Apache Solr open-source search software
LUCENE-9732: Hunspell: support dictionary entries starting with slash (#2301)
Peter Gromov825d8dbfd95c
diff --git a/lucene/analysis/common/src/java/org/apache/lucene/analysis/hunspell/Dictionary.java b/lucene/analysis/common/src/java/org/apache/lucene/analysis/hunspell/Dictionary.javaindex c4d902ef425..048f9c6ecaf 100644--- a/lucene/analysis/common/src/java/org/apache/lucene/analysis/hunspell/Dictionary.java+++ b/lucene/analysis/common/src/java/org/apache/lucene/analysis/hunspell/Dictionary.java@@ -832,7 +832,7 @@ public class Dictionary {if (ch == '\\' && i + 1 < entry.length()) {sb.append(entry.charAt(i + 1));i++;- } else if (ch == '/') {+ } else if (ch == '/' && i > 0) {sb.append(FLAG_SEPARATOR);} else if (!shouldSkipEscapedChar(ch)) {sb.append(ch);@@ -902,10 +902,7 @@ public class Dictionary {String line;while ((line = lines.readLine()) != null) {// wild and unpredictable code comment rules- if (line.isEmpty()- || line.charAt(0) == '/'- || line.charAt(0) == '#'- || line.charAt(0) == '\t') {+ if (line.isEmpty() || line.charAt(0) == '#' || line.charAt(0) == '\t') {continue;}line = unescapeEntry(line);diff --git a/lucene/analysis/common/src/java/org/apache/lucene/analysis/hunspell/WordCase.java b/lucene/analysis/common/src/java/org/apache/lucene/analysis/hunspell/WordCase.javaindex 1499ee46ca0..434eefe220e 100644--- a/lucene/analysis/common/src/java/org/apache/lucene/analysis/hunspell/WordCase.java+++ b/lucene/analysis/common/src/java/org/apache/lucene/analysis/hunspell/WordCase.java@@ -17,13 +17,23 @@package org.apache.lucene.analysis.hunspell;enum WordCase {+ /** e.g. WORD */UPPER,++ /** e.g. Word */TITLE,++ /** e.g. word */LOWER,- MIXED;++ /** e.g. WoRd or wOrd */+ MIXED,++ /** e.g "-" or "/" or "42" */+ NEUTRAL;static WordCase caseOf(char[] word, int length) {- boolean startsWithLower = Character.isLowerCase(word[0]);+ CharCase startCase = charCase(word[0]);boolean seenUpper = false;boolean seenLower = false;@@ -34,7 +44,7 @@ enum WordCase {if (seenUpper && seenLower) break;}- return get(startsWithLower, seenUpper, seenLower);+ return get(startCase, seenUpper, seenLower);}static WordCase caseOf(CharSequence word) {@@ -42,7 +52,7 @@ enum WordCase {}static WordCase caseOf(CharSequence word, int length) {- boolean startsWithLower = Character.isLowerCase(word.charAt(0));+ CharCase startCase = charCase(word.charAt(0));boolean seenUpper = false;boolean seenLower = false;@@ -53,14 +63,19 @@ enum WordCase {if (seenUpper && seenLower) break;}- return get(startsWithLower, seenUpper, seenLower);+ return get(startCase, seenUpper, seenLower);}- private static WordCase get(boolean startsWithLower, boolean seenUpper, boolean seenLower) {- if (!startsWithLower) {- return !seenLower ? UPPER : !seenUpper ? TITLE : MIXED;+ private static WordCase get(CharCase startCase, boolean seenUpper, boolean seenLower) {+ if (seenLower && seenUpper) return MIXED;+ switch (startCase) {+ case LOWER:+ return seenUpper ? MIXED : LOWER;+ case UPPER:+ return !seenLower ? UPPER : TITLE;+ default:+ return seenLower ? LOWER : seenUpper ? UPPER : NEUTRAL;}- return seenUpper ? MIXED : LOWER;}private static CharCase charCase(char c) {diff --git a/lucene/analysis/common/src/test/org/apache/lucene/analysis/hunspell/TestEscaped.java b/lucene/analysis/common/src/test/org/apache/lucene/analysis/hunspell/TestEscaped.javaindex 3038385665f..93b96cc9aa5 100644--- a/lucene/analysis/common/src/test/org/apache/lucene/analysis/hunspell/TestEscaped.java+++ b/lucene/analysis/common/src/test/org/apache/lucene/analysis/hunspell/TestEscaped.java@@ -25,11 +25,16 @@ public class TestEscaped extends StemmerTestBase {}public void testStemming() {+ assertStemsTo("/", "/");assertStemsTo("works", "work");assertStemsTo("work", "work");assertStemsTo("R2/D2", "R2/D2", "R2/d2");assertStemsTo("R2/D2s", "R2/D2");assertStemsTo("N/A", "N/A");assertStemsTo("N/As");++ assertStemsTo("/", "/");+ assertStemsTo("/a", "/a");+ assertStemsTo("//");}}diff --git a/lucene/analysis/common/src/test/org/apache/lucene/analysis/hunspell/escaped.dic b/lucene/analysis/common/src/test/org/apache/lucene/analysis/hunspell/escaped.dicindex 93602944260..ac4d4806069 100644--- a/lucene/analysis/common/src/test/org/apache/lucene/analysis/hunspell/escaped.dic+++ b/lucene/analysis/common/src/test/org/apache/lucene/analysis/hunspell/escaped.dic@@ -1,4 +1,6 @@-3+5+/+/awork/AR2\/D2/AN\/A