assert current /bag-info.txt behavior #8760

Also, add a superuser-only API for downloading files
(such as bags) from the file system so we can make
assertions about them in our tests.

src/main/java/edu/harvard/iq/dataverse/api/AbstractApiBean.java

@@ -45,11 +45,13 @@
 import edu.harvard.iq.dataverse.search.savedsearch.SavedSearchServiceBean;
 import edu.harvard.iq.dataverse.settings.SettingsServiceBean;
 import edu.harvard.iq.dataverse.util.BundleUtil;
+import edu.harvard.iq.dataverse.util.FileUtil;
 import edu.harvard.iq.dataverse.util.SystemConfig;
 import edu.harvard.iq.dataverse.util.json.JsonParser;
 import edu.harvard.iq.dataverse.util.json.JsonUtil;
 import edu.harvard.iq.dataverse.util.json.NullSafeJsonBuilder;
 import edu.harvard.iq.dataverse.validation.PasswordValidatorServiceBean;
+import java.io.InputStream;
 import java.net.URI;
 import java.util.Arrays;
 import java.util.Collections;
@@ -726,6 +728,11 @@ protected Response ok(String data, MediaType mediaType, String downloadFilename)
         return res.build();
     }
 
+    protected Response ok(InputStream inputStream) {
+        ResponseBuilder res = Response.ok().entity(inputStream).type(MediaType.valueOf(FileUtil.MIME_TYPE_UNDETERMINED_DEFAULT));
+        return res.build();
+    }
+
     protected Response created( String uri, JsonObjectBuilder bld ) {
         return Response.created( URI.create(uri) )
                 .entity( Json.createObjectBuilder()

src/main/java/edu/harvard/iq/dataverse/api/Admin.java

@@ -107,6 +107,7 @@
 import edu.harvard.iq.dataverse.util.SystemConfig;
 import edu.harvard.iq.dataverse.util.UrlSignerUtil;
 
+import java.io.FileInputStream;
 import java.io.IOException;
 import java.io.OutputStream;
 
@@ -2425,5 +2426,27 @@ public Response getSignedUrl(@Context ContainerRequestContext crc, JsonObject ur
 
         return ok(Json.createObjectBuilder().add(ExternalToolHandler.SIGNED_URL, signedUrl));
     }
-
+
+    /**
+     * For testing only. Download a file from the file system.
+     */
+    @GET
+    @AuthRequired
+    @Path("/localfile")
+    public Response getLocalFile(@Context ContainerRequestContext crc, @QueryParam("pathToFile") String pathToFile) {
+        try {
+            AuthenticatedUser user = getRequestAuthenticatedUserOrDie(crc);
+            if (!user.isSuperuser()) {
+                return error(Response.Status.FORBIDDEN, "Superusers only.");
+            }
+        } catch (WrappedResponse wr) {
+            return wr.getResponse();
+        }
+        try {
+            return ok(new FileInputStream(pathToFile));
+        } catch (IOException ex) {
+            return error(Status.BAD_REQUEST, ex.toString());
+        }
+    }
+
 }

src/test/java/edu/harvard/iq/dataverse/api/BagIT.java

@@ -1,17 +1,32 @@
 package edu.harvard.iq.dataverse.api;
 
-import io.restassured.RestAssured;
-import io.restassured.response.Response;
 import edu.harvard.iq.dataverse.engine.command.impl.LocalSubmitToArchiveCommand;
 import edu.harvard.iq.dataverse.settings.SettingsServiceBean;
+import io.restassured.RestAssured;
+import static io.restassured.RestAssured.given;
+import io.restassured.response.Response;
 import static jakarta.ws.rs.core.Response.Status.CREATED;
 import static jakarta.ws.rs.core.Response.Status.OK;
+import java.io.IOException;
+import java.io.InputStream;
+import java.nio.file.Path;
+import java.nio.file.Paths;
+import java.nio.file.StandardCopyOption;
+import java.util.Enumeration;
+import java.util.Scanner;
+import java.util.logging.Level;
+import java.util.logging.Logger;
+import java.util.zip.ZipEntry;
+import java.util.zip.ZipFile;
 import org.junit.jupiter.api.AfterAll;
+import static org.junit.jupiter.api.Assertions.assertEquals;
 import org.junit.jupiter.api.BeforeAll;
 import org.junit.jupiter.api.Test;
 
 public class BagIT {
 
+    static String bagitExportDir = "/tmp";
+
     @BeforeAll
     public static void setUpClass() {
 
@@ -25,14 +40,14 @@ public static void setUpClass() {
         setArchiverSettings.then().assertThat()
                 .statusCode(OK.getStatusCode());
 
-        Response setBagItLocalPath = UtilIT.setSetting(":BagItLocalPath", "/tmp");
+        Response setBagItLocalPath = UtilIT.setSetting(":BagItLocalPath", bagitExportDir);
         setBagItLocalPath.then().assertThat()
                 .statusCode(OK.getStatusCode());
 
     }
 
     @Test
-    public void testBagItExport() {
+    public void testBagItExport() throws IOException {
 
         Response createUser = UtilIT.createRandomUser();
         createUser.then().assertThat().statusCode(OK.getStatusCode());
@@ -63,6 +78,78 @@ public void testBagItExport() {
         archiveDataset.prettyPrint();
         archiveDataset.then().assertThat().statusCode(OK.getStatusCode());
 
+        // spaceName comes from LocalSubmitToArchiveCommand
+        String spaceName = datasetPid.replace(':', '-').replace('/', '-')
+                .replace('.', '-').toLowerCase();
+        // spacename: doi-10-5072-fk2-fosg5q
+
+        String pathToZip = bagitExportDir + "/" + spaceName + "v1.0" + ".zip";
+
+        try {
+            // give the bag time to generate
+            Thread.sleep(3000);
+        } catch (InterruptedException ex) {
+        }
+
+        // A bag could look like this:
+        //doi-10-5072-FK2-DKUTDUv-1-0/data/
+        //doi-10-5072-FK2-DKUTDUv-1-0/data/Darwin's Finches/
+        //doi-10-5072-FK2-DKUTDUv-1-0/metadata/
+        //doi-10-5072-FK2-DKUTDUv-1-0/metadata/pid-mapping.txt
+        //doi-10-5072-FK2-DKUTDUv-1-0/manifest-md5.txt
+        //doi-10-5072-FK2-DKUTDUv-1-0/bagit.txt
+        //doi-10-5072-FK2-DKUTDUv-1-0/metadata/oai-ore.jsonld
+        //doi-10-5072-FK2-DKUTDUv-1-0/metadata/datacite.xml
+        //doi-10-5072-FK2-DKUTDUv-1-0/bag-info.txt
+        // ---
+        // bag-info.txt could look like this:
+        //Contact-Name: Finch, Fiona
+        //Contact-Email: finch@mailinator.com
+        //Source-Organization: Dataverse Installation (<Site Url>)
+        //Organization-Address: <Full address>
+        //Organization-Email: <Email address>
+        //External-Description: Darwin's finches (also known as the Galápagos finches) are a group of about
+        // fifteen species of passerine birds.
+        //Bagging-Date: 2023-11-14
+        //External-Identifier: https://doi.org/10.5072/FK2/LZIGBC
+        //Bag-Size: 0 bytes
+        //Payload-Oxum: 0.0
+        //Internal-Sender-Identifier: Root:Darwin's Finches
+        Response downloadBag = downloadLocalFile(pathToZip, apiToken);
+        downloadBag.then().assertThat().statusCode(OK.getStatusCode());
+        Path outputPath = Paths.get("/tmp/foo.zip");
+        java.nio.file.Files.copy(downloadBag.getBody().asInputStream(), outputPath, StandardCopyOption.REPLACE_EXISTING);
+
+        ZipFile zipFile = new ZipFile(outputPath.toString());
+        Enumeration<? extends ZipEntry> entries = zipFile.entries();
+        String sourceOrg = null;
+        String orgAddress = null;
+        String orgEmail = null;
+        while (entries.hasMoreElements()) {
+            ZipEntry entry = entries.nextElement();
+            String name = entry.getName();
+            System.out.println("name: " + name);
+            if (name.endsWith("bag-info.txt")) {
+                InputStream stream = zipFile.getInputStream(entry);
+                Scanner s = new Scanner(stream).useDelimiter("\\A");
+                String result = s.hasNext() ? s.next() : "";
+                System.out.println("result: " + result);
+                String[] lines = result.split("\n");
+                for (String line : lines) {
+                    if (line.startsWith("Source-Organization")) {
+                        sourceOrg = line;
+                    } else if (line.startsWith("Organization-Address")) {
+                        orgAddress = line;
+                    } else if (line.startsWith("Organization-Email")) {
+                        orgEmail = line;
+                    } else {
+                    }
+                }
+            }
+        }
+        assertEquals("Source-Organization: Dataverse Installation (<Site Url>)", sourceOrg.trim());
+        assertEquals("Organization-Address: <Full address>", orgAddress.trim());
+        assertEquals("Organization-Email: <Email address>", orgEmail.trim());
     }
 
     @AfterAll
@@ -75,4 +162,10 @@ public static void tearDownClass() {
 
     }
 
+    static Response downloadLocalFile(String pathToFile, String apiToken) {
+        return given()
+                .header("X-Dataverse-key", apiToken)
+                .get("/api/admin/localfile?pathToFile=" + pathToFile);
+    }
+
 }