Disallow empty regex in PQ+

Change-Id: Ifcda331007866616e6bd079e8d331dec9d1f880a Reviewed-on: https://korap.ids-mannheim.de/gerrit/c/KorAP/Koral/+/7512 Reviewed-by: Nils Diewald <[email protected]>
KorAP · Sep 13, 2023 · 56dcb86 · 56dcb86
1 parent d964323
commit 56dcb86
Show file tree

Hide file tree

Showing 4 changed files with 7 additions and 4 deletions.
diff --git a/Changes b/Changes
@@ -1,3 +1,6 @@
+0.41 2023-09-13
+    - [bugfix] Disallow empty regex in PQ+ (diewald)
+
 0.40 2023-07-26
     - [feature] Initial support for CQP
       (irimia, margaretha, diewald)

diff --git a/pom.xml b/pom.xml
@@ -4,7 +4,7 @@
 
 	<groupId>de.ids_mannheim.korap</groupId>
 	<artifactId>Koral</artifactId>
-	<version>0.40.0</version>
+	<version>0.41.0</version>
 	<packaging>jar</packaging>
 	<name>Koral</name>
 	<url>https://korap.ids-mannheim.de</url>

diff --git a/src/main/antlr/poliqarpplus/Regex.g4 b/src/main/antlr/poliqarpplus/Regex.g4
@@ -15,4 +15,4 @@ fragment RE_chgroup  : '[' '^'? RE_char+ ']';
 fragment RE_quant    : ('.' | RE_char | RE_chgroup | ( '(' RE_expr ')')) ('?' | '*' | '+' | RE_focc) '?'?;
 fragment RE_group    : '(' RE_expr ')';
 fragment RE_expr     : ('.' | RE_char | RE_alter | RE_chgroup | RE_quant | RE_group | '^' )+;
-fragment RE_dquote   : ('"'|'„'|'“')  (RE_expr | '\'' | ':' )* ('"'|'“'|'”');
+fragment RE_dquote   : ('"'|'„'|'“')  (RE_expr | '\'' | ':' )+ ('"'|'“'|'”');
diff --git a/src/test/java/de/ids_mannheim/korap/query/test/poliqarpplus/PQGrammarTest.java b/src/test/java/de/ids_mannheim/korap/query/test/poliqarpplus/PQGrammarTest.java
@@ -59,7 +59,7 @@ public void dquoutes () {
 
         // how it behaves
       assertEquals(
-        "(request (query (segment (token (key (regex \"\"))))) <EOF>)",
+        "(request query <EOF>)",  // see different behaviour of " and '; for ", the query is parsed and an empty regex is generated
         treeString("\"\"\"")
         );
        // how it should behave
@@ -68,7 +68,7 @@ public void dquoutes () {
     	 	 treeString("\"\"\"")
     	 	 );
         assertEquals(
-        "(request (query (sequence (segment (token (key (regex \"\")))) (segment (token (key -key))))) <EOF>)",  // see different behaviour of " and '; for ", the query is parsed and an empty regex is generated
+        "(request (query (segment (token (key -key)))) <EOF>)",  // see different behaviour of " and '; for ", the query is parsed and an empty regex is generated
         treeString("\"\"-key\"")
         );
         assertEquals(