Feat: Use tika to detect content type more accurately

jenkinsci · Jul 29, 2022 · e421cc7 · e421cc7
1 parent 1a4d466
commit e421cc7
Show file tree

Hide file tree

Showing 2 changed files with 14 additions and 0 deletions.
diff --git a/pom.xml b/pom.xml
@@ -47,6 +47,11 @@
       <artifactId>jaxb</artifactId>
       <version>2.3.6-1</version>
     </dependency>
+    <dependency>
+      <groupId>org.apache.tika</groupId>
+      <artifactId>tika-core</artifactId>
+      <version>1.18</version>
+    </dependency>
     <dependency>
       <groupId>org.apache.jclouds.provider</groupId>
       <artifactId>aws-s3</artifactId>

diff --git a/src/main/java/io/jenkins/plugins/artifact_manager_jclouds/JCloudsArtifactManager.java b/src/main/java/io/jenkins/plugins/artifact_manager_jclouds/JCloudsArtifactManager.java
@@ -61,6 +61,7 @@
 import jenkins.model.ArtifactManager;
 import jenkins.util.VirtualFile;
 import org.apache.http.client.methods.HttpGet;
+import org.apache.tika.Tika;
 import org.jclouds.blobstore.BlobStore;
 import org.jclouds.blobstore.BlobStoreContext;
 import org.jclouds.blobstore.BlobStores;
@@ -159,13 +160,21 @@ public Map<String, String> invoke(File f, VirtualChannel channel) {
                     if (contentType == null) {
                         contentType = URLConnection.guessContentTypeFromName(theFile.getName());
                     }
+                    if (contentType == null){
+                        contentType = detectByTika(theFile);
+                    }
                     contentTypes.put(relPath, contentType);
                 } catch (IOException e) {
                     Functions.printStackTrace(e, listener.error("Unable to determine content type for file: " + theFile));
                 }
             }
             return contentTypes;
         }
+
+        static String detectByTika(File f) throws IOException {
+            Tika tika = new Tika();
+            return tika.detect(f);
+        }
     }
 
     private static class UploadToBlobStorage extends MasterToSlaveFileCallable<Void> {