Improving tika PDF handling (#101486)

Correctly handling PDF files with public key encryption in attachment processor

docs/changelog/101486.yaml

@@ -0,0 +1,5 @@
+pr: 101486
+summary: Improving tika handling
+area: Ingest Node
+type: bug
+issues: []

modules/ingest-attachment/src/main/java/org/elasticsearch/ingest/attachment/TikaImpl.java

@@ -107,6 +107,15 @@ final class TikaImpl {
             } else {
                 throw new AssertionError(cause);
             }
+        } catch (LinkageError e) {
+            if (e.getMessage().contains("bouncycastle")) {
+                /*
+                 * Elasticsearch does not ship with bouncycastle. It is only used for public-key-encrypted PDFs, which this module does
+                 * not support anyway.
+                 */
+                throw new RuntimeException("document is encrypted", e);
+            }
+            throw new RuntimeException(e);
         }
     }
 

modules/ingest-attachment/src/test/java/org/elasticsearch/ingest/attachment/AttachmentProcessorTests.java

@@ -243,11 +243,25 @@ public class AttachmentProcessorTests extends ESTestCase {
         assertThat(attachmentData.get("content_length"), is(0L));
     }
 
-    public void testEncryptedPdf() throws Exception {
+    public void testEncryptedWithPasswordPdf() throws Exception {
+        /*
+         * This tests that a PDF that has been encrypted with a password fails in the way expected
+         */
         ElasticsearchParseException e = expectThrows(ElasticsearchParseException.class, () -> parseDocument("encrypted.pdf", processor));
         assertThat(e.getDetailedMessage(), containsString("document is encrypted"));
     }
 
+    public void testEncryptedWithKeyPdf() throws Exception {
+        /*
+         * This tests that a PDF that has been encrypted with a public key fails in the way expected
+         */
+        ElasticsearchParseException e = expectThrows(
+            ElasticsearchParseException.class,
+            () -> parseDocument("encrypted-with-key.pdf", processor)
+        );
+        assertThat(e.getDetailedMessage(), containsString("document is encrypted"));
+    }
+
     public void testHtmlDocument() throws Exception {
         Map<String, Object> attachmentData = parseDocument("htmlWithEmptyDateMeta.html", processor);