feat(storage): add datasets resource-type prefix to dataset logical paths

common/workflow-core/src/main/scala/org/apache/texera/amber/core/storage/FileResolver.scala

@@ -75,9 +75,13 @@ object FileResolver {
     filePath.toUri
   }
 
+  private val RESOURCE_TYPE_PREFIXES = Set("datasets")
+
   /**
     * Parses a dataset file path and extracts its components.
-    * Expected format: /ownerEmail/datasetName/versionName/fileRelativePath
+    * Expected format: /datasets/ownerEmail/datasetName/versionName/fileRelativePath
+    *
+    * The first segment is a resource type prefix (e.g. "datasets") and is stripped before parsing.
     *
     * @param fileName The file path to parse
     * @return Some((ownerEmail, datasetName, versionName, fileRelativePath)) if valid, None otherwise
@@ -86,7 +90,12 @@ object FileResolver {
       fileName: String
   ): Option[(String, String, String, Array[String])] = {
     val filePath = Paths.get(fileName)
-    val pathSegments = (0 until filePath.getNameCount).map(filePath.getName(_).toString).toArray
+    var pathSegments = (0 until filePath.getNameCount).map(filePath.getName(_).toString).toArray
+
+    // Strip known resource type prefix if present
+    if (pathSegments.nonEmpty && RESOURCE_TYPE_PREFIXES.contains(pathSegments(0))) {
+      pathSegments = pathSegments.drop(1)
+    }
 
     if (pathSegments.length < 4) {
       return None
@@ -103,8 +112,8 @@ object FileResolver {
   /**
     * Attempts to resolve a given fileName to a URI.
     *
-    * The fileName format should be: /ownerEmail/datasetName/versionName/fileRelativePath
-    *   e.g. /bob@texera.com/twitterDataset/v1/california/irvine/tw1.csv
+    * The fileName format should be: /datasets/ownerEmail/datasetName/versionName/fileRelativePath
+    *   e.g. /datasets/bob@texera.com/twitterDataset/v1/california/irvine/tw1.csv
     * The output dataset URI format is: {DATASET_FILE_URI_SCHEME}:///{repositoryName}/{versionHash}/fileRelativePath
     *   e.g. {DATASET_FILE_URI_SCHEME}:///dataset-15/adeq233td/some/dir/file.txt
     *
@@ -195,7 +204,7 @@ object FileResolver {
 
   /**
     * Parses a dataset file path to extract owner email and dataset name.
-    * Expected format: /ownerEmail/datasetName/versionName/fileRelativePath
+    * Expected format: /datasets/ownerEmail/datasetName/versionName/fileRelativePath
     *
     * @param path The file path from operator properties
     * @return Some((ownerEmail, datasetName)) if path is valid, None otherwise

common/workflow-core/src/test/scala/org/apache/texera/amber/storage/FileResolverSpec.scala

@@ -79,9 +79,9 @@ class FileResolverSpec
 
   private val localCsvFilePath = "common/workflow-core/src/test/resources/country_sales_small.csv"
 
-  private val datasetACsvFilePath = "/test_user@test.com/test_dataset/v2/directory/a.csv"
+  private val datasetACsvFilePath = "/datasets/test_user@test.com/test_dataset/v2/directory/a.csv"
 
-  private val dataset1TxtFilePath = "/test_user@test.com/test_dataset/v1/1.txt"
+  private val dataset1TxtFilePath = "/datasets/test_user@test.com/test_dataset/v1/1.txt"
 
   override protected def beforeAll(): Unit = {
     initializeDBAndReplaceDSLContext()

file-service/src/main/scala/org/apache/texera/service/resource/DatasetResource.scala

@@ -1188,7 +1188,7 @@ class DatasetResource extends LazyLogging {
         throw new NotFoundException(ERR_DATASET_VERSION_NOT_FOUND_MESSAGE)
       )
 
-      val ownerNode = DatasetFileNode
+      val datasetsNode = DatasetFileNode
         .fromLakeFSRepositoryCommittedObjects(
           Map(
             (user.getEmail, dataset.getName, latestVersion.getName) -> LakeFSStorageClient
@@ -1197,6 +1197,8 @@ class DatasetResource extends LazyLogging {
         )
         .head
 
+      val ownerNode = datasetsNode.children.get.head
+
       DashboardDatasetVersion(
         latestVersion,
         ownerNode.children.get
@@ -1404,7 +1406,7 @@ class DatasetResource extends LazyLogging {
     val datasetName = dataset.dataset.getName
     val repositoryName = dataset.dataset.getRepositoryName
 
-    val ownerFileNode = DatasetFileNode
+    val datasetsNode = DatasetFileNode
       .fromLakeFSRepositoryCommittedObjects(
         Map(
           (dataset.ownerEmail, datasetName, datasetVersion.getName) -> LakeFSStorageClient
@@ -1413,6 +1415,8 @@ class DatasetResource extends LazyLogging {
       )
       .head
 
+    val ownerFileNode = datasetsNode.children.get.head
+
     DatasetVersionRootFileNodesResponse(
       ownerFileNode.children.get
         .find(_.getName == datasetName)
@@ -1423,7 +1427,7 @@ class DatasetResource extends LazyLogging {
         .head
         .children
         .get,
-      DatasetFileNode.calculateTotalSize(List(ownerFileNode))
+      DatasetFileNode.calculateTotalSize(List(datasetsNode))
     )
   }
 

file-service/src/main/scala/org/apache/texera/service/type/dataset/DatasetFileNode.scala

@@ -81,15 +81,19 @@ object DatasetFileNode {
   ): List[DatasetFileNode] = {
     val rootNode = new DatasetFileNode("/", "directory", null, "")
 
+    // Add "datasets" prefix node
+    val datasetsNode = new DatasetFileNode("datasets", "directory", rootNode, "")
+    rootNode.children = Some(List(datasetsNode))
+
     // Owner level nodes map
     val ownerNodes = mutable.Map[String, DatasetFileNode]()
 
     map.foreach {
       case ((ownerEmail, datasetName, versionName), objects) =>
         val ownerNode = ownerNodes.getOrElseUpdate(
           ownerEmail, {
-            val newNode = new DatasetFileNode(ownerEmail, "directory", rootNode, ownerEmail)
-            rootNode.children = Some(rootNode.getChildren :+ newNode)
+            val newNode = new DatasetFileNode(ownerEmail, "directory", datasetsNode, ownerEmail)
+            datasetsNode.children = Some(datasetsNode.getChildren :+ newNode)
             newNode
           }
         )

frontend/src/app/common/type/dataset-file.ts

@@ -17,8 +17,8 @@
  * under the License.
  */
 
-// user given filePath is /ownerEmail/datasetName/versionName/fileRelativePath
-// e.g. /bob@texera.com/twitterDataset/v1/california/irvine/tw1.csv
+// user given filePath is /datasets/ownerEmail/datasetName/versionName/fileRelativePath
+// e.g. /datasets/bob@texera.com/twitterDataset/v1/california/irvine/tw1.csv
 export interface DatasetFile {
   ownerEmail: string;
   datasetName: string;
@@ -28,11 +28,17 @@ export interface DatasetFile {
 
 /**
  * Parses a file path string to a DatasetFile interface.
+ * The first segment "datasets" is stripped before parsing.
  * @param filePath - The file path string to parse.
  * @returns The parsed DatasetFile object.
  */
 export function parseFilePathToDatasetFile(filePath: string): DatasetFile {
-  const parts = filePath.split("/").filter(part => part.length > 0);
+  let parts = filePath.split("/").filter(part => part.length > 0);
+
+  // Strip the "datasets" prefix if present
+  if (parts.length > 0 && parts[0] === "datasets") {
+    parts = parts.slice(1);
+  }
 
   if (parts.length < 4) {
     throw new Error("Invalid file path format");
@@ -56,5 +62,5 @@ export function parseFilePathToDatasetFile(filePath: string): DatasetFile {
  */
 export function parseDatasetFileToFilePath(datasetFile: DatasetFile): string {
   const { ownerEmail, datasetName, versionName, fileRelativePath } = datasetFile;
-  return `/${ownerEmail}/${datasetName}/${versionName}/${fileRelativePath}`;
+  return `/datasets/${ownerEmail}/${datasetName}/${versionName}/${fileRelativePath}`;
 }

frontend/src/app/common/type/datasetVersionFileTree.ts

@@ -31,19 +31,20 @@ export function getFullPathFromDatasetFileNode(node: DatasetFileNode): string {
 }
 
 /**
- * Returns the relative path of a DatasetFileNode by stripping the first three segments.
+ * Returns the relative path of a DatasetFileNode by stripping the first four segments
+ * (datasets/ownerEmail/datasetName/versionName).
  * @param node The DatasetFileNode whose relative path is needed.
- * @returns The relative path (without the first three segments and without a leading slash).
+ * @returns The relative path (without the first four segments and without a leading slash).
  */
 export function getRelativePathFromDatasetFileNode(node: DatasetFileNode): string {
   const fullPath = getFullPathFromDatasetFileNode(node); // Get the full path
   const pathSegments = fullPath.split("/").filter(segment => segment.length > 0); // Split and remove empty segments
 
-  if (pathSegments.length <= 3) {
-    return ""; // If there are 3 or fewer segments, return an empty string (no relative path exists)
+  if (pathSegments.length <= 4) {
+    return ""; // If there are 4 or fewer segments, return an empty string (no relative path exists)
   }
 
-  return pathSegments.slice(3).join("/"); // Join remaining segments as the relative path
+  return pathSegments.slice(4).join("/"); // Join remaining segments as the relative path
 }
 
 export function getPathsUnderOrEqualDatasetFileNode(node: DatasetFileNode): string[] {