Use 'lower' for all ignoreCase operations.

To avoid developers having to maintain multiples indices, use 'lower' for all ignoreCase operations (JSqlParser, QueryByExample, Querydsl).

Resolves #2420.

src/main/java/org/springframework/data/jpa/repository/query/JSqlParserUtils.java

@@ -72,4 +72,20 @@ public static Function getJSqlLower(String column) {
 				.withName("lower") //
 				.withParameters(lowerParamExpression);
 	}
+
+	/**
+	 * Generates a upper function call, based on the {@code column}.
+	 *
+	 * @param column the non-empty column to use as param for upper
+	 * @return the generated upper function call
+	 */
+	public static Function getJSqlUpper(String column) {
+
+		List<Expression> expressions = Collections.singletonList(new Column(column));
+		ExpressionList upperParamExpression = new ExpressionList(expressions);
+
+		return new Function() //
+				.withName("upper") //
+				.withParameters(upperParamExpression);
+	}
 }

src/main/java/org/springframework/data/jpa/repository/query/QueryUtils.java

@@ -610,8 +610,8 @@ private static javax.persistence.criteria.Order toJpaOrder(Order order, From<?,
 		Expression<?> expression = toExpressionRecursively(from, property);
 
 		if (order.isIgnoreCase() && String.class.equals(expression.getJavaType())) {
-			Expression<String> lower = cb.lower((Expression<String>) expression);
-			return order.isAscending() ? cb.asc(lower) : cb.desc(lower);
+			Expression<String> upper = cb.lower((Expression<String>) expression);
+			return order.isAscending() ? cb.asc(upper) : cb.desc(upper);
 		} else {
 			return order.isAscending() ? cb.asc(expression) : cb.desc(expression);
 		}