1
package org.apache.lucene.analysis.standard;
4
* Licensed to the Apache Software Foundation (ASF) under one or more
5
* contributor license agreements. See the NOTICE file distributed with
6
* this work for additional information regarding copyright ownership.
7
* The ASF licenses this file to You under the Apache License, Version 2.0
8
* (the "License"); you may not use this file except in compliance with
9
* the License. You may obtain a copy of the License at
11
* http://www.apache.org/licenses/LICENSE-2.0
13
* Unless required by applicable law or agreed to in writing, software
14
* distributed under the License is distributed on an "AS IS" BASIS,
15
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
16
* See the License for the specific language governing permissions and
17
* limitations under the License.
20
import org.apache.lucene.analysis.tokenattributes.CharTermAttribute;
23
* This class implements Word Break rules from the Unicode Text Segmentation
24
* algorithm, as specified in
25
* <a href="http://unicode.org/reports/tr29/">Unicode Standard Annex #29</a>
26
* URLs and email addresses are also tokenized according to the relevant RFCs.
28
* Tokens produced are of the following types:
30
* <li><ALPHANUM>: A sequence of alphabetic and numeric characters</li>
31
* <li><NUM>: A number</li>
32
* <li><URL>: A URL</li>
33
* <li><EMAIL>: An email address</li>
34
* <li><SOUTHEAST_ASIAN>: A sequence of characters from South and Southeast
35
* Asian languages, including Thai, Lao, Myanmar, and Khmer</li>
36
* <li><IDEOGRAPHIC>: A single CJKV ideographic character</li>
37
* <li><HIRAGANA>: A single hiragana character</li>
46
%class UAX29URLEmailTokenizerImpl
47
%implements StandardTokenizerInterface
48
%function getNextToken
51
%include src/java/org/apache/lucene/analysis/standard/SUPPLEMENTARY.jflex-macro
52
ALetter = ([\p{WB:ALetter}] | {ALetterSupp})
53
Format = ([\p{WB:Format}] | {FormatSupp})
54
Numeric = ([\p{WB:Numeric}] | {NumericSupp})
55
Extend = ([\p{WB:Extend}] | {ExtendSupp})
56
Katakana = ([\p{WB:Katakana}] | {KatakanaSupp})
57
MidLetter = ([\p{WB:MidLetter}] | {MidLetterSupp})
58
MidNum = ([\p{WB:MidNum}] | {MidNumSupp})
59
MidNumLet = ([\p{WB:MidNumLet}] | {MidNumLetSupp})
60
ExtendNumLet = ([\p{WB:ExtendNumLet}] | {ExtendNumLetSupp})
61
ComplexContext = ([\p{LB:Complex_Context}] | {ComplexContextSupp})
62
Han = ([\p{Script:Han}] | {HanSupp})
63
Hiragana = ([\p{Script:Hiragana}] | {HiraganaSupp})
65
// Script=Hangul & Aletter
66
HangulEx = (!(!\p{Script:Hangul}|!\p{WB:ALetter})) ({Format} | {Extend})*
67
// UAX#29 WB4. X (Extend | Format)* --> X
69
ALetterEx = {ALetter} ({Format} | {Extend})*
70
// TODO: Convert hard-coded full-width numeric range to property intersection (something like [\p{Full-Width}&&\p{Numeric}]) once JFlex supports it
71
NumericEx = ({Numeric} | [\uFF10-\uFF19]) ({Format} | {Extend})*
72
KatakanaEx = {Katakana} ({Format} | {Extend})*
73
MidLetterEx = ({MidLetter} | {MidNumLet}) ({Format} | {Extend})*
74
MidNumericEx = ({MidNum} | {MidNumLet}) ({Format} | {Extend})*
75
ExtendNumLetEx = {ExtendNumLet} ({Format} | {Extend})*
77
HanEx = {Han} ({Format} | {Extend})*
78
HiraganaEx = {Hiragana} ({Format} | {Extend})*
80
// URL and E-mail syntax specifications:
82
// RFC-952: DOD INTERNET HOST TABLE SPECIFICATION
83
// RFC-1035: DOMAIN NAMES - IMPLEMENTATION AND SPECIFICATION
84
// RFC-1123: Requirements for Internet Hosts - Application and Support
85
// RFC-1738: Uniform Resource Locators (URL)
86
// RFC-3986: Uniform Resource Identifier (URI): Generic Syntax
87
// RFC-5234: Augmented BNF for Syntax Specifications: ABNF
88
// RFC-5321: Simple Mail Transfer Protocol
89
// RFC-5322: Internet Message Format
91
%include src/java/org/apache/lucene/analysis/standard/ASCIITLD.jflex-macro
93
DomainLabel = [A-Za-z0-9] ([-A-Za-z0-9]* [A-Za-z0-9])?
94
DomainNameStrict = {DomainLabel} ("." {DomainLabel})* {ASCIITLD}
95
DomainNameLoose = {DomainLabel} ("." {DomainLabel})*
97
IPv4DecimalOctet = "0"{0,2} [0-9] | "0"? [1-9][0-9] | "1" [0-9][0-9] | "2" ([0-4][0-9] | "5" [0-5])
98
IPv4Address = {IPv4DecimalOctet} ("." {IPv4DecimalOctet}){3}
99
IPv6Hex16Bit = [0-9A-Fa-f]{1,4}
100
IPv6LeastSignificant32Bits = {IPv4Address} | ({IPv6Hex16Bit} ":" {IPv6Hex16Bit})
101
IPv6Address = ({IPv6Hex16Bit} ":"){6} {IPv6LeastSignificant32Bits}
102
| "::" ({IPv6Hex16Bit} ":"){5} {IPv6LeastSignificant32Bits}
103
| {IPv6Hex16Bit}? "::" ({IPv6Hex16Bit} ":"){4} {IPv6LeastSignificant32Bits}
104
| (({IPv6Hex16Bit} ":"){0,1} {IPv6Hex16Bit})? "::" ({IPv6Hex16Bit} ":"){3} {IPv6LeastSignificant32Bits}
105
| (({IPv6Hex16Bit} ":"){0,2} {IPv6Hex16Bit})? "::" ({IPv6Hex16Bit} ":"){2} {IPv6LeastSignificant32Bits}
106
| (({IPv6Hex16Bit} ":"){0,3} {IPv6Hex16Bit})? "::" {IPv6Hex16Bit} ":" {IPv6LeastSignificant32Bits}
107
| (({IPv6Hex16Bit} ":"){0,4} {IPv6Hex16Bit})? "::" {IPv6LeastSignificant32Bits}
108
| (({IPv6Hex16Bit} ":"){0,5} {IPv6Hex16Bit})? "::" {IPv6Hex16Bit}
109
| (({IPv6Hex16Bit} ":"){0,6} {IPv6Hex16Bit})? "::"
111
URIunreserved = [-._~A-Za-z0-9]
112
URIpercentEncoded = "%" [0-9A-Fa-f]{2}
113
URIsubDelims = [!$&'()*+,;=]
114
URIloginSegment = ({URIunreserved} | {URIpercentEncoded} | {URIsubDelims})*
115
URIlogin = {URIloginSegment} (":" {URIloginSegment})? "@"
116
URIquery = "?" ({URIunreserved} | {URIpercentEncoded} | {URIsubDelims} | [:@/?])*
117
URIfragment = "#" ({URIunreserved} | {URIpercentEncoded} | {URIsubDelims} | [:@/?])*
118
URIport = ":" [0-9]{1,5}
119
URIhostStrict = ("[" {IPv6Address} "]") | {IPv4Address} | {DomainNameStrict}
120
URIhostLoose = ("[" {IPv6Address} "]") | {IPv4Address} | {DomainNameLoose}
122
URIauthorityStrict = {URIhostStrict} {URIport}?
123
URIauthorityLoose = {URIlogin}? {URIhostLoose} {URIport}?
125
HTTPsegment = ({URIunreserved} | {URIpercentEncoded} | [;:@&=])*
126
HTTPpath = ("/" {HTTPsegment})*
127
HTTPscheme = [hH][tT][tT][pP][sS]? "://"
128
HTTPurlFull = {HTTPscheme} {URIauthorityLoose} {HTTPpath}? {URIquery}? {URIfragment}?
129
// {HTTPurlNoScheme} excludes {URIlogin}, because it could otherwise accept e-mail addresses
130
HTTPurlNoScheme = {URIauthorityStrict} {HTTPpath}? {URIquery}? {URIfragment}?
131
HTTPurl = {HTTPurlFull} | {HTTPurlNoScheme}
133
FTPorFILEsegment = ({URIunreserved} | {URIpercentEncoded} | [?:@&=])*
134
FTPorFILEpath = "/" {FTPorFILEsegment} ("/" {FTPorFILEsegment})*
135
FTPtype = ";" [tT][yY][pP][eE] "=" [aAiIdD]
136
FTPscheme = [fF][tT][pP] "://"
137
FTPurl = {FTPscheme} {URIauthorityLoose} {FTPorFILEpath} {FTPtype}? {URIfragment}?
139
FILEscheme = [fF][iI][lL][eE] "://"
140
FILEurl = {FILEscheme} {URIhostLoose}? {FTPorFILEpath} {URIfragment}?
142
URL = {HTTPurl} | {FTPurl} | {FILEurl}
144
EMAILquotedString = [\"] ([\u0001-\u0008\u000B\u000C\u000E-\u0021\u0023-\u005B\u005D-\u007E] | [\\] [\u0000-\u007F])* [\"]
145
EMAILatomText = [A-Za-z0-9!#$%&'*+-/=?\^_`{|}~]
146
EMAILlabel = {EMAILatomText}+ | {EMAILquotedString}
147
EMAILlocalPart = {EMAILlabel} ("." {EMAILlabel})*
148
EMAILdomainLiteralText = [\u0001-\u0008\u000B\u000C\u000E-\u005A\u005E-\u007F] | [\\] [\u0000-\u007F]
149
// DFA minimization allows {IPv6Address} and {IPv4Address} to be included
150
// in the {EMAILbracketedHost} definition without incurring any size penalties,
151
// since {EMAILdomainLiteralText} recognizes all valid IP addresses.
152
// The IP address regexes are included in {EMAILbracketedHost} simply as a
153
// reminder that they are acceptable bracketed host forms.
154
EMAILbracketedHost = "[" ({EMAILdomainLiteralText}* | {IPv4Address} | [iI][pP][vV] "6:" {IPv6Address}) "]"
155
EMAIL = {EMAILlocalPart} "@" ({DomainNameStrict} | {EMAILbracketedHost})
159
/** Alphanumeric sequences */
160
public static final int WORD_TYPE = UAX29URLEmailTokenizer.ALPHANUM;
163
public static final int NUMERIC_TYPE = UAX29URLEmailTokenizer.NUM;
166
* Chars in class \p{Line_Break = Complex_Context} are from South East Asian
167
* scripts (Thai, Lao, Myanmar, Khmer, etc.). Sequences of these are kept
168
* together as as a single token rather than broken up, because the logic
169
* required to break them at word boundaries is too complex for UAX#29.
171
* See Unicode Line Breaking Algorithm: http://www.unicode.org/reports/tr14/#SA
173
public static final int SOUTH_EAST_ASIAN_TYPE = UAX29URLEmailTokenizer.SOUTHEAST_ASIAN;
175
public static final int IDEOGRAPHIC_TYPE = UAX29URLEmailTokenizer.IDEOGRAPHIC;
177
public static final int HIRAGANA_TYPE = UAX29URLEmailTokenizer.HIRAGANA;
179
public static final int KATAKANA_TYPE = UAX29URLEmailTokenizer.KATAKANA;
181
public static final int HANGUL_TYPE = UAX29URLEmailTokenizer.HANGUL;
183
public static final int EMAIL_TYPE = UAX29URLEmailTokenizer.EMAIL;
185
public static final int URL_TYPE = UAX29URLEmailTokenizer.URL;
187
public final int yychar()
193
* Fills CharTermAttribute with the current token text.
195
public final void getText(CharTermAttribute t) {
196
t.copyBuffer(zzBuffer, zzStartRead, zzMarkedPos-zzStartRead);
205
<<EOF>> { return StandardTokenizerInterface.YYEOF; }
207
{URL} { return URL_TYPE; }
208
{EMAIL} { return EMAIL_TYPE; }
210
// UAX#29 WB8. Numeric × Numeric
211
// WB11. Numeric (MidNum | MidNumLet) × Numeric
212
// WB12. Numeric × (MidNum | MidNumLet) Numeric
213
// WB13a. (ALetter | Numeric | Katakana | ExtendNumLet) × ExtendNumLet
214
// WB13b. ExtendNumLet × (ALetter | Numeric | Katakana)
216
{ExtendNumLetEx}* {NumericEx} ({ExtendNumLetEx}+ {NumericEx}
217
| {MidNumericEx} {NumericEx}
220
{ return NUMERIC_TYPE; }
222
// subset of the below for typing purposes only!
224
{ return HANGUL_TYPE; }
227
{ return KATAKANA_TYPE; }
229
// UAX#29 WB5. ALetter × ALetter
230
// WB6. ALetter × (MidLetter | MidNumLet) ALetter
231
// WB7. ALetter (MidLetter | MidNumLet) × ALetter
232
// WB9. ALetter × Numeric
233
// WB10. Numeric × ALetter
234
// WB13. Katakana × Katakana
235
// WB13a. (ALetter | Numeric | Katakana | ExtendNumLet) × ExtendNumLet
236
// WB13b. ExtendNumLet × (ALetter | Numeric | Katakana)
238
{ExtendNumLetEx}* ( {KatakanaEx} ({ExtendNumLetEx}* {KatakanaEx})*
239
| ( {NumericEx} ({ExtendNumLetEx}+ {NumericEx} | {MidNumericEx} {NumericEx} | {NumericEx})*
240
| {ALetterEx} ({ExtendNumLetEx}+ {ALetterEx} | {MidLetterEx} {ALetterEx} | {ALetterEx})* )+ )
241
({ExtendNumLetEx}+ ( {KatakanaEx} ({ExtendNumLetEx}* {KatakanaEx})*
242
| ( {NumericEx} ({ExtendNumLetEx}+ {NumericEx} | {MidNumericEx} {NumericEx} | {NumericEx})*
243
| {ALetterEx} ({ExtendNumLetEx}+ {ALetterEx} | {MidLetterEx} {ALetterEx} | {ALetterEx})* )+ ) )*
245
{ return WORD_TYPE; }
250
// [C]haracters with the Line_Break property values of Contingent_Break (CB),
251
// Complex_Context (SA/South East Asian), and XX (Unknown) are assigned word
252
// boundary property values based on criteria outside of the scope of this
253
// annex. That means that satisfactory treatment of languages like Chinese
254
// or Thai requires special handling.
256
// In Unicode 6.0, only one character has the \p{Line_Break = Contingent_Break}
257
// property: U+FFFC (  ) OBJECT REPLACEMENT CHARACTER.
259
// In the ICU implementation of UAX#29, \p{Line_Break = Complex_Context}
260
// character sequences (from South East Asian scripts like Thai, Myanmar, Khmer,
261
// Lao, etc.) are kept together. This grammar does the same below.
263
// See also the Unicode Line Breaking Algorithm:
265
// http://www.unicode.org/reports/tr14/#SA
267
{ComplexContext}+ { return SOUTH_EAST_ASIAN_TYPE; }
269
// UAX#29 WB14. Any ÷ Any
271
{HanEx} { return IDEOGRAPHIC_TYPE; }
272
{HiraganaEx} { return HIRAGANA_TYPE; }
275
// UAX#29 WB3. CR × LF
276
// WB3a. (Newline | CR | LF) ÷
277
// WB3b. ÷ (Newline | CR | LF)
280
[^] { /* Break so we don't hit fall-through warning: */ break;/* Not numeric, word, ideographic, hiragana, or SE Asian -- ignore it. */ }