<notlc>=[<letter>^<lc>];"
305:
306: // whitespace (line separators are treated as whitespace)
307: + "<space>=[\t\r\f\n\u2028:Zs:];"
308:
309: // punctuation which may occur at the beginning of a sentence: "starting
310: // punctuation" and quotation marks
311: + "<start-punctuation>=[:Ps::Pi:\\\"\\\'];"
312:
313: // punctuation with may occur at the end of a sentence: "ending punctuation"
314: // and quotation marks
315: + "<end>=[:Pe::Pf:\\\"\\\'];"
316:
317: // digits
318: + "<digit>=[:N:];"
319:
320: // characters that unambiguously signal the end of a sentence
321: + "<term>=[\\!\\?\u3002\uff01\uff1f];"
322:
323: // periods, which MAY signal the end of a sentence
324: + "<period>=[\\.\uff0e];"
325:
326: // characters that may occur at the beginning of a sentence: basically anything
327: // not mentioned above (letters and digits are specifically excluded)
328: + "<sent-start>=[^[:L:<space><start-punctuation><end><digit><term><period>\u2029<ignore>]];"
329:
330: // Hindi phrase separator
331: + "<danda>=[\u0964\u0965];"
332:
333: // always break sentences after paragraph separators
334: + ".*?{\u2029};"
335:
336: // always break after a danda, if it's followed by whitespace
337: + ".*?<danda><space>*;"
338:
339: // if you see a period, skip over additional periods and ending punctuation
340: // and if the next character is a paragraph separator, break after the
341: // paragraph separator
342: //+ ".*?<period>[<period><end>]*<space>*\u2029;"
343: //+ ".*?[<period><end>]*<space>*\u2029;"
344:
345: // if you see a period, skip over additional periods and ending punctuation,
346: // followed by optional whitespace, followed by optional starting punctuation,
347: // and if the next character is something that can start a sentence
348: // (basically, a capital letter), then put the sentence break between the
349: // whitespace and the opening punctuation
350: + ".*?<period>[<period><end>]*<space><space>*/<notlc>;"
351: + ".*?<period>[<period><end>]*<space>*/[<start-punctuation><sent-start>][<start-punctuation><sent-start>]*<letter>;"
352:
353: // if you see a sentence-terminating character, skip over any additional
354: // terminators, periods, or ending punctuation, followed by any whitespace,
355: // followed by a SINGLE optional paragraph separator, and put the break there
356: + ".*?<term>[<term><period><end>]*<space>*{\u2029};"
357:
358: // The following rules are here to aid in backwards iteration. The automatically
359: // generated backwards state table will rewind to the beginning of the
360: // paragraph all the time (or all the way to the beginning of the document
361: // if the document doesn't use the Unicode PS character) because the only
362: // unambiguous character pairs are those involving paragraph separators.
363: // These specify a few more unambiguous breaking situations.
364:
365: // if you see a sentence-starting character, followed by starting punctuation
366: // (remember, we're iterating backwards), followed by an optional run of
367: // whitespace, followed by an optional run of ending punctuation, followed
368: // by a period, this is a safe place to turn around
369: + "!<sent-start><start-punctuation>*<space>*<end>*<period>;"
370:
371: // if you see a letter or a digit, followed by an optional run of
372: // starting punctuation, followed by an optional run of whitespace,
373: // followed by an optional run of ending punctuation, followed by
374: // a sentence terminator, this is a safe place to turn around
375: + "![<sent-start><lc><digit>]<start-punctuation>*<space>*<end>*<term>;" } };
376: }
377: }
|