LUCENE-9445 Add support for case insensitive regex searches in QueryParser

lucene/CHANGES.txt

@@ -61,8 +61,14 @@ API Changes
 * LUCENE-9462: Fields without positions should still return MatchIterator.
   (Alan Woodward, Dawid Weiss)
 
+* LUCENE-9445: QueryParserBase.getRegexpQuery and newRegexpQuery now take a 
+  caseSensitive flag to allow new case insensitive matching option. (Mark Harwood)
+
 Improvements
 
+* LUCENE-9445: QueryParser syntax for regular expressions extended to support 
+  trailing i for case insensitive matching e.g. /.*Foo/i (Mark Harwood)
+
 * LUCENE-9463: Query match region retrieval component, passage scoring and formatting
   for building custom highlighters. (Alan Woodward, Dawid Weiss)
 

lucene/queryparser/src/java/org/apache/lucene/queryparser/classic/MultiFieldQueryParser.java

@@ -261,16 +261,16 @@ protected Query getRangeQuery(String field, String part1, String part2, boolean
 
 
   @Override
-  protected Query getRegexpQuery(String field, String termStr)
+  protected Query getRegexpQuery(String field, String termStr, boolean caseSensitive)
       throws ParseException {
     if (field == null) {
       List<Query> clauses = new ArrayList<>();
       for (int i = 0; i < fields.length; i++) {
-        clauses.add(getRegexpQuery(fields[i], termStr));
+        clauses.add(getRegexpQuery(fields[i], termStr, caseSensitive));
       }
       return getMultiFieldQuery(clauses);
     }
-    return super.getRegexpQuery(field, termStr);
+    return super.getRegexpQuery(field, termStr, caseSensitive);
   }
 
   /** Creates a multifield query */

lucene/queryparser/src/java/org/apache/lucene/queryparser/classic/QueryParser.jj

@@ -209,7 +209,7 @@ PARSER_END(QueryParser)
 | <FUZZY_SLOP:    "~" ((<_NUM_CHAR>)+ (( "." (<_NUM_CHAR>)+ )? (<_TERM_CHAR>)*) | (<_TERM_CHAR>)*) >
 | <PREFIXTERM:    ("*") | ( <_TERM_START_CHAR> (<_TERM_CHAR>)* "*" ) >
 | <WILDTERM:      (<_TERM_START_CHAR> | [ "*", "?" ]) (<_TERM_CHAR> | ( [ "*", "?" ] ))* >
-| <REGEXPTERM:    "/" (~[ "/" ] | "\\/" )* "/" >
+| <REGEXPTERM:    "/" (~[ "/" ] | "\\/" )* ("/" | "/i") >
 | <RANGEIN_START: "[" > : Range
 | <RANGEEX_START: "{" > : Range
 }

lucene/queryparser/src/java/org/apache/lucene/queryparser/classic/QueryParserBase.java

@@ -572,10 +572,12 @@ protected Query newPrefixQuery(Term prefix){
   /**
    * Builds a new RegexpQuery instance
    * @param regexp Regexp term
+   * @param caseSensitive if the term matching should be case sensitive
    * @return new RegexpQuery instance
    */
-  protected Query newRegexpQuery(Term regexp) {
-    RegexpQuery query = new RegexpQuery(regexp, RegExp.ALL,
+  protected Query newRegexpQuery(Term regexp, boolean caseSensitive) {
+    int matchFlags = caseSensitive ? 0 : RegExp.ASCII_CASE_INSENSITIVE;
+    RegexpQuery query = new RegexpQuery(regexp, RegExp.ALL, matchFlags,
       maxDeterminizedStates);
     query.setRewriteMethod(multiTermRewriteMethod);
     return query;
@@ -746,18 +748,19 @@ private BytesRef analyzeWildcard(String field, String termStr) {
    *
    * @param field Name of the field query will use.
    * @param termStr Term token that contains a regular expression
+   * @param caseSensitive if token matching should be case sensitive
    *
    * @return Resulting {@link org.apache.lucene.search.Query} built for the term
    * @exception org.apache.lucene.queryparser.classic.ParseException throw in overridden method to disallow
    */
-  protected Query getRegexpQuery(String field, String termStr) throws ParseException
+  protected Query getRegexpQuery(String field, String termStr, boolean caseSensitive) throws ParseException
   {
     // We need to pass the whole string to #normalize, which will not work with
     // custom attribute factories for the binary term impl, and may not work
     // with some analyzers
     BytesRef term = getAnalyzer().normalize(field, termStr);
     Term t = new Term(field, term);
-    return newRegexpQuery(t);
+    return newRegexpQuery(t, caseSensitive);
   }
 
   /**
@@ -823,7 +826,9 @@ Query handleBareTokenQuery(String qfield, Token term, Token fuzzySlop, boolean p
           discardEscapeChar(term.image.substring
               (0, term.image.length()-1)));
     } else if (regexp) {
-      q = getRegexpQuery(qfield, term.image.substring(1, term.image.length()-1));
+      boolean caseSensitive = !term.image.endsWith("i");
+      int lastSlash = term.image.lastIndexOf("/");
+      q = getRegexpQuery(qfield, term.image.substring(1, lastSlash), caseSensitive);
     } else if (fuzzy) {
       q = handleBareFuzzy(qfield, fuzzySlop, termImage);
     } else {