Merge pull request #11269 from tdonohue/port_11268_to_7x

[Port dspace-7_x] fix(#11191): Align Content-Disposition with RFC 5987/6266

dspace-server-webapp/src/main/java/org/dspace/app/rest/utils/HttpHeadersInitializer.java

@@ -9,9 +9,11 @@ package org.dspace.app.rest.utils;
 
 import static java.util.Objects.isNull;
 import static java.util.Objects.nonNull;
-import static javax.mail.internet.MimeUtility.encodeText;
 
 import java.io.IOException;
+import java.net.URLEncoder;
+import java.nio.charset.StandardCharsets;
+import java.text.Normalizer;
 import java.util.Arrays;
 import java.util.Collections;
 import java.util.Objects;
@@ -171,9 +173,16 @@ public class HttpHeadersInitializer {
 
         // distposition may be null here if contentType is null
         if (!isNullOrEmpty(disposition)) {
-            httpHeaders.put(CONTENT_DISPOSITION, Collections.singletonList(String.format(CONTENT_DISPOSITION_FORMAT,
+            String fallbackAsciiName = createFallbackAsciiName(this.fileName);
+            String encodedUtf8Name = createEncodedUtf8Name(this.fileName);
+
+            String headerValue = String.format(
+                "%s; filename=\"%s\"; filename*=UTF-8''%s",
                 disposition,
-                                                                                         encodeText(fileName))));
+                fallbackAsciiName,
+                encodedUtf8Name
+            );
+            httpHeaders.put(CONTENT_DISPOSITION, Collections.singletonList(headerValue));
         }
         log.debug("Content-Disposition : {}", disposition);
 
@@ -261,4 +270,41 @@ public class HttpHeadersInitializer {
         return Arrays.binarySearch(matchValues, toMatch) > -1 || Arrays.binarySearch(matchValues, "*") > -1;
     }
 
+    /**
+     * Creates a safe ASCII-only fallback filename by removing diacritics (accents)
+     * and replacing any remaining non-ASCII characters.
+     * E.g., "ä-ö-é.pdf" becomes "a-o-e.pdf".
+     * @param originalFilename The original filename.
+     * @return A string containing only ASCII characters.
+     */
+    private String createFallbackAsciiName(String originalFilename) {
+        if (originalFilename == null) {
+            return "";
+        }
+        String normalized = Normalizer.normalize(originalFilename, Normalizer.Form.NFD);
+        String withoutAccents = normalized.replaceAll("\\p{InCombiningDiacriticalMarks}+", "");
+        return withoutAccents.replaceAll("[^\\x00-\\x7F]", "");
+    }
+
+    /**
+     * Creates a percent-encoded UTF-8 filename according to RFC 5987.
+     * This is for the `filename*` parameter.
+     * E.g., "ä ö é.pdf" becomes "%C3%A4%20%C3%B6%20%C3%A9.pdf".
+     * @param originalFilename The original filename.
+     * @return A percent-encoded string.
+     */
+    private String createEncodedUtf8Name(String originalFilename) {
+        if (originalFilename == null) {
+            return "";
+        }
+        try {
+            String encoded = URLEncoder.encode(originalFilename, StandardCharsets.UTF_8.toString());
+            return encoded.replace("+", "%20");
+        } catch (java.io.UnsupportedEncodingException e) {
+            // Fallback to a simple ASCII name if encoding fails.
+            log.error("UTF-8 encoding not supported, which should not happen.", e);
+            return createFallbackAsciiName(originalFilename);
+        }
+    }
+
 }

dspace-server-webapp/src/test/java/org/dspace/app/rest/BitstreamRestControllerIT.java

@@ -8,7 +8,6 @@
 package org.dspace.app.rest;
 
 import static java.util.UUID.randomUUID;
-import static javax.mail.internet.MimeUtility.encodeText;
 import static org.apache.commons.codec.CharEncoding.UTF_8;
 import static org.apache.commons.collections.CollectionUtils.isEmpty;
 import static org.apache.commons.io.IOUtils.toInputStream;
@@ -347,7 +346,11 @@ public class BitstreamRestControllerIT extends AbstractControllerIntegrationTest
         //2. A public item with a bitstream
 
         String bitstreamContent = "0123456789";
-        String bitstreamName = "ภาษาไทย";
+        String bitstreamName = "ภาษาไทย-com-acentuação.pdf";
+        String expectedAscii = "-com-acentuacao.pdf";
+        String expectedUtf8Encoded =
+        "%E0%B8%A0%E0%B8%B2%E0%B8%A9%E0%B8%B2%E0%B9%84%E0%B8%97%E0%B8%A2-"
+        + "com-acentua%C3%A7%C3%A3o.pdf";
 
         try (InputStream is = IOUtils.toInputStream(bitstreamContent, CharEncoding.UTF_8)) {
 
@@ -371,7 +374,9 @@ public class BitstreamRestControllerIT extends AbstractControllerIntegrationTest
             //We expect the content disposition to have the encoded bitstream name
             .andExpect(header().string(
                 "Content-Disposition",
-                "attachment;filename=\"" + encodeText(bitstreamName) + "\""
+                String.format("attachment; filename=\"%s\"; filename*=UTF-8''%s",
+                              expectedAscii,
+                              expectedUtf8Encoded)
             ));
     }