Merge pull request #303 from topikachu/feature/tika

Feat: Use tika to detect content type more accurately
jenkinsci · Aug 1, 2022 · 6c461c3 · 6c461c3
2 parents 282a772 + 9627797
commit 6c461c3
Show file tree

Hide file tree

Showing 4 changed files with 30 additions and 2 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
@@ -72,6 +72,8 @@
 import org.kohsuke.accmod.Restricted;
 import org.kohsuke.accmod.restrictions.NoExternalUse;
 
+import static io.jenkins.plugins.artifact_manager_jclouds.TikaUtil.detectByTika;
+
 /**
  * Jenkins artifact/stash implementation using any blob store supported by Apache jclouds.
  * To offer a new backend, implement {@link BlobStoreProvider}.
@@ -159,6 +161,9 @@ 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));

diff --git a/src/main/java/io/jenkins/plugins/artifact_manager_jclouds/TikaUtil.java b/src/main/java/io/jenkins/plugins/artifact_manager_jclouds/TikaUtil.java
@@ -0,0 +1,15 @@
+package io.jenkins.plugins.artifact_manager_jclouds;
+
+import org.apache.tika.Tika;
+
+import java.io.File;
+import java.io.IOException;
+
+public class TikaUtil {
+
+    private static Tika tika = new Tika();
+
+    static String detectByTika(File f) throws IOException {
+        return tika.detect(f);
+    }
+}
diff --git a/src/test/java/io/jenkins/plugins/artifact_manager_jclouds/s3/JCloudsArtifactManagerTest.java b/src/test/java/io/jenkins/plugins/artifact_manager_jclouds/s3/JCloudsArtifactManagerTest.java
@@ -283,13 +283,13 @@ public void nonAdmin() throws Exception {
     public void contentType() throws Exception {
         String text = "some regular text";
         String html = "<html><header></header><body>Test file contents</body></html>";
-
+        String json = "{\"key\":\"value\"}";
         ArtifactManagerConfiguration.get().getArtifactManagerFactories().add(getArtifactManagerFactory(null, null));
 
         j.createSlave("remote", null, null);
 
         WorkflowJob p = j.createProject(WorkflowJob.class, "p");
-        p.setDefinition(new CpsFlowDefinition("node('remote') {writeFile file: 'f.txt', text: '" + text + "'; writeFile file: 'f.html', text: '" + html + "'; writeFile file: 'f', text: '\\u0000'; archiveArtifacts 'f*'}", true));
+        p.setDefinition(new CpsFlowDefinition("node('remote') {writeFile file: 'f.txt', text: '" + text + "'; writeFile file: 'f.html', text: '" + html + "'; writeFile file: 'f', text: '\\u0000';writeFile file: 'f.json', text: '" + json +"'; archiveArtifacts 'f*'}", true));
         j.buildAndAssertSuccess(p);
 
         WebResponse response = j.createWebClient().goTo("job/p/1/artifact/f.txt", null).getWebResponse();
@@ -301,6 +301,9 @@ public void contentType() throws Exception {
         response = j.createWebClient().goTo("job/p/1/artifact/f", null).getWebResponse();
         assertThat(response.getContentLength(), equalTo(1L));
         assertThat(response.getContentType(), containsString("/octet-stream"));
+        response = j.createWebClient().goTo("job/p/1/artifact/f.json", null).getWebResponse();
+        assertThat(response.getContentAsString(), equalTo(json));
+        assertThat(response.getContentType(), equalTo("application/json"));
     }
 
     //@Test