Merge pull request #168 from apache/XALANJ-2618

Improved encoding lookup
apache · Jan 27, 2024 · d83b90e · d83b90e
2 parents 289dd5f + 1aad64a
commit d83b90e
Show file tree

Hide file tree

Showing 2 changed files with 11 additions and 2 deletions.
diff --git a/serializer/src/main/java/org/apache/xml/serializer/Encodings.java b/serializer/src/main/java/org/apache/xml/serializer/Encodings.java
@@ -36,6 +36,12 @@
 /**
  * Provides information about encodings. Depends on the Java runtime
  * to provides writers for the different encodings.
+ *
+ * NOTE: When a Java or MIME names appears more than once in the
+ * encodings.properties table, priority is given to the first instance
+ * seen. For example, MIME name ISO-8859-1 will be mapped to Java name
+ * ISO8859-1, even though the file includes synonyms such as ISO8859_1
+ * and 8859-1.
  * <p>
  * This class is not a public API. It is only public because it
  * is used outside of this package.
@@ -369,9 +375,9 @@ private static EncodingInfo[] loadEncodingInfo()
                         mimeName = st.nextToken();
                         EncodingInfo ei = new EncodingInfo(mimeName, javaName, highChar);
                         encodingInfo_list.add(ei);
-                        _encodingTableKeyMime.put(mimeName.toUpperCase(), ei);
+                        _encodingTableKeyMime.putIfAbsent(mimeName.toUpperCase(), ei);
                         if (first)
-                            _encodingTableKeyJava.put(javaName.toUpperCase(), ei);
+                            _encodingTableKeyJava.putIfAbsent(javaName.toUpperCase(), ei);
                     }
                 }
             }

diff --git a/serializer/src/main/resources/org/apache/xml/serializer/Encodings.properties b/serializer/src/main/resources/org/apache/xml/serializer/Encodings.properties
@@ -37,6 +37,9 @@
 # Higher values above this char might be in the encoding, although in the 
 # case of this particular encoding there are no higher chars.
 #
+# NOTE: When a Java or MIME names appears more than once in this
+# table, priority is given to the first instance seen. For example,
+# MIME name ISO-8859-1 will be mapped to Java name ISO8859-1.
 #
 # <JAVA name encoding>, <PREFERRED name MIME>
 #