Add #12810: Implement escaping for keyword separators (FKA #12888)

CHANGELOG.md

@@ -11,6 +11,7 @@ Note that this project **does not** adhere to [Semantic Versioning](https://semv
 
 ### Added
 
+- We implemented escaping characters `\ ,` when parsing KeywordLists [#12810](https://github.com/JabRef/jabref/issues/12810)
 - We added a field for the citation count field on the General tab. [#13477](https://github.com/JabRef/jabref/issues/13477)
 - We added automatic lookup of DOI at citation relations [#13234](https://github.com/JabRef/jabref/issues/13234)
 - We added focus on the field Link in the "Add file link" dialog. [#13486](https://github.com/JabRef/jabref/issues/13486)

jablib/src/main/java/org/jabref/model/entry/KeywordList.java

@@ -8,7 +8,6 @@
 import java.util.List;
 import java.util.Objects;
 import java.util.Set;
-import java.util.StringTokenizer;
 import java.util.stream.Collectors;
 import java.util.stream.Stream;
 
@@ -52,13 +51,39 @@ public static KeywordList parse(String keywordString, Character delimiter, Chara
         Objects.requireNonNull(hierarchicalDelimiter);
 
         KeywordList keywordList = new KeywordList();
+        List<String> hierarchy = new ArrayList<>();
+        StringBuilder currentToken = new StringBuilder();
+        boolean isEscaping = false;
+
+        for (int i = 0; i < keywordString.length(); i++) {
+            char currentChar = keywordString.charAt(i);
+
+            if (isEscaping) {
+                currentToken.append(currentChar);
+                isEscaping = false;
+            } else if (currentChar == '\\') {
+                isEscaping = true;
+            } else if (currentChar == hierarchicalDelimiter) {
+                // Hierarchical delimiter reached: push current token as sublevel
+                hierarchy.add(currentToken.toString().trim());
+                currentToken.setLength(0);
+            } else if (currentChar == delimiter) {
+                // Keyword delimiter reached: push current token and then flush full hierarchy as keyword
+                hierarchy.add(currentToken.toString().trim());
+                currentToken.setLength(0);
+                keywordList.add(Keyword.of(hierarchy.toArray(new String[0])));
+                hierarchy.clear();
+            } else {
+                currentToken.append(currentChar);
+            }
+        }
 
-        StringTokenizer tok = new StringTokenizer(keywordString, delimiter.toString());
-        while (tok.hasMoreTokens()) {
-            String chain = tok.nextToken();
-            Keyword chainRoot = Keyword.of(chain.split(hierarchicalDelimiter.toString()));
-            keywordList.add(chainRoot);
+        // Handle the final token
+        if (!currentToken.isEmpty() || !hierarchy.isEmpty()) {
+            hierarchy.add(currentToken.toString().trim());
+            keywordList.add(Keyword.of(hierarchy.toArray(new String[0])));
         }
+
         return keywordList;
     }
 

jablib/src/test/java/org/jabref/model/entry/KeywordListTest.java

@@ -115,4 +115,41 @@ void mergeTwoDistinctKeywordsShouldReturnTheTwoKeywordsMerged() {
     void mergeTwoListsOfKeywordsShouldReturnTheKeywordsMerged() {
         assertEquals(new KeywordList("Figma", "Adobe", "JabRef", "Eclipse", "JetBrains"), KeywordList.merge("Figma, Adobe, JetBrains, Eclipse", "Adobe, JabRef", ','));
     }
+
+    @Test
+    void parseKeywordWithEscapedDelimiterDoesNotSplitKeyword() {
+        assertEquals(new KeywordList("keyword,one", "keywordTwo"),
+                KeywordList.parse("keyword\\,one, keywordTwo", ',', '>'));
+    }
+
+    @Test
+    void parseKeywordWithEscapedDelimiterAtEndPreservesDelimiter() {
+        assertEquals(new KeywordList("keywordOne,", "keywordTwo"),
+                KeywordList.parse("keywordOne\\,, keywordTwo", ',', '>'));
+    }
+
+    @Test
+    void parseKeywordWithEscapedBackslashTreatsItAsLiteral() {
+        assertEquals(new KeywordList("keyword\\", "keywordTwo"),
+                KeywordList.parse("keyword\\\\, keywordTwo", ',', '>'));
+    }
+
+    @Test
+    void parseKeywordWithEscapedDelimiterAndHierarchicalDelimiterCreatesHierarchy() {
+        Keyword expected = Keyword.of("keyword,one", "sub");
+        assertEquals(new KeywordList(expected),
+                KeywordList.parse("keyword\\,one > sub", ',', '>'));
+    }
+
+    @Test
+    void parseKeywordWithMultipleEscapedDelimitersTreatsThemAsLiteral() {
+        assertEquals(new KeywordList("one,two,three", "four"),
+                KeywordList.parse("one\\,two\\,three, four", ',', '>'));
+    }
+
+    @Test
+    void parseKeywordWithTrailingEscapeCharacterTreatsItAsLiteralBackslash() {
+        assertEquals(new KeywordList("keywordOne\\"),
+                KeywordList.parse("keywordOne\\\\", ',', '>'));
+    }
 }