TIKA-2714 -- add detection for rar4 and rar5 files; throw an Unsuppor…

…tedFormatException for rar5 files
apache · Mar 11, 2020 · fec8db0 · fec8db0
1 parent 8c33080
commit fec8db0
Show file tree

Hide file tree

Showing 3 changed files with 21 additions and 2 deletions.
diff --git a/tika-core/src/main/resources/org/apache/tika/mime/tika-mimetypes.xml b/tika-core/src/main/resources/org/apache/tika/mime/tika-mimetypes.xml
@@ -4035,7 +4035,20 @@
     </magic>
     <glob pattern="*.rar"/>
   </mime-type>
-
+  <mime-type type="application/x-rar-compressed;version=4">
+    <_comment>RAR archive</_comment>
+    <magic priority="60">
+      <match value="\x52\x61\x72\x21\x1a\x07\x00" type="string" offset="0"/>
+    </magic>
+    <sub-class-of type="application/x-rar-compressed"/>
+  </mime-type>
+  <mime-type type="application/x-rar-compressed;version=5">
+    <_comment>RAR archive</_comment>
+    <magic priority="60">
+      <match value="\x52\x61\x72\x21\x1a\x07\x01\x00" type="string" offset="0"/>
+    </magic>
+    <sub-class-of type="application/x-rar-compressed"/>
+  </mime-type>
   <mime-type type="application/x-roxio-toast">
     <glob pattern="*.toast"/>
     <sub-class-of type="application/x-iso9660-image"/>

diff --git a/tika-parsers/src/main/java/org/apache/tika/parser/pkg/RarParser.java b/tika-parsers/src/main/java/org/apache/tika/parser/pkg/RarParser.java
@@ -27,11 +27,13 @@
 import com.github.junrar.rarfile.FileHeader;
 import org.apache.tika.exception.EncryptedDocumentException;
 import org.apache.tika.exception.TikaException;
+import org.apache.tika.exception.UnsupportedFormatException;
 import org.apache.tika.extractor.EmbeddedDocumentExtractor;
 import org.apache.tika.extractor.EmbeddedDocumentUtil;
 import org.apache.tika.io.TemporaryResources;
 import org.apache.tika.io.TikaInputStream;
 import org.apache.tika.metadata.Metadata;
+import org.apache.tika.metadata.TikaCoreProperties;
 import org.apache.tika.mime.MediaType;
 import org.apache.tika.parser.AbstractParser;
 import org.apache.tika.parser.ParseContext;
@@ -62,7 +64,11 @@ public void parse(InputStream stream, ContentHandler handler,
         xhtml.startDocument();
 
         EmbeddedDocumentExtractor extractor = EmbeddedDocumentUtil.getEmbeddedDocumentExtractor(context);
+        String mediaType = metadata.get(Metadata.CONTENT_TYPE);
 
+        if (mediaType != null && mediaType.contains("version=5")) {
+            throw new UnsupportedFormatException("Tika does not yet support rar version 5.");
+        }
         Archive rar = null;
         try (TemporaryResources tmp = new TemporaryResources()) {
             TikaInputStream tis = TikaInputStream.get(stream, tmp);

diff --git a/tika-parsers/src/test/java/org/apache/tika/parser/pkg/RarParserTest.java b/tika-parsers/src/test/java/org/apache/tika/parser/pkg/RarParserTest.java
@@ -48,7 +48,7 @@ public void testRarParsing() throws Exception {
             AUTO_DETECT_PARSER.parse(stream, handler, metadata, recursingContext);
         }
 
-        assertEquals("application/x-rar-compressed", metadata.get(Metadata.CONTENT_TYPE));
+        assertEquals("application/x-rar-compressed; version=4", metadata.get(Metadata.CONTENT_TYPE));
         String content = handler.toString();
         assertContains("test-documents/testEXCEL.xls", content);
         assertContains("Sample Excel Worksheet", content);