[TIKA-1800] Decode the escape characters in front of special characters

longphan98 · May 25, 2022 · 9150753 · 9150753
1 parent 13cf016
commit 9150753
Show file tree

Hide file tree

Showing 2 changed files with 25 additions and 3 deletions.
diff --git a/tika-core/src/main/java/org/apache/tika/mime/MediaType.java b/tika-core/src/main/java/org/apache/tika/mime/MediaType.java
@@ -39,15 +39,15 @@ public final class MediaType implements Comparable<MediaType>, Serializable {
      */
     private static final long serialVersionUID = -3831000556189036392L;
 
-    private static final Pattern SPECIAL = Pattern.compile("[\\(\\)<>@,;:\\\\\"/\\[\\]\\?=]");
+    private static final Pattern SPECIAL = Pattern.compile("[()<>@,;:\\\\\"/\\[\\]?=]");
 
     private static final Pattern SPECIAL_OR_WHITESPACE =
-            Pattern.compile("[\\(\\)<>@,;:\\\\\"/\\[\\]\\?=\\s]");
+            Pattern.compile("[()<>@,;:\\\\\"/\\[\\]?=\\s]");
 
     /**
      * See http://www.ietf.org/rfc/rfc2045.txt for valid mime-type characters.
      */
-    private static final String VALID_CHARS = "([^\\c\\(\\)<>@,;:\\\\\"/\\[\\]\\?=\\s]+)";
+    private static final String VALID_CHARS = "([^\\c\\()<>@,;:\\\\\"/\\[\\]?=\\s]+)";
 
     private static final Pattern TYPE_PATTERN =
             Pattern.compile("(?s)\\s*" + VALID_CHARS + "\\s*/\\s*" + VALID_CHARS + "\\s*($|;.*)");
@@ -331,6 +331,7 @@ private static Map<String, String> parseParameters(String string) {
     /**
      * Fuzzy unquoting mechanism that works also with somewhat malformed
      * quotes.
+     * TIKA-1800: get rid of the escape characters which are in front of special characters for further usage (.e.g to parse on to a new MediaType as parameters)
      *
      * @param s string to unquote
      * @return unquoted string
@@ -342,6 +343,14 @@ private static String unquote(String s) {
         while (s.endsWith("\"") || s.endsWith("'")) {
             s = s.substring(0, s.length() - 1);
         }
+        for (int i = 0; i < s.length() - 1; i++) {
+            if (s.charAt(i) == '\\' && !('0' <= s.charAt(i + 1) && s.charAt(i + 1) <= '9') &&
+                    !('a' <= s.charAt(i + 1) && s.charAt(i + 1) <= 'z') &&
+                    s.charAt(i + 1) != '-' && s.charAt(i + 1) != '+' &&
+                    s.charAt(i + 1) != '.' && s.charAt(i + 1) != '_') {
+                s = s.substring(0, i) + s.substring(i + 1);
+            }
+        }
         return s;
     }
 

diff --git a/tika-core/src/test/java/org/apache/tika/mime/MediaTypeTest.java b/tika-core/src/test/java/org/apache/tika/mime/MediaTypeTest.java
@@ -21,13 +21,26 @@
 import static org.junit.jupiter.api.Assertions.assertNotNull;
 import static org.junit.jupiter.api.Assertions.assertTrue;
 
+import java.io.IOException;
 import java.util.HashMap;
 import java.util.Map;
 
+import org.apache.tika.exception.TikaException;
 import org.junit.jupiter.api.Test;
+import org.xml.sax.SAXException;
 
 public class MediaTypeTest {
 
+    // TIKA-1800
+    @Test
+    public void testEscapedSpecialChar() {
+        MediaType mType = new MediaType(MediaType.APPLICATION_XML, "x-report", "#report?");
+        String cType = mType.toString();
+        mType = MediaType.parse(cType);
+        String report = mType.getParameters().get("x-report");
+        assertEquals("#report?", report);
+    }
+
     @Test
     public void testBasics() {
         assertEquals("application/octet-stream",