diff --git a/lucene/CHANGES.txt b/lucene/CHANGES.txt index e6125cec330..477f2610752 100644 --- a/lucene/CHANGES.txt +++ b/lucene/CHANGES.txt @@ -123,6 +123,10 @@ Changes in Runtime Behavior: * LUCENE-7976: TieredMergePolicy now respects maxSegmentSizeMB by default when executing findForcedMerges and findForcedDeletesMerges (Erick Erickson) +Other: + +* LUCENE-8366: Upgrade to ICU 62.1. Emoji handling now uses Unicode 11's + Extended_Pictographic property. (Robert Muir) ======================= Lucene 7.4.1 ======================= diff --git a/lucene/analysis/common/src/java/org/apache/lucene/analysis/util/UnicodeProps.java b/lucene/analysis/common/src/java/org/apache/lucene/analysis/util/UnicodeProps.java index 86870d0becf..3e1006546bc 100644 --- a/lucene/analysis/common/src/java/org/apache/lucene/analysis/util/UnicodeProps.java +++ b/lucene/analysis/common/src/java/org/apache/lucene/analysis/util/UnicodeProps.java @@ -24,15 +24,15 @@ import org.apache.lucene.util.SparseFixedBitSet; /** * This file contains unicode properties used by various {@link CharTokenizer}s. - * The data was created using ICU4J v61.1.0.0 + * The data was created using ICU4J v62.1.0.0 *
- * Unicode version: 10.0.0.0 + * Unicode version: 11.0.0.0 */ public final class UnicodeProps { private UnicodeProps() {} /** Unicode version that was used to generate this file: {@value} */ - public static final String UNICODE_VERSION = "10.0.0.0"; + public static final String UNICODE_VERSION = "11.0.0.0"; /** Bitset with Unicode WHITESPACE code points. */ public static final Bits WHITESPACE = createBits( diff --git a/lucene/analysis/icu/src/data/uax29/Default.rbbi b/lucene/analysis/icu/src/data/uax29/Default.rbbi index afda68f47b5..b87fe7fcbf1 100644 --- a/lucene/analysis/icu/src/data/uax29/Default.rbbi +++ b/lucene/analysis/icu/src/data/uax29/Default.rbbi @@ -63,14 +63,10 @@ $MidNumLet = [\p{Word_Break = MidNumLet}]; $MidLetter = [\p{Word_Break = MidLetter}]; $MidNum = [\p{Word_Break = MidNum}]; $Numeric = [\p{Word_Break = Numeric}[[:Decomposition_Type=Wide:]&[:General_Category=Decimal_Number:]]]; -$ExtendNumLet = [\p{Word_Break = ExtendNumLet}]; -$E_Base = [\p{Word_Break = EB}]; -$E_Modifier = [\p{Word_Break = EM}]; -# Data for Extended Pictographic scraped from CLDR common/properties/ExtendedPictographic.txt, r13267 -$Extended_Pict = [\U0001F774-\U0001F77F\U00002700-\U00002701\U00002703-\U00002704\U0000270E\U00002710-\U00002711\U00002765-\U00002767\U0001F030-\U0001F093\U0001F094-\U0001F09F\U0001F10D-\U0001F10F\U0001F12F\U0001F16C-\U0001F16F\U0001F1AD-\U0001F1E5\U0001F260-\U0001F265\U0001F203-\U0001F20F\U0001F23C-\U0001F23F\U0001F249-\U0001F24F\U0001F252-\U0001F25F\U0001F266-\U0001F2FF\U0001F7D5-\U0001F7FF\U0001F000-\U0001F003\U0001F005-\U0001F02B\U0001F02C-\U0001F02F\U0001F322-\U0001F323\U0001F394-\U0001F395\U0001F398\U0001F39C-\U0001F39D\U0001F3F1-\U0001F3F2\U0001F3F6\U0001F4FE\U0001F53E-\U0001F548\U0001F54F\U0001F568-\U0001F56E\U0001F571-\U0001F572\U0001F57B-\U0001F586\U0001F588-\U0001F589\U0001F58E-\U0001F58F\U0001F591-\U0001F594\U0001F597-\U0001F5A3\U0001F5A6-\U0001F5A7\U0001F5A9-\U0001F5B0\U0001F5B3-\U0001F5BB\U0001F5BD-\U0001F5C1\U0001F5C5-\U0001F5D0\U0001F5D4-\U0001F5DB\U0001F5DF-\U0001F5E0\U0001F5E2\U0001F5E4-\U0001F5E7\U0001F5E9-\U0001F5EE\U0001F5F0-\U0001F5F2\U0001F5F4-\U0001F5F9\U00002605\U00002607-\U0000260D\U0000260F-\U00002610\U00002612\U00002616-\U00002617\U00002619-\U0000261C\U0000261E-\U0000261F\U00002621\U00002624-\U00002625\U00002627-\U00002629\U0000262B-\U0000262D\U00002630-\U00002637\U0000263B-\U00002647\U00002654-\U0000265F\U00002661-\U00002662\U00002664\U00002667\U00002669-\U0000267A\U0000267C-\U0000267E\U00002680-\U00002691\U00002695\U00002698\U0000269A\U0000269D-\U0000269F\U000026A2-\U000026A9\U000026AC-\U000026AF\U000026B2-\U000026BC\U000026BF-\U000026C3\U000026C6-\U000026C7\U000026C9-\U000026CD\U000026D0\U000026D2\U000026D5-\U000026E8\U000026EB-\U000026EF\U000026F6\U000026FB-\U000026FC\U000026FE-\U000026FF\U00002388\U0001FA00-\U0001FFFD\U0001F0A0-\U0001F0AE\U0001F0B1-\U0001F0BF\U0001F0C1-\U0001F0CF\U0001F0D1-\U0001F0F5\U0001F0AF-\U0001F0B0\U0001F0C0\U0001F0D0\U0001F0F6-\U0001F0FF\U0001F80C-\U0001F80F\U0001F848-\U0001F84F\U0001F85A-\U0001F85F\U0001F888-\U0001F88F\U0001F8AE-\U0001F8FF\U0001F900-\U0001F90B\U0001F91F\U0001F928-\U0001F92F\U0001F931-\U0001F932\U0001F94C\U0001F95F-\U0001F96B\U0001F992-\U0001F997\U0001F9D0-\U0001F9E6\U0001F90C-\U0001F90F\U0001F93F\U0001F94D-\U0001F94F\U0001F96C-\U0001F97F\U0001F998-\U0001F9BF\U0001F9C1-\U0001F9CF\U0001F9E7-\U0001F9FF\U0001F6C6-\U0001F6CA\U0001F6D3-\U0001F6D4\U0001F6E6-\U0001F6E8\U0001F6EA\U0001F6F1-\U0001F6F2\U0001F6F7-\U0001F6F8\U0001F6D5-\U0001F6DF\U0001F6ED-\U0001F6EF\U0001F6F9-\U0001F6FF]; -$EBG = [\p{Word_Break = EBG}]; -$EmojiNRK = [[\p{Emoji}] - [\p{Word_Break = Regional_Indicator}\u002a\u00230-9©®™〰〽]]; +$ExtendNumLet = [\p{Word_Break = ExtendNumLet}]; +$WSegSpace = [\p{Word_Break = WSegSpace}]; +$Extended_Pict = [:ExtPict:]; $Han = [:Han:]; $Hiragana = [:Hiragana:]; @@ -115,17 +111,17 @@ $IdeographicEx = $Ideographic ($Extend | $Format | $ZWJ)*; ## ------------------------------------------------- -!!forward; - - # Rule 3 - CR x LF # $CR $LF; # Rule 3c ZWJ x (Extended_Pict | EmojiNRK). Precedes WB4, so no intervening Extend chars allowed. # -$ZWJ ($Extended_Pict | $EmojiNRK); +$ZWJ $Extended_Pict; +# Rule 3d - Keep horizontal whitespace together. +# +$WSegSpace $WSegSpace; # Rule 4 - ignore Format and Extend characters, except when they appear at the beginning # of a region of Text. The rule here comes into play when the start of text @@ -142,8 +138,6 @@ $KatakanaEx {300}; # note: these status values override those from rule 5 $HiraganaEx {300}; # by virtue of being numerically larger. $IdeographicEx {400}; # -$E_Base ($Extend | $Format | $ZWJ)*; -$E_Modifier ($Extend | $Format | $ZWJ)*; $Extended_Pict ($Extend | $Format | $ZWJ)*; # @@ -193,11 +187,6 @@ $ExtendNumLetEx $Hebrew_Letter {200}; # (13b) $ExtendNumLetEx $NumericEx {100}; # (13b) $ExtendNumLetEx $KatakanaEx {300}; # (13b) -# rule 14 -# Do not break within emoji modifier sequences - -($E_Base | $EBG) ($Format | $Extend | $ZWJ)* $E_Modifier; - # rules 15 - 17 # Pairs of Regional Indicators stay together. # With rule chaining disabled by ^, this rule will match exactly two of them. @@ -211,24 +200,3 @@ $HangulSyllable $HangulSyllable {200}; # Rule 999 # Match a single code point if no other rule applies. .; - - -## ------------------------------------------------- - -!!safe_reverse; - -# rule 3 -($Extend | $Format | $ZWJ)+ .?; - -# rule 6 -($MidLetter | $MidNumLet | $Single_Quote) ($Format | $Extend | $ZWJ)* ($Hebrew_Letter | $ALetterPlus); - -# rule 7b -$Double_Quote ($Format | $Extend | $ZWJ)* $Hebrew_Letter; - - -# rule 11 -($MidNum | $MidNumLet | $Single_Quote) ($Format | $Extend | $ZWJ)* $Numeric; - -# rule 13c -$Regional_Indicator ($Format | $Extend | $ZWJ)* $Regional_Indicator; diff --git a/lucene/analysis/icu/src/data/utr30/DiacriticFolding.txt b/lucene/analysis/icu/src/data/utr30/DiacriticFolding.txt index 806a4f9baf8..ac6ac1eb05b 100644 --- a/lucene/analysis/icu/src/data/utr30/DiacriticFolding.txt +++ b/lucene/analysis/icu/src/data/utr30/DiacriticFolding.txt @@ -159,6 +159,8 @@ FF9E..FF9F> FFE3> 102E0> 10AE5..10AE6> +10D22..10D27> +10F46..10F50> 110B9..110BA> 11133..11134> 11173> @@ -177,12 +179,14 @@ FFE3> 1163F> 116B6..116B7> 1172B> +11839..1183A> 11A34> 11A47> 11A99> 11C3F> 11D42> 11D44..11D45> +11D97> 16AF0..16AF4> 16F8F..16F9F> 1D167..1D169> diff --git a/lucene/analysis/icu/src/data/utr30/NativeDigitFolding.txt b/lucene/analysis/icu/src/data/utr30/NativeDigitFolding.txt index 707674e299d..382a2795601 100644 --- a/lucene/analysis/icu/src/data/utr30/NativeDigitFolding.txt +++ b/lucene/analysis/icu/src/data/utr30/NativeDigitFolding.txt @@ -442,6 +442,16 @@ ABF9>0039 # MEETEI MAYEK DIGIT NINE 10A41>0032 # KHAROSHTHI DIGIT TWO 10A42>0033 # KHAROSHTHI DIGIT THREE 10A43>0034 # KHAROSHTHI DIGIT FOUR +10D30>0030 # HANIFI ROHINGYA DIGIT ZERO +10D31>0031 # HANIFI ROHINGYA DIGIT ONE +10D32>0032 # HANIFI ROHINGYA DIGIT TWO +10D33>0033 # HANIFI ROHINGYA DIGIT THREE +10D34>0034 # HANIFI ROHINGYA DIGIT FOUR +10D35>0035 # HANIFI ROHINGYA DIGIT FIVE +10D36>0036 # HANIFI ROHINGYA DIGIT SIX +10D37>0037 # HANIFI ROHINGYA DIGIT SEVEN +10D38>0038 # HANIFI ROHINGYA DIGIT EIGHT +10D39>0039 # HANIFI ROHINGYA DIGIT NINE 10E60>0031 # RUMI DIGIT ONE 10E61>0032 # RUMI DIGIT TWO 10E62>0033 # RUMI DIGIT THREE @@ -590,6 +600,16 @@ ABF9>0039 # MEETEI MAYEK DIGIT NINE 11D57>0037 # MASARAM GONDI DIGIT SEVEN 11D58>0038 # MASARAM GONDI DIGIT EIGHT 11D59>0039 # MASARAM GONDI DIGIT NINE +11DA0>0030 # GUNJALA GONDI DIGIT ZERO +11DA1>0031 # GUNJALA GONDI DIGIT ONE +11DA2>0032 # GUNJALA GONDI DIGIT TWO +11DA3>0033 # GUNJALA GONDI DIGIT THREE +11DA4>0034 # GUNJALA GONDI DIGIT FOUR +11DA5>0035 # GUNJALA GONDI DIGIT FIVE +11DA6>0036 # GUNJALA GONDI DIGIT SIX +11DA7>0037 # GUNJALA GONDI DIGIT SEVEN +11DA8>0038 # GUNJALA GONDI DIGIT EIGHT +11DA9>0039 # GUNJALA GONDI DIGIT NINE 16A60>0030 # MRO DIGIT ZERO 16A61>0031 # MRO DIGIT ONE 16A62>0032 # MRO DIGIT TWO diff --git a/lucene/analysis/icu/src/java/org/apache/lucene/analysis/icu/segmentation/BreakIteratorWrapper.java b/lucene/analysis/icu/src/java/org/apache/lucene/analysis/icu/segmentation/BreakIteratorWrapper.java index 9e5050d55b8..e21d2fdee92 100644 --- a/lucene/analysis/icu/src/java/org/apache/lucene/analysis/icu/segmentation/BreakIteratorWrapper.java +++ b/lucene/analysis/icu/src/java/org/apache/lucene/analysis/icu/segmentation/BreakIteratorWrapper.java @@ -16,8 +16,6 @@ */ package org.apache.lucene.analysis.icu.segmentation; -import com.ibm.icu.lang.UCharacter; -import com.ibm.icu.lang.UProperty; import com.ibm.icu.text.BreakIterator; import com.ibm.icu.text.RuleBasedBreakIterator; import com.ibm.icu.text.UTF16; @@ -65,18 +63,18 @@ final class BreakIteratorWrapper { } } - // See unicode doc L2/16-315 and also the RBBI rules for rationale. - // we don't include regional indicators here, because they aren't ambiguous for tagging, - // they need only be treated special for segmentation. + // See unicode doc L2/16-315 for rationale. + // basically for us the ambiguous cases (keycap/etc) as far as types go. static final UnicodeSet EMOJI_RK = new UnicodeSet("[\u002a\u00230-9©®™〰〽]").freeze(); + // faster than doing hasBinaryProperty() checks, at the cost of 1KB ram + static final UnicodeSet EMOJI = new UnicodeSet("[[:Emoji:][:Extended_Pictographic:]]").freeze(); /** Returns true if the current text represents emoji character or sequence */ private boolean isEmoji(int current, int next) { int begin = start + current; int end = start + next; int codepoint = UTF16.charAt(text, 0, end, begin); - // TODO: this can be made more aggressive and future-proof if it uses [:Extended_Pictographic:] - if (UCharacter.hasBinaryProperty(codepoint, UProperty.EMOJI)) { + if (EMOJI.contains(codepoint)) { if (EMOJI_RK.contains(codepoint)) { // if its in EmojiRK, we don't treat it as emoji unless there is evidence it forms emoji sequence, // an emoji presentation selector or keycap follows. diff --git a/lucene/analysis/icu/src/java/overview.html b/lucene/analysis/icu/src/java/overview.html index 6fa5821c242..6e0a5d70f58 100644 --- a/lucene/analysis/icu/src/java/overview.html +++ b/lucene/analysis/icu/src/java/overview.html @@ -353,7 +353,7 @@ and
This module exists to provide up-to-date Unicode functionality that supports
-the most recent version of Unicode (currently 10.0). However, some users who wish
+the most recent version of Unicode (currently 11.0). However, some users who wish
for stronger backwards compatibility can restrict
{@link org.apache.lucene.analysis.icu.ICUNormalizer2Filter} to operate on only
a specific Unicode Version by using a {@link com.ibm.icu.text.FilteredNormalizer2}.
diff --git a/lucene/analysis/icu/src/resources/org/apache/lucene/analysis/icu/segmentation/Default.brk b/lucene/analysis/icu/src/resources/org/apache/lucene/analysis/icu/segmentation/Default.brk
index e6c0ef770e7..9333a40bec9 100644
Binary files a/lucene/analysis/icu/src/resources/org/apache/lucene/analysis/icu/segmentation/Default.brk and b/lucene/analysis/icu/src/resources/org/apache/lucene/analysis/icu/segmentation/Default.brk differ
diff --git a/lucene/analysis/icu/src/resources/org/apache/lucene/analysis/icu/segmentation/MyanmarSyllable.brk b/lucene/analysis/icu/src/resources/org/apache/lucene/analysis/icu/segmentation/MyanmarSyllable.brk
index d6bfdb4ebf3..7a138834e91 100644
Binary files a/lucene/analysis/icu/src/resources/org/apache/lucene/analysis/icu/segmentation/MyanmarSyllable.brk and b/lucene/analysis/icu/src/resources/org/apache/lucene/analysis/icu/segmentation/MyanmarSyllable.brk differ
diff --git a/lucene/analysis/icu/src/resources/org/apache/lucene/analysis/icu/utr30.nrm b/lucene/analysis/icu/src/resources/org/apache/lucene/analysis/icu/utr30.nrm
index 1c3de121cad..4ded6c97762 100644
Binary files a/lucene/analysis/icu/src/resources/org/apache/lucene/analysis/icu/utr30.nrm and b/lucene/analysis/icu/src/resources/org/apache/lucene/analysis/icu/utr30.nrm differ
diff --git a/lucene/analysis/icu/src/test/org/apache/lucene/analysis/icu/segmentation/Latin-dont-break-on-hyphens.rbbi b/lucene/analysis/icu/src/test/org/apache/lucene/analysis/icu/segmentation/Latin-dont-break-on-hyphens.rbbi
index 0a4f0686a4f..caa784d1734 100644
--- a/lucene/analysis/icu/src/test/org/apache/lucene/analysis/icu/segmentation/Latin-dont-break-on-hyphens.rbbi
+++ b/lucene/analysis/icu/src/test/org/apache/lucene/analysis/icu/segmentation/Latin-dont-break-on-hyphens.rbbi
@@ -14,35 +14,71 @@
# See the License for the specific language governing permissions and
# limitations under the License.
#
-# Based on Default.rbbi, the default RBBI rules, based on UAX#29.
-# Added dashes to $MidLetter, so that words aren't broken on single dashes.
+# This test is a modified Default.rbbi that adds hyphens to MidLetter
#
+# Default.rbbi is from ICU (with some small modifications, to avoid CJK dictionary break,
+# and status code change related to that)
+#
+# Copyright (C) 2016 and later: Unicode, Inc. and others.
+# License & terms of use: http://www.unicode.org/copyright.html
+# Copyright (C) 2002-2016, International Business Machines Corporation
+# and others. All Rights Reserved.
+#
+# file: word.txt
+#
+# ICU Word Break Rules
+# See Unicode Standard Annex #29.
+# These rules are based on UAX #29 Revision 29 for Unicode Version 9.0
+# with additions for Emoji Sequences from https://goo.gl/cluFCn
+# Plus additional characters introduces with Emoji 5, http://www.unicode.org/reports/tr51/proposed.html
+#
+# Note: Updates to word.txt will usually need to be merged into
+# word_POSIX.txt also.
+
+##############################################################################
+#
+# Character class definitions from TR 29
+#
+##############################################################################
!!chain;
+!!quoted_literals_only;
+
#
# Character Class Definitions.
#
-$CR = [\p{Word_Break = CR}];
-$LF = [\p{Word_Break = LF}];
-$Newline = [\p{Word_Break = Newline}];
-$Extend = [\p{Word_Break = Extend}];
-$Format = [\p{Word_Break = Format}];
-$Katakana = [\p{Word_Break = Katakana}];
-$ALetter = [\p{Word_Break = ALetter}];
-$MidNumLet = [\p{Word_Break = MidNumLet}];
+$CR = [\p{Word_Break = CR}];
+$LF = [\p{Word_Break = LF}];
+$Newline = [\p{Word_Break = Newline} ];
+$Extend = [\p{Word_Break = Extend}];
+$ZWJ = [\p{Word_Break = ZWJ}];
+$Regional_Indicator = [\p{Word_Break = Regional_Indicator}];
+$Format = [\p{Word_Break = Format}];
+$Katakana = [\p{Word_Break = Katakana}];
+$Hebrew_Letter = [\p{Word_Break = Hebrew_Letter}];
+$ALetter = [\p{Word_Break = ALetter}];
+$Single_Quote = [\p{Word_Break = Single_Quote}];
+$Double_Quote = [\p{Word_Break = Double_Quote}];
+$MidNumLet = [\p{Word_Break = MidNumLet}];
# Don't use [:Dash:] here - it contains lots of chars that should continue to trigger word breaks
-$Dash = [\N{HYPHEN-MINUS}
- \N{HYPHEN}
- \N{EN DASH}
- \N{MINUS SIGN}
- \N{SMALL HYPHEN-MINUS}
- \N{FULLWIDTH HYPHEN-MINUS}];
-$MidLetter = [\p{Word_Break = MidLetter}$Dash]; # Don't break on (single) hyphen
-$MidNum = [\p{Word_Break = MidNum}];
-$Numeric = [\p{Word_Break = Numeric}[[:Decomposition_Type=Wide:]&[:General_Category=Decimal_Number:]]];
-$ExtendNumLet = [\p{Word_Break = ExtendNumLet}];
+$Dash = [\N{HYPHEN-MINUS}
+ \N{HYPHEN}
+ \N{EN DASH}
+ \N{MINUS SIGN}
+ \N{SMALL HYPHEN-MINUS}
+ \N{FULLWIDTH HYPHEN-MINUS}];
+$MidLetter = [\p{Word_Break = MidLetter}$Dash]; # Don't break on (single) hyphen
+$MidNum = [\p{Word_Break = MidNum}];
+$Numeric = [\p{Word_Break = Numeric}[[:Decomposition_Type=Wide:]&[:General_Category=Decimal_Number:]]];
+
+$ExtendNumLet = [\p{Word_Break = ExtendNumLet}];
+$WSegSpace = [\p{Word_Break = WSegSpace}];
+$Extended_Pict = [:ExtPict:];
+
+$Han = [:Han:];
+$Hiragana = [:Hiragana:];
# Dictionary character set, for triggering language-based break engines. Currently
@@ -50,58 +86,83 @@ $ExtendNumLet = [\p{Word_Break = ExtendNumLet}];
# 5.0 or later as the definition of Complex_Context was corrected to include all
# characters requiring dictionary break.
-$dictionary = [:LineBreak = Complex_Context:];
-$Control = [\p{Grapheme_Cluster_Break = Control}];
-$ALetterPlus = [$ALetter [$dictionary-$Extend-$Control]]; # Note: default ALetter does not
- # include the dictionary characters.
+$Control = [\p{Grapheme_Cluster_Break = Control}];
+$HangulSyllable = [\uac00-\ud7a3];
+$ComplexContext = [:LineBreak = Complex_Context:];
+$KanaKanji = [$Han $Hiragana $Katakana];
+$dictionaryCJK = [$Han $Hiragana $HangulSyllable];
+$dictionary = [$ComplexContext];
+
+# leave CJK scripts out of ALetterPlus
+$ALetterPlus = [$ALetter-$dictionaryCJK [$ComplexContext-$Extend-$Control]];
+
#
-# Rules 4 Ignore Format and Extend characters,
+# Rules 4 Ignore Format and Extend characters,
# except when they appear at the beginning of a region of text.
#
-$KatakanaEx = $Katakana ($Extend | $Format)*;
-$ALetterEx = $ALetterPlus ($Extend | $Format)*;
-$MidNumLetEx = $MidNumLet ($Extend | $Format)*;
-$MidLetterEx = $MidLetter ($Extend | $Format)*;
-$MidNumEx = $MidNum ($Extend | $Format)*;
-$NumericEx = $Numeric ($Extend | $Format)*;
-$ExtendNumLetEx = $ExtendNumLet ($Extend | $Format)*;
+# TODO: check if handling of katakana in dictionary makes rules incorrect/void
+$KatakanaEx = $Katakana ($Extend | $Format | $ZWJ)*;
+$Hebrew_LetterEx = $Hebrew_Letter ($Extend | $Format | $ZWJ)*;
+$ALetterEx = $ALetterPlus ($Extend | $Format | $ZWJ)*;
+$Single_QuoteEx = $Single_Quote ($Extend | $Format | $ZWJ)*;
+$Double_QuoteEx = $Double_Quote ($Extend | $Format | $ZWJ)*;
+$MidNumLetEx = $MidNumLet ($Extend | $Format | $ZWJ)*;
+$MidLetterEx = $MidLetter ($Extend | $Format | $ZWJ)*;
+$MidNumEx = $MidNum ($Extend | $Format | $ZWJ)*;
+$NumericEx = $Numeric ($Extend | $Format | $ZWJ)*;
+$ExtendNumLetEx = $ExtendNumLet ($Extend | $Format | $ZWJ)*;
+$Regional_IndicatorEx = $Regional_Indicator ($Extend | $Format | $ZWJ)*;
-$Hiragana = [\p{script=Hiragana}];
$Ideographic = [\p{Ideographic}];
-$HiraganaEx = $Hiragana ($Extend | $Format)*;
-$IdeographicEx = $Ideographic ($Extend | $Format)*;
+$HiraganaEx = $Hiragana ($Extend | $Format | $ZWJ)*;
+$IdeographicEx = $Ideographic ($Extend | $Format | $ZWJ)*;
## -------------------------------------------------
-!!forward;
-
-
# Rule 3 - CR x LF
#
$CR $LF;
+# Rule 3c ZWJ x (Extended_Pict | EmojiNRK). Precedes WB4, so no intervening Extend chars allowed.
+#
+$ZWJ $Extended_Pict;
+
+# Rule 3d - Keep horizontal whitespace together.
+#
+$WSegSpace $WSegSpace;
+
# Rule 4 - ignore Format and Extend characters, except when they appear at the beginning
# of a region of Text. The rule here comes into play when the start of text
# begins with a group of Format chars, or with a "word" consisting of a single
# char that is not in any of the listed word break categories followed by
-# format char(s).
-[^$CR $LF $Newline]? ($Extend | $Format)+;
+# format char(s), or is not a CJK dictionary character.
+[^$CR $LF $Newline]? ($Extend | $Format | $ZWJ)+;
$NumericEx {100};
$ALetterEx {200};
+$HangulSyllable {200};
+$Hebrew_LetterEx{200};
$KatakanaEx {300}; # note: these status values override those from rule 5
-$HiraganaEx {300}; # by virtual of being numerically larger.
+$HiraganaEx {300}; # by virtue of being numerically larger.
$IdeographicEx {400}; #
+$Extended_Pict ($Extend | $Format | $ZWJ)*;
+
#
# rule 5
# Do not break between most letters.
#
-$ALetterEx $ALetterEx {200};
+($ALetterEx | $Hebrew_LetterEx) ($ALetterEx | $Hebrew_LetterEx) {200};
# rule 6 and 7
-$ALetterEx ($MidLetterEx | $MidNumLetEx) $ALetterEx {200};
+($ALetterEx | $Hebrew_LetterEx) ($MidLetterEx | $MidNumLetEx | $Single_QuoteEx) ($ALetterEx | $Hebrew_LetterEx) {200};
+
+# rule 7a
+$Hebrew_LetterEx $Single_QuoteEx {200};
+
+# rule 7b and 7c
+$Hebrew_LetterEx $Double_QuoteEx $Hebrew_LetterEx {200};
# rule 8
@@ -109,27 +170,43 @@ $NumericEx $NumericEx {100};
# rule 9
-$ALetterEx $NumericEx {200};
+($ALetterEx | $Hebrew_LetterEx) $NumericEx {200};
# rule 10
-$NumericEx $ALetterEx {200};
+$NumericEx ($ALetterEx | $Hebrew_LetterEx) {200};
-# rule 11 and 12
+# rule 11 and 12
-$NumericEx ($MidNumEx | $MidNumLetEx) $NumericEx {100};
+$NumericEx ($MidNumEx | $MidNumLetEx | $Single_QuoteEx) $NumericEx {100};
# rule 13
-
$KatakanaEx $KatakanaEx {300};
# rule 13a/b
-$ALetterEx $ExtendNumLetEx {200}; # (13a)
-$NumericEx $ExtendNumLetEx {100}; # (13a)
-$KatakanaEx $ExtendNumLetEx {300}; # (13a)
-$ExtendNumLetEx $ExtendNumLetEx {200}; # (13a)
+$ALetterEx $ExtendNumLetEx {200}; # (13a)
+$Hebrew_LetterEx $ExtendNumLetEx {200}; # (13a)
+$NumericEx $ExtendNumLetEx {100}; # (13a)
+$KatakanaEx $ExtendNumLetEx {300}; # (13a)
+$ExtendNumLetEx $ExtendNumLetEx {200}; # (13a)
+
+$ExtendNumLetEx $ALetterEx {200}; # (13b)
+$ExtendNumLetEx $Hebrew_Letter {200}; # (13b)
+$ExtendNumLetEx $NumericEx {100}; # (13b)
+$ExtendNumLetEx $KatakanaEx {300}; # (13b)
+
+# rules 15 - 17
+# Pairs of Regional Indicators stay together.
+# With rule chaining disabled by ^, this rule will match exactly two of them.
+# No other rule begins with a Regional_Indicator, so chaining cannot extend the match.
+#
+^$Regional_IndicatorEx $Regional_IndicatorEx;
+
+# special handling for CJK characters: chain for later dictionary segmentation
+$HangulSyllable $HangulSyllable {200};
+
+# Rule 999
+# Match a single code point if no other rule applies.
+.;
-$ExtendNumLetEx $ALetterEx {200}; # (13b)
-$ExtendNumLetEx $NumericEx {100}; # (13b)
-$ExtendNumLetEx $KatakanaEx {300}; # (13b)
diff --git a/lucene/analysis/icu/src/test/org/apache/lucene/analysis/icu/segmentation/TestICUTokenizer.java b/lucene/analysis/icu/src/test/org/apache/lucene/analysis/icu/segmentation/TestICUTokenizer.java
index 98939752cbe..91c3d929d89 100644
--- a/lucene/analysis/icu/src/test/org/apache/lucene/analysis/icu/segmentation/TestICUTokenizer.java
+++ b/lucene/analysis/icu/src/test/org/apache/lucene/analysis/icu/segmentation/TestICUTokenizer.java
@@ -23,6 +23,7 @@ import org.apache.lucene.analysis.Tokenizer;
import org.apache.lucene.analysis.icu.tokenattributes.ScriptAttribute;
import com.ibm.icu.lang.UScript;
+import com.ibm.icu.text.UnicodeSet;
import java.io.IOException;
import java.io.StringReader;
@@ -316,6 +317,20 @@ public class TestICUTokenizer extends BaseTokenStreamTestCase {
new String[] { "