Index: lucene/src/test/org/apache/lucene/analysis/TestUAX29Tokenizer.java =================================================================== --- lucene/src/test/org/apache/lucene/analysis/TestUAX29Tokenizer.java (revision 0) +++ lucene/src/test/org/apache/lucene/analysis/TestUAX29Tokenizer.java (revision 0) @@ -0,0 +1,186 @@ +package org.apache.lucene.analysis; + +import java.io.IOException; +import java.io.Reader; +import java.io.StringReader; +import java.util.Arrays; + +import org.apache.lucene.analysis.standard.UAX29Tokenizer; + +/** + * Licensed to the Apache Software Foundation (ASF) under one or more + * contributor license agreements. See the NOTICE file distributed with + * this work for additional information regarding copyright ownership. + * The ASF licenses this file to You under the Apache License, Version 2.0 + * (the "License"); you may not use this file except in compliance with + * the License. You may obtain a copy of the License at + * + * http://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, software + * distributed under the License is distributed on an "AS IS" BASIS, + * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. + * See the License for the specific language governing permissions and + * limitations under the License. + */ + +public class TestUAX29Tokenizer extends BaseTokenStreamTestCase { + + public void testHugeDoc() throws IOException { + StringBuilder sb = new StringBuilder(); + char whitespace[] = new char[4094]; + Arrays.fill(whitespace, ' '); + sb.append(whitespace); + sb.append("testing 1234"); + String input = sb.toString(); + UAX29Tokenizer tokenizer = new UAX29Tokenizer(new StringReader(input)); + assertTokenStreamContents(tokenizer, new String[] { "testing", "1234" }); + } + + private Analyzer a = new ReusableAnalyzerBase() { + @Override + protected TokenStreamComponents createComponents + (String fieldName, Reader reader) { + + Tokenizer tokenizer = new UAX29Tokenizer(reader); + return new TokenStreamComponents(tokenizer); + } + }; + + public void testArmenian() throws Exception { + assertAnalyzesTo(a, "Վիքիպեդիայի 13 միլիոն հոդվածները (4,600` հայերեն վիքիպեդիայում) գրվել են կամավորների կողմից ու համարյա բոլոր հոդվածները կարող է խմբագրել ցանկաց մարդ ով կարող է բացել Վիքիպեդիայի կայքը։", + new String[] { "Վիքիպեդիայի", "13", "միլիոն", "հոդվածները", "4,600", "հայերեն", "վիքիպեդիայում", "գրվել", "են", "կամավորների", "կողմից", + "ու", "համարյա", "բոլոր", "հոդվածները", "կարող", "է", "խմբագրել", "ցանկաց", "մարդ", "ով", "կարող", "է", "բացել", "Վիքիպեդիայի", "կայքը" } ); + } + + public void testAmharic() throws Exception { + assertAnalyzesTo(a, "ዊኪፔድያ የባለ ብዙ ቋንቋ የተሟላ ትክክለኛና ነጻ መዝገበ ዕውቀት (ኢንሳይክሎፒዲያ) ነው። ማንኛውም", + new String[] { "ዊኪፔድያ", "የባለ", "ብዙ", "ቋንቋ", "የተሟላ", "ትክክለኛና", "ነጻ", "መዝገበ", "ዕውቀት", "ኢንሳይክሎፒዲያ", "ነው", "ማንኛውም" } ); + } + + public void testArabic() throws Exception { + assertAnalyzesTo(a, "الفيلم الوثائقي الأول عن ويكيبيديا يسمى \"الحقيقة بالأرقام: قصة ويكيبيديا\" (بالإنجليزية: Truth in Numbers: The Wikipedia Story)، سيتم إطلاقه في 2008.", + new String[] { "الفيلم", "الوثائقي", "الأول", "عن", "ويكيبيديا", "يسمى", "الحقيقة", "بالأرقام", "قصة", "ويكيبيديا", + "بالإنجليزية", "Truth", "in", "Numbers", "The", "Wikipedia", "Story", "سيتم", "إطلاقه", "في", "2008" } ); + } + + public void testAramaic() throws Exception { + assertAnalyzesTo(a, "ܘܝܩܝܦܕܝܐ (ܐܢܓܠܝܐ: Wikipedia) ܗܘ ܐܝܢܣܩܠܘܦܕܝܐ ܚܐܪܬܐ ܕܐܢܛܪܢܛ ܒܠܫܢ̈ܐ ܣܓܝܐ̈ܐ܂ ܫܡܗ ܐܬܐ ܡܢ ܡ̈ܠܬܐ ܕ\"ܘܝܩܝ\" ܘ\"ܐܝܢܣܩܠܘܦܕܝܐ\"܀", + new String[] { "ܘܝܩܝܦܕܝܐ", "ܐܢܓܠܝܐ", "Wikipedia", "ܗܘ", "ܐܝܢܣܩܠܘܦܕܝܐ", "ܚܐܪܬܐ", "ܕܐܢܛܪܢܛ", "ܒܠܫܢ̈ܐ", "ܣܓܝܐ̈ܐ", "ܫܡܗ", + "ܐܬܐ", "ܡܢ", "ܡ̈ܠܬܐ", "ܕ", "ܘܝܩܝ", "ܘ", "ܐܝܢܣܩܠܘܦܕܝܐ"}); + } + + public void testBengali() throws Exception { + assertAnalyzesTo(a, "এই বিশ্বকোষ পরিচালনা করে উইকিমিডিয়া ফাউন্ডেশন (একটি অলাভজনক সংস্থা)। উইকিপিডিয়ার শুরু ১৫ জানুয়ারি, ২০০১ সালে। এখন পর্যন্ত ২০০টিরও বেশী ভাষায় উইকিপিডিয়া রয়েছে।", + new String[] { "এই", "বিশ্বকোষ", "পরিচালনা", "করে", "উইকিমিডিয়া", "ফাউন্ডেশন", "একটি", "অলাভজনক", "সংস্থা", "উইকিপিডিয়ার", + "শুরু", "১৫", "জানুয়ারি", "২০০১", "সালে", "এখন", "পর্যন্ত", "২০০টিরও", "বেশী", "ভাষায়", "উইকিপিডিয়া", "রয়েছে" }); + } + + public void testFarsi() throws Exception { + assertAnalyzesTo(a, "ویکی پدیای انگلیسی در تاریخ ۲۵ دی ۱۳۷۹ به صورت مکملی برای دانشنامهٔ تخصصی نوپدیا نوشته شد.", + new String[] { "ویکی", "پدیای", "انگلیسی", "در", "تاریخ", "۲۵", "دی", "۱۳۷۹", "به", "صورت", "مکملی", + "برای", "دانشنامهٔ", "تخصصی", "نوپدیا", "نوشته", "شد" }); + } + + public void testGreek() throws Exception { + assertAnalyzesTo(a, "Γράφεται σε συνεργασία από εθελοντές με το λογισμικό wiki, κάτι που σημαίνει ότι άρθρα μπορεί να προστεθούν ή να αλλάξουν από τον καθένα.", + new String[] { "Γράφεται", "σε", "συνεργασία", "από", "εθελοντές", "με", "το", "λογισμικό", "wiki", "κάτι", "που", + "σημαίνει", "ότι", "άρθρα", "μπορεί", "να", "προστεθούν", "ή", "να", "αλλάξουν", "από", "τον", "καθένα" }); + } + + public void testTibetan() throws Exception { + assertAnalyzesTo(a, "སྣོན་མཛོད་དང་ལས་འདིས་བོད་ཡིག་མི་ཉམས་གོང་འཕེལ་དུ་གཏོང་བར་ཧ་ཅང་དགེ་མཚན་མཆིས་སོ། །", + new String[] { "སྣོན", "མཛོད", "དང", "ལས", "འདིས", "བོད", "ཡིག", + "མི", "ཉམས", "གོང", "འཕེལ", "དུ", "གཏོང", "བར", + "ཧ", "ཅང", "དགེ", "མཚན", "མཆིས", "སོ" }); + } + + /* + * For chinese, tokenize as char (these can later form bigrams or whatever) + * TODO: why do full-width numerics have no word-break prop? + */ + public void testChinese() throws Exception { + assertAnalyzesTo(a, "我是中国人。 １２３４ Ｔｅｓｔｓ ", + new String[] { "我", "是", "中", "国", "人", "Ｔｅｓｔｓ"}); + } + + public void testEmpty() throws Exception { + assertAnalyzesTo(a, "", new String[] {}); + assertAnalyzesTo(a, ".", new String[] {}); + assertAnalyzesTo(a, " ", new String[] {}); + } + + /* test various jira issues this analyzer is related to */ + + public void testLUCENE1545() throws Exception { + /* + * Standard analyzer does not correctly tokenize combining character U+0364 COMBINING LATIN SMALL LETTRE E. + * The word "moͤchte" is incorrectly tokenized into "mo" "chte", the combining character is lost. + * Expected result is only on token "moͤchte". + */ + assertAnalyzesTo(a, "moͤchte", new String[] { "moͤchte" }); + } + + /* Tests from StandardAnalyzer, just to show behavior is similar */ + public void testAlphanumericSA() throws Exception { + // alphanumeric tokens + assertAnalyzesTo(a, "B2B", new String[]{"B2B"}); + assertAnalyzesTo(a, "2B", new String[]{"2B"}); + } + + public void testDelimitersSA() throws Exception { + // other delimiters: "-", "/", "," + assertAnalyzesTo(a, "some-dashed-phrase", new String[]{"some", "dashed", "phrase"}); + assertAnalyzesTo(a, "dogs,chase,cats", new String[]{"dogs", "chase", "cats"}); + assertAnalyzesTo(a, "ac/dc", new String[]{"ac", "dc"}); + } + + public void testApostrophesSA() throws Exception { + // internal apostrophes: O'Reilly, you're, O'Reilly's + assertAnalyzesTo(a, "O'Reilly", new String[]{"O'Reilly"}); + assertAnalyzesTo(a, "you're", new String[]{"you're"}); + assertAnalyzesTo(a, "she's", new String[]{"she's"}); + assertAnalyzesTo(a, "Jim's", new String[]{"Jim's"}); + assertAnalyzesTo(a, "don't", new String[]{"don't"}); + assertAnalyzesTo(a, "O'Reilly's", new String[]{"O'Reilly's"}); + } + + public void testNumericSA() throws Exception { + // floating point, serial, model numbers, ip addresses, etc. + // every other segment must have at least one digit + assertAnalyzesTo(a, "21.35", new String[]{"21.35"}); + assertAnalyzesTo(a, "R2D2 C3PO", new String[]{"R2D2", "C3PO"}); + assertAnalyzesTo(a, "216.239.63.104", new String[]{"216.239.63.104"}); + assertAnalyzesTo(a, "216.239.63.104", new String[]{"216.239.63.104"}); + } + + public void testTextWithNumbersSA() throws Exception { + // numbers + assertAnalyzesTo(a, "David has 5000 bones", new String[]{"David", "has", "5000", "bones"}); + } + + public void testVariousTextSA() throws Exception { + // various + assertAnalyzesTo(a, "C embedded developers wanted", new String[]{"C", "embedded", "developers", "wanted"}); + assertAnalyzesTo(a, "foo bar FOO BAR", new String[]{"foo", "bar", "FOO", "BAR"}); + assertAnalyzesTo(a, "foo bar . FOO <> BAR", new String[]{"foo", "bar", "FOO", "BAR"}); + assertAnalyzesTo(a, "\"QUOTED\" word", new String[]{"QUOTED", "word"}); + } + + public void testKoreanSA() throws Exception { + // Korean words + assertAnalyzesTo(a, "안녕하세요 한글입니다", new String[]{"안녕하세요", "한글입니다"}); + } + + public void testOffsets() throws Exception { + assertAnalyzesTo(a, "David has 5000 bones", + new String[] {"David", "has", "5000", "bones"}, + new int[] {0, 6, 10, 15}, + new int[] {5, 9, 14, 20}); + } + + public void testTypes() throws Exception { + assertAnalyzesTo(a, "David has 5000 bones", + new String[] {"David", "has", "5000", "bones"}, + new String[] { "", "", "", "" }); + }} Index: lucene/src/java/org/apache/lucene/analysis/standard/UAX29Tokenizer.java =================================================================== --- lucene/src/java/org/apache/lucene/analysis/standard/UAX29Tokenizer.java (revision 0) +++ lucene/src/java/org/apache/lucene/analysis/standard/UAX29Tokenizer.java (revision 0) @@ -0,0 +1,931 @@ +/* The following code was generated by JFlex 1.5.0-SNAPSHOT on 5/8/10 10:43 AM */ + +package org.apache.lucene.analysis.standard; + +/** + * Licensed to the Apache Software Foundation (ASF) under one or more + * contributor license agreements. See the NOTICE file distributed with + * this work for additional information regarding copyright ownership. + * The ASF licenses this file to You under the Apache License, Version 2.0 + * (the "License"); you may not use this file except in compliance with + * the License. You may obtain a copy of the License at + * + * http://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, software + * distributed under the License is distributed on an "AS IS" BASIS, + * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. + * See the License for the specific language governing permissions and + * limitations under the License. + */ + +import java.io.IOException; +import java.io.Reader; +import java.nio.CharBuffer; + +import org.apache.lucene.analysis.Tokenizer; +import org.apache.lucene.analysis.tokenattributes.OffsetAttribute; +import org.apache.lucene.analysis.tokenattributes.PositionIncrementAttribute; +import org.apache.lucene.analysis.tokenattributes.CharTermAttribute; +import org.apache.lucene.analysis.tokenattributes.TypeAttribute; +import org.apache.lucene.util.AttributeSource; + + +/** + * This class implements the Unicode Text Segmentation algorithm, as specified + * in Unicode Standard Annex #29 . + */ + +public class UAX29Tokenizer extends Tokenizer { + + /** This character denotes the end of file */ + public static final int YYEOF = -1; + + /** initial size of the lookahead buffer */ + private static final int ZZ_BUFFERSIZE = 16384; + + /** lexical states */ + public static final int YYINITIAL = 0; + + /** + * ZZ_LEXSTATE[l] is the state in the DFA for the lexical state l + * ZZ_LEXSTATE[l+1] is the state in the DFA for the lexical state l + * at the beginning of a line + * l is of the form l = 2*k, k a non negative integer + */ + private static final int ZZ_LEXSTATE[] = { + 0, 0 + }; + + /** + * Translates characters to character classes + */ + private static final String ZZ_CMAP_PACKED = + "\11\0\1\4\1\2\2\3\1\1\22\0\1\4\6\0\1\13\4\0"+ + "\1\12\1\0\1\13\1\0\12\5\1\11\1\12\5\0\32\10\4\0"+ + "\1\15\1\0\32\10\12\0\1\3\32\0\1\4\11\0\1\10\2\0"+ + "\1\7\7\0\1\10\1\0\1\11\2\0\1\10\5\0\27\10\1\0"+ + "\37\10\1\0\u01ca\10\4\0\14\10\16\0\5\10\7\0\1\10\1\0"+ + "\1\10\21\0\160\7\5\10\1\0\2\10\2\0\4\10\1\12\7\0"+ + "\1\10\1\11\3\10\1\0\1\10\1\0\24\10\1\0\123\10\1\0"+ + "\213\10\1\0\7\7\234\10\13\0\46\10\2\0\1\10\7\0\47\10"+ + "\1\0\1\12\7\0\55\7\1\0\1\7\1\0\2\7\1\0\2\7"+ + "\1\0\1\7\10\0\33\10\5\0\4\10\1\11\13\0\4\7\10\0"+ + "\2\12\2\0\13\7\6\0\52\10\24\7\1\0\12\5\1\0\1\5"+ + "\1\12\1\0\2\10\1\7\143\10\1\0\1\10\17\7\2\10\2\7"+ + "\1\0\4\7\2\10\12\5\3\10\2\0\1\10\17\0\1\7\1\10"+ + "\1\7\36\10\33\7\2\0\131\10\13\7\1\10\16\0\12\5\41\10"+ + "\11\7\2\10\2\0\1\12\1\0\1\10\5\0\26\10\4\7\1\10"+ + "\11\7\1\10\3\7\1\10\5\7\322\0\4\7\66\10\2\0\1\7"+ + "\1\10\21\7\1\0\1\10\5\7\2\0\12\10\2\7\2\0\12\5"+ + "\1\0\2\10\6\0\7\10\1\0\3\7\1\0\10\10\2\0\2\10"+ + "\2\0\26\10\1\0\7\10\1\0\1\10\3\0\4\10\2\0\1\7"+ + "\1\10\7\7\2\0\2\7\2\0\3\7\1\10\10\0\1\7\4\0"+ + "\2\10\1\0\3\10\2\7\2\0\12\5\2\10\17\0\3\7\1\0"+ + "\6\10\4\0\2\10\2\0\26\10\1\0\7\10\1\0\2\10\1\0"+ + "\2\10\1\0\2\10\2\0\1\7\1\0\5\7\4\0\2\7\2\0"+ + "\3\7\3\0\1\7\7\0\4\10\1\0\1\10\7\0\12\5\2\7"+ + "\3\10\1\7\13\0\3\7\1\0\11\10\1\0\3\10\1\0\26\10"+ + "\1\0\7\10\1\0\2\10\1\0\5\10\2\0\1\7\1\10\10\7"+ + "\1\0\3\7\1\0\3\7\2\0\1\10\17\0\2\10\2\7\2\0"+ + "\12\5\21\0\3\7\1\0\10\10\2\0\2\10\2\0\26\10\1\0"+ + "\7\10\1\0\2\10\1\0\5\10\2\0\1\7\1\10\7\7\2\0"+ + "\2\7\2\0\3\7\10\0\2\7\4\0\2\10\1\0\3\10\2\7"+ + "\2\0\12\5\1\0\1\10\20\0\1\7\1\10\1\0\6\10\3\0"+ + "\3\10\1\0\4\10\3\0\2\10\1\0\1\10\1\0\2\10\3\0"+ + "\2\10\3\0\3\10\3\0\14\10\4\0\5\7\3\0\3\7\1\0"+ + "\4\7\2\0\1\10\6\0\1\7\16\0\12\5\21\0\3\7\1\0"+ + "\10\10\1\0\3\10\1\0\27\10\1\0\12\10\1\0\5\10\3\0"+ + "\1\10\7\7\1\0\3\7\1\0\4\7\7\0\2\7\1\0\2\10"+ + "\6\0\2\10\2\7\2\0\12\5\22\0\2\7\1\0\10\10\1\0"+ + "\3\10\1\0\27\10\1\0\12\10\1\0\5\10\2\0\1\7\1\10"+ + "\7\7\1\0\3\7\1\0\4\7\7\0\2\7\7\0\1\10\1\0"+ + "\2\10\2\7\2\0\12\5\22\0\2\7\1\0\10\10\1\0\3\10"+ + "\1\0\27\10\1\0\20\10\3\0\1\10\7\7\1\0\3\7\1\0"+ + "\4\7\11\0\1\7\10\0\2\10\2\7\2\0\12\5\12\0\6\10"+ + "\2\0\2\7\1\0\22\10\3\0\30\10\1\0\11\10\1\0\1\10"+ + "\2\0\7\10\3\0\1\7\4\0\6\7\1\0\1\7\1\0\10\7"+ + "\22\0\2\7\75\0\1\7\2\0\7\7\14\0\10\7\1\0\12\5"+ + "\127\0\1\7\2\0\6\7\1\0\2\7\13\0\6\7\2\0\12\5"+ + "\46\0\1\10\27\0\2\7\6\0\12\5\13\0\1\7\1\0\1\7"+ + "\1\0\1\7\4\0\2\7\10\10\1\0\44\10\4\0\24\7\1\0"+ + "\2\7\4\10\4\0\10\7\1\0\44\7\11\0\1\7\144\0\24\7"+ + "\1\0\12\5\14\0\4\7\4\0\3\7\1\0\3\7\2\0\7\7"+ + "\3\0\4\7\15\0\14\7\1\0\1\7\12\5\4\7\2\0\46\10"+ + "\12\0\53\10\1\0\1\10\3\0\u0149\10\1\0\4\10\2\0\7\10"+ + "\1\0\1\10\1\0\4\10\2\0\51\10\1\0\4\10\2\0\41\10"+ + "\1\0\4\10\2\0\7\10\1\0\1\10\1\0\4\10\2\0\17\10"+ + "\1\0\71\10\1\0\4\10\2\0\103\10\4\0\1\7\40\0\20\10"+ + "\20\0\125\10\14\0\u026c\10\2\0\21\10\1\4\32\10\5\0\113\10"+ + "\3\0\3\10\17\0\15\10\1\0\4\10\3\7\13\0\22\10\3\7"+ + "\13\0\22\10\2\7\14\0\15\10\1\0\3\10\1\0\2\7\100\0"+ + "\40\7\11\0\1\7\2\0\12\5\41\0\3\7\1\4\1\0\12\5"+ + "\6\0\130\10\10\0\51\10\1\7\1\10\5\0\106\10\12\0\35\10"+ + "\3\0\14\7\4\0\14\7\12\0\12\5\140\0\21\7\7\0\2\7"+ + "\6\0\13\5\45\0\27\10\5\7\71\0\12\7\1\0\35\7\2\0"+ + "\1\7\12\5\6\0\12\5\146\0\5\7\57\10\21\7\7\10\4\0"+ + "\12\5\21\0\11\7\14\0\3\7\36\10\12\7\3\0\2\10\12\5"+ + "\106\0\44\10\24\7\10\0\12\5\3\0\3\10\12\5\44\10\122\0"+ + "\3\7\1\0\25\7\4\10\1\7\4\10\1\7\15\0\300\10\47\7"+ + "\26\0\3\7\u0116\10\2\0\6\10\2\0\46\10\2\0\6\10\2\0"+ + "\10\10\1\0\1\10\1\0\1\10\1\0\1\10\1\0\37\10\2\0"+ + "\65\10\1\0\7\10\1\0\1\10\3\0\3\10\1\0\7\10\3\0"+ + "\4\10\2\0\6\10\4\0\15\10\5\0\3\10\1\0\7\10\3\0"+ + "\13\4\1\0\4\7\10\0\2\13\12\0\1\13\2\0\1\11\2\3"+ + "\5\7\1\4\17\0\2\15\3\0\1\12\17\0\1\15\12\0\1\4"+ + "\5\7\5\0\6\7\1\0\1\10\15\0\1\10\20\0\5\10\73\0"+ + "\41\7\21\0\1\10\4\0\1\10\2\0\12\10\1\0\1\10\3\0"+ + "\5\10\6\0\1\10\1\0\1\10\1\0\1\10\1\0\4\10\1\0"+ + "\13\10\2\0\4\10\5\0\5\10\4\0\1\10\21\0\51\10\u032d\0"+ + "\64\10\u0716\0\57\10\1\0\57\10\1\0\205\10\6\0\4\10\3\7"+ + "\16\0\46\10\12\0\66\10\11\0\1\10\20\0\27\10\11\0\7\10"+ + "\1\0\7\10\1\0\7\10\1\0\7\10\1\0\7\10\1\0\7\10"+ + "\1\0\7\10\1\0\7\10\1\0\40\7\57\0\1\10\120\0\32\6"+ + "\1\0\131\6\14\0\326\6\52\0\1\4\4\0\1\10\1\0\1\6"+ + "\31\0\11\6\6\7\1\0\5\14\2\0\3\6\2\10\134\0\2\7"+ + "\2\14\3\0\133\14\1\0\4\14\5\0\51\10\3\0\136\10\21\0"+ + "\30\10\70\0\20\14\320\0\57\14\1\0\130\14\250\0\u19b6\6\112\0"+ + "\u51cc\6\64\0\u048d\10\103\0\56\10\2\0\u010d\10\3\0\20\10\12\5"+ + "\2\10\24\0\40\10\2\0\15\10\4\7\11\0\2\7\1\0\31\10"+ + "\10\0\120\10\2\7\45\0\11\10\2\0\147\10\2\0\2\10\156\0"+ + "\7\10\1\7\3\10\1\7\4\10\1\7\27\10\5\7\30\0\64\10"+ + "\14\0\2\7\62\10\21\7\13\0\12\5\6\0\22\7\6\10\3\0"+ + "\1\10\4\0\12\5\34\10\10\7\2\0\27\10\15\7\14\0\35\10"+ + "\3\0\4\7\57\10\16\7\16\0\1\10\12\5\46\0\51\10\16\7"+ + "\11\0\3\10\1\7\10\10\2\7\2\0\12\5\41\0\1\7\64\0"+ + "\1\7\1\0\3\7\2\0\2\7\5\0\2\7\1\0\1\7\376\0"+ + "\43\10\10\7\1\0\2\7\2\0\12\5\6\0\u2ba4\10\14\0\27\10"+ + "\4\0\61\10\u2104\0\u012e\6\2\0\76\6\2\0\152\6\46\0\7\10"+ + "\14\0\5\10\5\0\1\10\1\7\12\10\1\0\15\10\1\0\5\10"+ + "\1\0\1\10\1\0\2\10\1\0\2\10\1\0\154\10\41\0\u016b\10"+ + "\22\0\100\10\2\0\66\10\50\0\14\10\4\0\20\7\1\12\2\0"+ + "\1\11\1\12\13\0\7\7\14\0\2\15\30\0\3\15\1\12\1\0"+ + "\1\13\1\0\1\12\1\11\32\0\5\10\1\0\207\10\2\0\1\7"+ + "\7\0\1\13\4\0\1\12\1\0\1\13\13\0\1\11\1\12\5\0"+ + "\32\10\4\0\1\15\1\0\32\10\13\0\70\14\2\7\37\10\3\0"+ + "\6\10\2\0\6\10\2\0\6\10\2\0\3\10\34\0\3\7\4\0"; + + /** + * Translates characters to character classes + */ + private static final char [] ZZ_CMAP = zzUnpackCMap(ZZ_CMAP_PACKED); + + /** + * Translates DFA states to action switch labels. + */ + private static final int [] ZZ_ACTION = zzUnpackAction(); + + private static final String ZZ_ACTION_PACKED_0 = + "\1\0\4\1\1\2\3\3\1\1\1\0\1\4\1\5"+ + "\1\6\2\0\1\7\1\10\1\11\2\0\1\12\1\13"+ + "\1\12\1\0\1\12\1\4\1\14\1\15\1\16\1\17"+ + "\1\20\1\0\1\21\1\7\1\22\1\0\1\23\1\0"+ + "\1\13\2\0"; + + private static int [] zzUnpackAction() { + int [] result = new int[42]; + int offset = 0; + offset = zzUnpackAction(ZZ_ACTION_PACKED_0, offset, result); + return result; + } + + private static int zzUnpackAction(String packed, int offset, int [] result) { + int i = 0; /* index in packed string */ + int j = offset; /* index in unpacked array */ + int l = packed.length(); + while (i < l) { + int count = packed.charAt(i++); + int value = packed.charAt(i++); + do result[j++] = value; while (--count > 0); + } + return j; + } + + + /** + * Translates a state to a row index in the transition table + */ + private static final int [] ZZ_ROWMAP = zzUnpackRowMap(); + + private static final String ZZ_ROWMAP_PACKED_0 = + "\0\0\0\16\0\34\0\52\0\70\0\106\0\124\0\142"+ + "\0\160\0\176\0\214\0\232\0\250\0\266\0\250\0\304"+ + "\0\322\0\340\0\356\0\374\0\u010a\0\u0118\0\u0126\0\u0134"+ + "\0\u0142\0\u0150\0\u015e\0\232\0\232\0\232\0\232\0\232"+ + "\0\340\0\232\0\232\0\232\0\u0118\0\232\0\u0134\0\232"+ + "\0\u0150\0\u015e"; + + private static int [] zzUnpackRowMap() { + int [] result = new int[42]; + int offset = 0; + offset = zzUnpackRowMap(ZZ_ROWMAP_PACKED_0, offset, result); + return result; + } + + private static int zzUnpackRowMap(String packed, int offset, int [] result) { + int i = 0; /* index in packed string */ + int j = offset; /* index in unpacked array */ + int l = packed.length(); + while (i < l) { + int high = packed.charAt(i++) << 16; + result[j++] = high | packed.charAt(i++); + } + return j; + } + + /** + * The transition table of the DFA + */ + private static final int [] ZZ_TRANS = zzUnpackTrans(); + + private static final String ZZ_TRANS_PACKED_0 = + "\1\2\1\3\2\4\1\5\1\6\1\7\1\2\1\10"+ + "\3\2\1\11\1\12\1\0\3\13\3\0\1\14\6\0"+ + "\2\15\1\16\31\15\1\0\3\17\3\0\1\14\7\0"+ + "\3\20\1\0\1\21\1\0\1\22\1\23\1\0\2\24"+ + "\1\0\1\21\1\0\3\25\3\0\1\26\5\0\1\27"+ + "\1\0\3\25\1\0\1\27\1\0\1\30\1\27\1\31"+ + "\1\0\1\31\1\0\1\27\1\0\3\25\3\0\1\32"+ + "\4\0\2\27\1\0\3\13\1\0\1\23\1\0\1\33"+ + "\1\23\3\0\2\23\16\34\16\0\16\35\16\36\16\37"+ + "\16\40\5\0\1\21\1\0\1\41\1\23\1\0\2\24"+ + "\1\0\1\21\16\42\5\0\1\43\1\0\1\24\6\0"+ + "\16\44\7\0\1\45\5\0\1\27\16\46\5\0\1\27"+ + "\1\0\1\47\1\27\1\31\1\0\1\31\1\0\1\27"+ + "\7\0\1\31\1\50\14\0\1\51\4\0\2\27\5\0"+ + "\1\23\1\0\1\52\1\23\3\0\2\23"; + + private static int [] zzUnpackTrans() { + int [] result = new int[364]; + int offset = 0; + offset = zzUnpackTrans(ZZ_TRANS_PACKED_0, offset, result); + return result; + } + + private static int zzUnpackTrans(String packed, int offset, int [] result) { + int i = 0; /* index in packed string */ + int j = offset; /* index in unpacked array */ + int l = packed.length(); + while (i < l) { + int count = packed.charAt(i++); + int value = packed.charAt(i++); + value--; + do result[j++] = value; while (--count > 0); + } + return j; + } + + + /* error codes */ + private static final int ZZ_UNKNOWN_ERROR = 0; + private static final int ZZ_NO_MATCH = 1; + private static final int ZZ_PUSHBACK_2BIG = 2; + + /* error messages for the codes above */ + private static final String ZZ_ERROR_MSG[] = { + "Unkown internal scanner error", + "Error: could not match input", + "Error: pushback value was too large" + }; + + /** + * ZZ_ATTRIBUTE[aState] contains the attributes of state aState + */ + private static final int [] ZZ_ATTRIBUTE = zzUnpackAttribute(); + + private static final String ZZ_ATTRIBUTE_PACKED_0 = + "\1\0\11\1\1\0\1\11\2\1\2\0\3\1\2\0"+ + "\3\1\1\0\2\1\5\11\1\0\3\11\1\0\1\11"+ + "\1\0\1\11\2\0"; + + private static int [] zzUnpackAttribute() { + int [] result = new int[42]; + int offset = 0; + offset = zzUnpackAttribute(ZZ_ATTRIBUTE_PACKED_0, offset, result); + return result; + } + + private static int zzUnpackAttribute(String packed, int offset, int [] result) { + int i = 0; /* index in packed string */ + int j = offset; /* index in unpacked array */ + int l = packed.length(); + while (i < l) { + int count = packed.charAt(i++); + int value = packed.charAt(i++); + do result[j++] = value; while (--count > 0); + } + return j; + } + + /** the input device */ + private java.io.Reader zzReader; + + /** the current state of the DFA */ + private int zzState; + + /** the current lexical state */ + private int zzLexicalState = YYINITIAL; + + /** this buffer contains the current text to be matched and is + the source of the yytext() string */ + private char zzBuffer[] = new char[ZZ_BUFFERSIZE]; + + /** the textposition at the last accepting state */ + private int zzMarkedPos; + + /** the current text position in the buffer */ + private int zzCurrentPos; + + /** startRead marks the beginning of the yytext() string in the buffer */ + private int zzStartRead; + + /** endRead marks the last character in the buffer, that has been read + from input */ + private int zzEndRead; + + /** number of newlines encountered up to the start of the matched text */ + private int yyline; + + /** the number of characters up to the start of the matched text */ + private int yychar; + + /** + * the number of characters from the last newline up to the start of the + * matched text + */ + private int yycolumn; + + /** + * zzAtBOL == true <=> the scanner is currently at the beginning of a line + */ + private boolean zzAtBOL = true; + + /** zzAtEOF == true <=> the scanner is at the EOF */ + private boolean zzAtEOF; + + /** denotes if the user-EOF-code has already been executed */ + private boolean zzEOFDone; + + /* user code: */ + public static final String WORD_TYPE = ""; + public static final String NUM_TYPE = ""; + + private final CharTermAttribute termAtt = addAttribute(CharTermAttribute.class); + private final OffsetAttribute offsetAtt = addAttribute(OffsetAttribute.class); + private final PositionIncrementAttribute posIncrAtt + = addAttribute(PositionIncrementAttribute.class); + private final TypeAttribute typeAtt = addAttribute(TypeAttribute.class); + + private int maxTokenLength = StandardAnalyzer.DEFAULT_MAX_TOKEN_LENGTH; + + private int matchStart; + private int posIncr; + private boolean hasLetter = false; + private boolean hasNumber = false; + private boolean isTokenTooLong; // tracks partial match token lengths + + + /** + * @param source The AttributeSource to use + * @param input The input reader + */ + public UAX29Tokenizer(AttributeSource source, Reader input) { + super(source, input); + zzReader = input; + } + + /** + * @param factory The AttributeFactory to use + * @param input The input reader + */ + public UAX29Tokenizer(AttributeFactory factory, Reader input) { + super(factory, input); + zzReader = input; + } + + @Override + public final void end() { + // set final offset + int finalOffset = correctOffset(yychar + yylength()); + offsetAtt.setOffset(finalOffset, finalOffset); + } + + @Override + public void reset(Reader reader) throws IOException { + super.reset(reader); + if (zzBuffer.length > ZZ_BUFFERSIZE) { + zzBuffer = new char[ZZ_BUFFERSIZE]; + } + yyreset(reader); + } + + /** + * Appends the current (potentially partial token) match to this TokenStream's + * CharTermAttribute. + *

+ * If maxTokenLength is exceeded, the CharTermAttribute is set back to empty + * and false is returned. + * + * @param isTokenBoundary true if there is a token boundary after the current + * match; false otherwise. + * @return true there is a token available; false otherwise + */ + private boolean updateAttributes(boolean isTokenBoundary) { + boolean isTokenAvailable = false; + if (isTokenTooLong || termAtt.length() + yylength() > maxTokenLength) { + // previous or current partial match made the current token too long + if (isTokenBoundary) { + // When we skip a too-long token, we still increment the position + // increment, but only if the token would have otherwise been output + // (that is, only if it contains Letters and/or Numbers.) + if (hasLetter || hasNumber) { + ++posIncr; + } + clearAttributes(); + isTokenTooLong = false; // reset for the next token + } else { + // when the next token boundary is encountered, the CharTermAttribute + // will be emptied - no need to do it here. + isTokenTooLong = true; + } + } else if (isTokenBoundary) { + if (hasLetter || hasNumber) { + if (0 == termAtt.length()) { + matchStart = yychar; + } + int newLength = termAtt.length() + yylength(); + char targetBuf[] = termAtt.resizeBuffer(newLength); + System.arraycopy(zzBuffer, zzStartRead, targetBuf, termAtt.length(), yylength()); + termAtt.setLength(newLength); + posIncrAtt.setPositionIncrement(posIncr); + offsetAtt.setOffset(correctOffset(matchStart), + correctOffset(matchStart + termAtt.length())); + typeAtt.setType(hasLetter ? WORD_TYPE : NUM_TYPE); + isTokenAvailable = true; + } else { + clearAttributes(); + } + } else { // this is not a token boundary + if (0 == termAtt.length()) { + matchStart = yychar; + } + int newLength = termAtt.length() + yylength(); + char targetBuf[] = termAtt.resizeBuffer(newLength); + System.arraycopy(zzBuffer, zzStartRead, targetBuf, termAtt.length(), yylength()); + termAtt.setLength(newLength); + } + return isTokenAvailable; + } + + /** + * Set the max allowed token length. Any token longer than this is skipped. + * @param length the new max allowed token length + */ + public void setMaxTokenLength(int length) { + this.maxTokenLength = length; + } + + /** + * Returns the max allowed token length. Any token longer than this is + * skipped. + * @return the max allowed token length + */ + public int getMaxTokenLength() { + return maxTokenLength; + } + + @Override + public final boolean incrementToken() throws IOException { + // This method is required because of two JFlex limitations: + // 1. No way to insert code at the beginning of the generated scanning + // get-next-token method; and + // 2. No way to declare @Override on the generated scanning method. + clearAttributes(); + posIncr = 1; + hasLetter = hasNumber = false; + return getNextToken(); + } + + + /** + * Creates a new scanner + * There is also a java.io.InputStream version of this constructor. + * + * @param in the java.io.Reader to read input from. + */ + public UAX29Tokenizer(java.io.Reader in) { + super(in); + this.zzReader = in; + } + + /** + * Creates a new scanner. + * There is also java.io.Reader version of this constructor. + * + * @param in the java.io.Inputstream to read input from. + */ + public UAX29Tokenizer(java.io.InputStream in) { + this(new java.io.InputStreamReader(in)); + } + + /** + * Unpacks the compressed character translation table. + * + * @param packed the packed character translation table + * @return the unpacked character translation table + */ + private static char [] zzUnpackCMap(String packed) { + char [] map = new char[0x10000]; + int i = 0; /* index in packed string */ + int j = 0; /* index in unpacked array */ + while (i < 2060) { + int count = packed.charAt(i++); + char value = packed.charAt(i++); + do map[j++] = value; while (--count > 0); + } + return map; + } + + + /** + * Refills the input buffer. + * + * @return false, iff there was new input. + * + * @exception java.io.IOException if any I/O-Error occurs + */ + private boolean zzRefill() throws java.io.IOException { + + /* first: make room (if you can) */ + if (zzStartRead > 0) { + System.arraycopy(zzBuffer, zzStartRead, + zzBuffer, 0, + zzEndRead-zzStartRead); + + /* translate stored positions */ + zzEndRead-= zzStartRead; + zzCurrentPos-= zzStartRead; + zzMarkedPos-= zzStartRead; + zzStartRead = 0; + } + + /* is the buffer big enough? */ + if (zzCurrentPos >= zzBuffer.length) { + /* if not: blow it up */ + char newBuffer[] = new char[zzCurrentPos*2]; + System.arraycopy(zzBuffer, 0, newBuffer, 0, zzBuffer.length); + zzBuffer = newBuffer; + } + + /* finally: fill the buffer with new input */ + int numRead = zzReader.read(zzBuffer, zzEndRead, + zzBuffer.length-zzEndRead); + + if (numRead > 0) { + zzEndRead+= numRead; + return false; + } + // unlikely but not impossible: read 0 characters, but not at end of stream + if (numRead == 0) { + int c = zzReader.read(); + if (c == -1) { + return true; + } else { + zzBuffer[zzEndRead++] = (char) c; + return false; + } + } + + // numRead < 0 + return true; + } + + + /** + * Closes the input stream. + */ + public final void yyclose() throws java.io.IOException { + zzAtEOF = true; /* indicate end of file */ + zzEndRead = zzStartRead; /* invalidate buffer */ + + if (zzReader != null) + zzReader.close(); + } + + + /** + * Resets the scanner to read from a new input stream. + * Does not close the old reader. + * + * All internal variables are reset, the old input stream + * cannot be reused (internal buffer is discarded and lost). + * Lexical state is set to ZZ_INITIAL. + * + * @param reader the new input stream + */ + public final void yyreset(java.io.Reader reader) { + zzReader = reader; + zzAtBOL = true; + zzAtEOF = false; + zzEOFDone = false; + zzEndRead = zzStartRead = 0; + zzCurrentPos = zzMarkedPos = 0; + yyline = yychar = yycolumn = 0; + zzLexicalState = YYINITIAL; + } + + + /** + * Returns the current lexical state. + */ + public final int yystate() { + return zzLexicalState; + } + + + /** + * Enters a new lexical state + * + * @param newState the new lexical state + */ + public final void yybegin(int newState) { + zzLexicalState = newState; + } + + + /** + * Returns the text matched by the current regular expression. + */ + public final String yytext() { + return new String( zzBuffer, zzStartRead, zzMarkedPos-zzStartRead ); + } + + + /** + * Returns the character at position pos from the + * matched text. + * + * It is equivalent to yytext().charAt(pos), but faster + * + * @param pos the position of the character to fetch. + * A value from 0 to yylength()-1. + * + * @return the character at position pos + */ + public final char yycharat(int pos) { + return zzBuffer[zzStartRead+pos]; + } + + + /** + * Returns the length of the matched text region. + */ + public final int yylength() { + return zzMarkedPos-zzStartRead; + } + + + /** + * Reports an error that occured while scanning. + * + * In a wellformed scanner (no or only correct usage of + * yypushback(int) and a match-all fallback rule) this method + * will only be called with things that "Can't Possibly Happen". + * If this method is called, something is seriously wrong + * (e.g. a JFlex bug producing a faulty scanner etc.). + * + * Usual syntax/scanner level error handling should be done + * in error fallback rules. + * + * @param errorCode the code of the errormessage to display + */ + private void zzScanError(int errorCode) { + String message; + try { + message = ZZ_ERROR_MSG[errorCode]; + } + catch (ArrayIndexOutOfBoundsException e) { + message = ZZ_ERROR_MSG[ZZ_UNKNOWN_ERROR]; + } + + throw new Error(message); + } + + + /** + * Pushes the specified amount of characters back into the input stream. + * + * They will be read again by then next call of the scanning method + * + * @param number the number of characters to be read again. + * This number must not be greater than yylength()! + */ + public void yypushback(int number) { + if ( number > yylength() ) + zzScanError(ZZ_PUSHBACK_2BIG); + + zzMarkedPos -= number; + } + + + /** + * Resumes scanning until the next regular expression is matched, + * the end of input is encountered or an I/O-Error occurs. + * + * @return the next token + * @exception java.io.IOException if any I/O-Error occurs + */ + public boolean getNextToken() throws java.io.IOException { + int zzInput; + int zzAction; + + // cached fields: + int zzCurrentPosL; + int zzMarkedPosL; + int zzEndReadL = zzEndRead; + char [] zzBufferL = zzBuffer; + char [] zzCMapL = ZZ_CMAP; + + int [] zzTransL = ZZ_TRANS; + int [] zzRowMapL = ZZ_ROWMAP; + int [] zzAttrL = ZZ_ATTRIBUTE; + + while (true) { + zzMarkedPosL = zzMarkedPos; + + yychar+= zzMarkedPosL-zzStartRead; + + zzAction = -1; + + zzCurrentPosL = zzCurrentPos = zzStartRead = zzMarkedPosL; + + zzState = ZZ_LEXSTATE[zzLexicalState]; + + + zzForAction: { + while (true) { + + if (zzCurrentPosL < zzEndReadL) + zzInput = zzBufferL[zzCurrentPosL++]; + else if (zzAtEOF) { + zzInput = YYEOF; + break zzForAction; + } + else { + // store back cached positions + zzCurrentPos = zzCurrentPosL; + zzMarkedPos = zzMarkedPosL; + boolean eof = zzRefill(); + // get translated positions and possibly new buffer + zzCurrentPosL = zzCurrentPos; + zzMarkedPosL = zzMarkedPos; + zzBufferL = zzBuffer; + zzEndReadL = zzEndRead; + if (eof) { + zzInput = YYEOF; + break zzForAction; + } + else { + zzInput = zzBufferL[zzCurrentPosL++]; + } + } + int zzNext = zzTransL[ zzRowMapL[zzState] + zzCMapL[zzInput] ]; + if (zzNext == -1) break zzForAction; + zzState = zzNext; + + int zzAttributes = zzAttrL[zzState]; + if ( (zzAttributes & 1) == 1 ) { + zzAction = zzState; + zzMarkedPosL = zzCurrentPosL; + if ( (zzAttributes & 8) == 8 ) break zzForAction; + } + + } + } + + // store back cached position + zzMarkedPos = zzMarkedPosL; + + switch (zzAction < 0 ? zzAction : ZZ_ACTION[zzAction]) { + case 14: + // lookahead expression with fixed base length + zzMarkedPos = zzStartRead + 2; + { /* ignore whitespace */ + } + case 20: break; + case 6: + { /* ignore whitespace */ + } + case 21: break; + case 13: + // lookahead expression with fixed base length + zzMarkedPos = zzStartRead + 1; + { /* ignore whitespace */ + } + case 22: break; + case 16: + // lookahead expression with fixed lookahead length + yypushback(2); + { hasNumber = true; updateAttributes(false); + } + case 23: break; + case 7: + // lookahead expression with fixed lookahead length + yypushback(1); + { hasNumber = true; updateAttributes(false); + } + case 24: break; + case 8: + // lookahead expression with fixed base length + zzMarkedPos = zzStartRead + 1; + { hasNumber = true; updateAttributes(false); + } + case 25: break; + case 5: + // lookahead expression with fixed base length + zzMarkedPos = zzStartRead + 1; + { /* Ignore whitespace */ + } + case 26: break; + case 17: + // lookahead expression with fixed lookahead length + yypushback(2); + { updateAttributes(false); + } + case 27: break; + case 4: + // lookahead expression with fixed base length + zzMarkedPos = zzStartRead + 1; + { updateAttributes(false); + } + case 28: break; + case 9: + // lookahead expression with fixed lookahead length + yypushback(1); + { updateAttributes(false); + } + case 29: break; + case 19: + // lookahead expression with fixed lookahead length + yypushback(2); + { hasLetter = true; updateAttributes(false); + } + case 30: break; + case 10: + // lookahead expression with fixed base length + zzMarkedPos = zzStartRead + 1; + { hasLetter = true; updateAttributes(false); + } + case 31: break; + case 11: + // lookahead expression with fixed lookahead length + yypushback(1); + { hasLetter = true; updateAttributes(false); + } + case 32: break; + case 1: + { if (updateAttributes(true)) return true; + } + case 33: break; + case 12: + // lookahead expression with fixed base length + zzMarkedPos = zzStartRead + 1; + { if (updateAttributes(true)) return true; + } + case 34: break; + case 15: + // lookahead expression with fixed base length + zzMarkedPos = zzStartRead + 1; + { hasNumber = true; if (updateAttributes(true)) return true; + } + case 35: break; + case 2: + { hasNumber = true; if (updateAttributes(true)) return true; + } + case 36: break; + case 18: + // lookahead expression with fixed base length + zzMarkedPos = zzStartRead + 1; + { hasLetter = true; if (updateAttributes(true)) return true; + } + case 37: break; + case 3: + { hasLetter = true; if (updateAttributes(true)) return true; + } + case 38: break; + default: + if (zzInput == YYEOF && zzStartRead == zzCurrentPos) { + zzAtEOF = true; + { + return false; + } + } + else { + zzScanError(ZZ_NO_MATCH); + } + } + } + } + + +} Index: lucene/src/java/org/apache/lucene/analysis/standard/UAX29Tokenizer.jflex =================================================================== --- lucene/src/java/org/apache/lucene/analysis/standard/UAX29Tokenizer.jflex (revision 0) +++ lucene/src/java/org/apache/lucene/analysis/standard/UAX29Tokenizer.jflex (revision 0) @@ -0,0 +1,333 @@ +package org.apache.lucene.analysis.standard; + +/** + * Licensed to the Apache Software Foundation (ASF) under one or more + * contributor license agreements. See the NOTICE file distributed with + * this work for additional information regarding copyright ownership. + * The ASF licenses this file to You under the Apache License, Version 2.0 + * (the "License"); you may not use this file except in compliance with + * the License. You may obtain a copy of the License at + * + * http://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, software + * distributed under the License is distributed on an "AS IS" BASIS, + * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. + * See the License for the specific language governing permissions and + * limitations under the License. + */ + +import java.io.IOException; +import java.io.Reader; +import java.nio.CharBuffer; + +import org.apache.lucene.analysis.Tokenizer; +import org.apache.lucene.analysis.tokenattributes.OffsetAttribute; +import org.apache.lucene.analysis.tokenattributes.PositionIncrementAttribute; +import org.apache.lucene.analysis.tokenattributes.CharTermAttribute; +import org.apache.lucene.analysis.tokenattributes.TypeAttribute; +import org.apache.lucene.util.AttributeSource; + + +/** + * This class implements the Unicode Text Segmentation algorithm, as specified + * in Unicode Standard Annex #29 . + */ +%% + +%unicode 5.2 +%public +%class UAX29Tokenizer +%extends Tokenizer +%type boolean +%function getNextToken +%char + +%init{ + super(in); +%init} + + +%{ + public static final String WORD_TYPE = ""; + public static final String NUM_TYPE = ""; + + private final CharTermAttribute termAtt = addAttribute(CharTermAttribute.class); + private final OffsetAttribute offsetAtt = addAttribute(OffsetAttribute.class); + private final PositionIncrementAttribute posIncrAtt + = addAttribute(PositionIncrementAttribute.class); + private final TypeAttribute typeAtt = addAttribute(TypeAttribute.class); + + private int maxTokenLength = StandardAnalyzer.DEFAULT_MAX_TOKEN_LENGTH; + + private int matchStart; + private int posIncr; + private boolean hasLetter = false; + private boolean hasNumber = false; + private boolean isTokenTooLong; // tracks partial match token lengths + + + /** + * @param source The AttributeSource to use + * @param input The input reader + */ + public UAX29Tokenizer(AttributeSource source, Reader input) { + super(source, input); + zzReader = input; + } + + /** + * @param factory The AttributeFactory to use + * @param input The input reader + */ + public UAX29Tokenizer(AttributeFactory factory, Reader input) { + super(factory, input); + zzReader = input; + } + + @Override + public final void end() { + // set final offset + int finalOffset = correctOffset(yychar + yylength()); + offsetAtt.setOffset(finalOffset, finalOffset); + } + + @Override + public void reset(Reader reader) throws IOException { + super.reset(reader); + if (zzBuffer.length > ZZ_BUFFERSIZE) { + zzBuffer = new char[ZZ_BUFFERSIZE]; + } + yyreset(reader); + } + + /** + * Appends the current (potentially partial token) match to this TokenStream's + * CharTermAttribute. + *

+ * If maxTokenLength is exceeded, the CharTermAttribute is set back to empty + * and false is returned. + * + * @param isTokenBoundary true if there is a token boundary after the current + * match; false otherwise. + * @return true there is a token available; false otherwise + */ + private boolean updateAttributes(boolean isTokenBoundary) { + boolean isTokenAvailable = false; + if (isTokenTooLong || termAtt.length() + yylength() > maxTokenLength) { + // previous or current partial match made the current token too long + if (isTokenBoundary) { + // When we skip a too-long token, we still increment the position + // increment, but only if the token would have otherwise been output + // (that is, only if it contains Letters and/or Numbers.) + if (hasLetter || hasNumber) { + ++posIncr; + } + clearAttributes(); + isTokenTooLong = false; // reset for the next token + } else { + // when the next token boundary is encountered, the CharTermAttribute + // will be emptied - no need to do it here. + isTokenTooLong = true; + } + } else if (isTokenBoundary) { + if (hasLetter || hasNumber) { + if (0 == termAtt.length()) { + matchStart = yychar; + } + int newLength = termAtt.length() + yylength(); + char targetBuf[] = termAtt.resizeBuffer(newLength); + System.arraycopy(zzBuffer, zzStartRead, targetBuf, termAtt.length(), yylength()); + termAtt.setLength(newLength); + posIncrAtt.setPositionIncrement(posIncr); + offsetAtt.setOffset(correctOffset(matchStart), + correctOffset(matchStart + termAtt.length())); + typeAtt.setType(hasLetter ? WORD_TYPE : NUM_TYPE); + isTokenAvailable = true; + } else { + clearAttributes(); + } + } else { // this is not a token boundary + if (0 == termAtt.length()) { + matchStart = yychar; + } + int newLength = termAtt.length() + yylength(); + char targetBuf[] = termAtt.resizeBuffer(newLength); + System.arraycopy(zzBuffer, zzStartRead, targetBuf, termAtt.length(), yylength()); + termAtt.setLength(newLength); + } + return isTokenAvailable; + } + + /** + * Set the max allowed token length. Any token longer than this is skipped. + * @param length the new max allowed token length + */ + public void setMaxTokenLength(int length) { + this.maxTokenLength = length; + } + + /** + * Returns the max allowed token length. Any token longer than this is + * skipped. + * @return the max allowed token length + */ + public int getMaxTokenLength() { + return maxTokenLength; + } + + @Override + public final boolean incrementToken() throws IOException { + // This method is required because of two JFlex limitations: + // 1. No way to insert code at the beginning of the generated scanning + // get-next-token method; and + // 2. No way to declare @Override on the generated scanning method. + clearAttributes(); + posIncr = 1; + hasLetter = hasNumber = false; + return getNextToken(); + } +%} + +%% + +// Break at the start and end of text. +// WB1. sot ÷ +// WB2. ÷ eot +<> { return false; } + + +// +// Do not break within CRLF. +// +// WB3. CR × LF +// +\p{WB:CR} \p{WB:LF} / [^] { /* ignore whitespace */ } +\p{WB:CR} \p{WB:LF} { /* ignore whitespace */ } + + +// Otherwise break before and after Newlines (including CR and LF) +// +// WB3a. (Newline | CR | LF) ÷ +[\p{WB:Newline}\p{WB:CR}\p{WB:LF}] / [^]{2} { /* ignore whitespace */ } +[\p{WB:Newline}\p{WB:CR}\p{WB:LF}] / [^] { /* Ignore whitespace */ } + + +// WB3b. ÷ (Newline | CR | LF) +// +\s / [\p{WB:Newline}\p{WB:CR}\p{WB:LF}] [^] { /* ignore whitespace */ } +\p{WB:Numeric} / [\p{WB:Newline}\p{WB:CR}\p{WB:LF}] [^] { hasNumber = true; if (updateAttributes(true)) return true; } +[\p{WB:Katakana}\p{WB:ALetter}\p{Script:Han}] / [\p{WB:Newline}\p{WB:CR}\p{WB:LF}] [^] { hasLetter = true; if (updateAttributes(true)) return true; } +[^] / [\p{WB:Newline}\p{WB:CR}\p{WB:LF}] [^] { if (updateAttributes(true)) return true; } + + +// Ignore Format and Extend characters, except when they appear at the +// beginning of a region of text. +// +// (See Section 6.2, Replacing Ignore Rules.) +// +// WB4. X (Extend | Format)* ? X +// +// --> [^ Newline CR LF ] × [Format Extend] +// +\p{WB:Numeric} / [\p{WB:Format}\p{WB:Extend}] { hasNumber = true; updateAttributes(false); } +[\p{WB:Katakana}\p{WB:ALetter}\p{Script:Han}] / [\p{WB:Format}\p{WB:Extend}] { hasLetter = true; updateAttributes(false); } +[^\p{WB:Newline}\p{WB:CR}\p{WB:LF}] / [\p{WB:Format}\p{WB:Extend}] { updateAttributes(false); } + + +// Do not break between most letters. +// +// WB5. ALetter × ALetter +// +// [included WB4. X (Extend | Format)* ? X] +// +\p{WB:ALetter} [\p{WB:Format}\p{WB:Extend}]* / \p{WB:ALetter} [^] { hasLetter = true; updateAttributes(false); } +\p{WB:ALetter} [\p{WB:Format}\p{WB:Extend}]* / \p{WB:ALetter} { hasLetter = true; updateAttributes(false); } + + +// Do not break letters across certain punctuation. +// +// WB6. ALetter × (MidLetter | MidNumLet) ALetter +// WB7. ALetter (MidLetter | MidNumLet) × ALetter +// +// [included WB4. X (Extend | Format)* ? X] +// +\p{WB:ALetter} [\p{WB:Format}\p{WB:Extend}]* [\p{WB:MidLetter}\p{WB:MidNumLet}] [\p{WB:Format}\p{WB:Extend}]* / \p{WB:ALetter} { hasLetter = true; updateAttributes(false); } + + +// Do not break within sequences of digits, or digits adjacent to letters +// (“3a”, or “A3”). +// +// WB8. Numeric × Numeric +// +// [included WB4. X (Extend | Format)* ? X] +// +\p{WB:Numeric} [\p{WB:Format}\p{WB:Extend}]* / \p{WB:Numeric} [^] { hasNumber = true; updateAttributes(false); } +\p{WB:Numeric} [\p{WB:Format}\p{WB:Extend}]* / \p{WB:Numeric} { hasNumber = true; updateAttributes(false); } + + +// WB9. ALetter × Numeric +// +// [included WB4. X (Extend | Format)* ? X] +// +\p{WB:ALetter} [\p{WB:Format}\p{WB:Extend}]* / \p{WB:Numeric} [^] { hasLetter = true; updateAttributes(false); } +\p{WB:ALetter} [\p{WB:Format}\p{WB:Extend}]* / \p{WB:Numeric} { hasLetter = true; updateAttributes(false); } + + +// WB10. Numeric × ALetter +// +// [included WB4. X (Extend | Format)* ? X] +// +\p{WB:Numeric} [\p{WB:Format}\p{WB:Extend}]* / \p{WB:ALetter} [^] { updateAttributes(false); } +\p{WB:Numeric} [\p{WB:Format}\p{WB:Extend}]* / \p{WB:ALetter} { updateAttributes(false); } + + +// Do not break within sequences, such as “3.2” or “3,456.789”. +// +// WB11. Numeric (MidNum | MidNumLet) × Numeric +// WB12. Numeric × (MidNum | MidNumLet) Numeric +// +// [included WB4. X (Extend | Format)* ? X] +// +\p{WB:Numeric} [\p{WB:Format}\p{WB:Extend}]* [\p{WB:MidNum}\p{WB:MidNumLet}] [\p{WB:Format}\p{WB:Extend}]* / \p{WB:Numeric} { hasNumber = true; updateAttributes(false); } + + +// Do not break between Katakana. +// +// WB13. Katakana × Katakana +// +// [included WB4. X (Extend | Format)* ? X] +// +\p{WB:Katakana} [\p{WB:Format}\p{WB:Extend}]* / \p{WB:Katakana} [^] { hasLetter = true; updateAttributes(false); } +\p{WB:Katakana} [\p{WB:Format}\p{WB:Extend}]* / \p{WB:Katakana} { hasLetter = true; updateAttributes(false); } + + +// Do not break from extenders. +// +// WB13a. (ALetter | Numeric | Katakana | ExtendNumLet) × ExtendNumLet +// +// [included WB4. X (Extend | Format)* ? X] +// +\p{WB:Numeric} [\p{WB:Format}\p{WB:Extend}]* / \p{WB:ExtendNumLet} [^] { hasNumber = true; updateAttributes(false); } +[\p{WB:ALetter}\p{WB:Katakana}\p{Script:Han}] [\p{WB:Format}\p{WB:Extend}]* / \p{WB:ExtendNumLet} [^] { hasLetter = true; updateAttributes(false); } +\p{WB:ExtendNumLet} [\p{WB:Format}\p{WB:Extend}]* / \p{WB:ExtendNumLet} [^] { updateAttributes(false); } +\p{WB:Numeric} [\p{WB:Format}\p{WB:Extend}]* / \p{WB:ExtendNumLet} { hasNumber = true; updateAttributes(false); } +[\p{WB:ALetter}\p{WB:Katakana}\p{Script:Han}] [\p{WB:Format}\p{WB:Extend}]* / \p{WB:ExtendNumLet} { hasLetter = true; updateAttributes(false); } +\p{WB:ExtendNumLet} [\p{WB:Format}\p{WB:Extend}]* / \p{WB:ExtendNumLet} { updateAttributes(false); } + + +// WB13b. ExtendNumLet × (ALetter | Numeric | Katakana) +// +// [included WB4. X (Extend | Format)* ? X] +// +\p{WB:ExtendNumLet} [\p{WB:Format}\p{WB:Extend}]* / [\p{WB:ALetter}\p{WB:Numeric}\p{WB:Katakana}] [^] { updateAttributes(false); } +\p{WB:ExtendNumLet} [\p{WB:Format}\p{WB:Extend}]* / [\p{WB:ALetter}\p{WB:Numeric}\p{WB:Katakana}] { updateAttributes(false); } + + +// Otherwise, break everywhere (including around ideographs). +// +// WB14. Any ÷ Any +// +\p{WB:Numeric} { hasNumber = true; if (updateAttributes(true)) return true; } +[\p{WB:ALetter}\p{WB:Katakana}\p{Script:Han}] { hasLetter = true; if (updateAttributes(true)) return true; } +[^] { if (updateAttributes(true)) return true; } Index: lucene/build.xml =================================================================== --- lucene/build.xml (revision 941990) +++ lucene/build.xml (working copy) @@ -661,6 +661,9 @@ +