Output metrics about remove_orphan_files execution

alter table lineitem execute remove_orphan_files(retention_threshold => '0d');

        metric_name         | metric_value
----------------------------+--------------
 processed_manifests_count  |            2
 valid_files_count          |           98
 scanned_files_count        |           97
 deleted_files_count        |            0

Author: raunaqmorarka

docs/src/main/sphinx/connector/iceberg.md

@@ -907,12 +907,39 @@ time is recommended to keep size of a table's data directory under control.
 ALTER TABLE test_table EXECUTE remove_orphan_files(retention_threshold => '7d');
 ```
 
+```text
+        metric_name         | metric_value
+----------------------------+--------------
+ processed_manifests_count  |            2
+ valid_files_count          |           98
+ scanned_files_count        |           97
+ deleted_files_count        |            0
+```
+
 The value for `retention_threshold` must be higher than or equal to
 `iceberg.remove-orphan-files.min-retention` in the catalog otherwise the
 procedure fails with a similar message: `Retention specified (1.00d) is shorter
 than the minimum retention configured in the system (7.00d)`. The default value
 for this property is `7d`.
 
+The output of the query has the following metrics:
+
+:::{list-table} Output
+:widths: 40, 60
+:header-rows: 1
+
+* - Property name
+  - Description
+* - `processed_manifests_count`
+  - The count of manifest files read by remove_orphan_files.
+* - `valid_files_count`
+  - The count of valid files found in the manifest files.
+* - `scanned_files_count`
+  - The count of files scanned from the file system.
+* - `deleted_files_count`
+  - The count of files deleted by remove_orphan_files.
+:::
+
 (drop-extended-stats)=
 ##### drop_extended_stats
 

plugin/trino-iceberg/src/main/java/io/trino/plugin/iceberg/IcebergMetadata.java

@@ -1989,8 +1989,7 @@ public Map<String, Long> executeTableExecute(ConnectorSession session, Connector
                 executeExpireSnapshots(session, executeHandle);
                 return ImmutableMap.of();
             case REMOVE_ORPHAN_FILES:
-                executeRemoveOrphanFiles(session, executeHandle);
-                return ImmutableMap.of();
+                return executeRemoveOrphanFiles(session, executeHandle);
             case ADD_FILES:
                 executeAddFiles(session, executeHandle);
                 return ImmutableMap.of();
@@ -2118,7 +2117,7 @@ private static void validateTableExecuteParameters(
                 sessionMinRetentionParameterName);
     }
 
-    public void executeRemoveOrphanFiles(ConnectorSession session, IcebergTableExecuteHandle executeHandle)
+    public Map<String, Long> executeRemoveOrphanFiles(ConnectorSession session, IcebergTableExecuteHandle executeHandle)
     {
         IcebergRemoveOrphanFilesHandle removeOrphanFilesHandle = (IcebergRemoveOrphanFilesHandle) executeHandle.procedureHandle();
 
@@ -2135,14 +2134,14 @@ public void executeRemoveOrphanFiles(ConnectorSession session, IcebergTableExecu
 
         if (table.currentSnapshot() == null) {
             log.debug("Skipping remove_orphan_files procedure for empty table %s", table);
-            return;
+            return ImmutableMap.of();
         }
 
         Instant expiration = session.getStart().minusMillis(retention.toMillis());
-        removeOrphanFiles(table, session, executeHandle.schemaTableName(), expiration, executeHandle.fileIoProperties());
+        return removeOrphanFiles(table, session, executeHandle.schemaTableName(), expiration, executeHandle.fileIoProperties());
     }
 
-    private void removeOrphanFiles(Table table, ConnectorSession session, SchemaTableName schemaTableName, Instant expiration, Map<String, String> fileIoProperties)
+    private Map<String, Long> removeOrphanFiles(Table table, ConnectorSession session, SchemaTableName schemaTableName, Instant expiration, Map<String, String> fileIoProperties)
     {
         Set<String> processedManifestFilePaths = new HashSet<>();
         // Similarly to issues like https://github.com/trinodb/trino/issues/13759, equivalent paths may have different String
@@ -2205,7 +2204,18 @@ private void removeOrphanFiles(Table table, ConnectorSession session, SchemaTabl
             // Ensure any futures still running are canceled in case of failure
             manifestScanFutures.forEach(future -> future.cancel(true));
         }
-        scanAndDeleteInvalidFiles(table, session, schemaTableName, expiration, validFileNames, fileIoProperties);
+        ScanAndDeleteResult result = scanAndDeleteInvalidFiles(table, session, schemaTableName, expiration, validFileNames, fileIoProperties);
+        log.info("remove_orphan_files for table %s completed. Processed %d manifest files, found %d valid files, scanned %d files, deleted %d files",
+                schemaTableName,
+                processedManifestFilePaths.size(),
+                validFileNames.size() - 1, // excluding version-hint.text
+                result.scannedFilesCount(),
+                result.deletedFilesCount());
+        return ImmutableMap.of(
+                "processed_manifests_count", (long) processedManifestFilePaths.size(),
+                "valid_files_count", (long) validFileNames.size() - 1, // excluding version-hint.text
+                "scanned_files_count", result.scannedFilesCount(),
+                "deleted_files_count", result.deletedFilesCount());
     }
 
     public void executeAddFiles(ConnectorSession session, IcebergTableExecuteHandle executeHandle)
@@ -2240,17 +2250,21 @@ public void executeAddFilesFromTable(ConnectorSession session, IcebergTableExecu
                 icebergScanExecutor);
     }
 
-    private void scanAndDeleteInvalidFiles(Table table, ConnectorSession session, SchemaTableName schemaTableName, Instant expiration, Set<String> validFiles, Map<String, String> fileIoProperties)
+    private ScanAndDeleteResult scanAndDeleteInvalidFiles(Table table, ConnectorSession session, SchemaTableName schemaTableName, Instant expiration, Set<String> validFiles, Map<String, String> fileIoProperties)
     {
         List<Future<?>> deleteFutures = new ArrayList<>();
+        long deletedFilesCount = 0;
+        long scannedFilesCount = 0;
         try {
             List<Location> filesToDelete = new ArrayList<>(DELETE_BATCH_SIZE);
             TrinoFileSystem fileSystem = fileSystemFactory.create(session.getIdentity(), fileIoProperties);
             FileIterator allFiles = fileSystem.listFiles(Location.of(table.location()));
             while (allFiles.hasNext()) {
                 FileEntry entry = allFiles.next();
+                scannedFilesCount++;
                 if (entry.lastModified().isBefore(expiration) && !validFiles.contains(entry.location().fileName())) {
                     filesToDelete.add(entry.location());
+                    deletedFilesCount++;
                     if (filesToDelete.size() >= DELETE_BATCH_SIZE) {
                         List<Location> finalFilesToDelete = filesToDelete;
                         deleteFutures.add(icebergFileDeleteExecutor.submit(() -> deleteFiles(finalFilesToDelete, schemaTableName, fileSystem)));
@@ -2277,8 +2291,11 @@ private void scanAndDeleteInvalidFiles(Table table, ConnectorSession session, Sc
             // Ensure any futures still running are canceled in case of failure
             deleteFutures.forEach(future -> future.cancel(true));
         }
+        return new ScanAndDeleteResult(deletedFilesCount, scannedFilesCount);
     }
 
+    private record ScanAndDeleteResult(long deletedFilesCount, long scannedFilesCount) {}
+
     private void deleteFiles(List<Location> files, SchemaTableName schemaTableName, TrinoFileSystem fileSystem)
     {
         log.debug("Deleting files while removing orphan files for table %s [%s]", schemaTableName, files);

plugin/trino-iceberg/src/test/java/io/trino/plugin/iceberg/BaseIcebergConnectorTest.java

@@ -6645,7 +6645,10 @@ public void testRemoveOrphanFiles()
         List<String> initialDataFiles = getAllDataFilesFromTableDirectory(tableName);
         assertThat(initialDataFiles).contains(orphanFile);
 
-        assertQuerySucceeds(sessionWithShortRetentionUnlocked, "ALTER TABLE " + tableName + " EXECUTE REMOVE_ORPHAN_FILES (retention_threshold => '0s')");
+        assertUpdate(
+                sessionWithShortRetentionUnlocked,
+                "ALTER TABLE " + tableName + " EXECUTE REMOVE_ORPHAN_FILES (retention_threshold => '0s')",
+                "VALUES ('processed_manifests_count', 3), ('valid_files_count', 16), ('scanned_files_count', 17), ('deleted_files_count', 1)");
         assertQuery("SELECT * FROM " + tableName, "VALUES ('one', 1), ('three', 3)");
 
         List<String> updatedDataFiles = getAllDataFilesFromTableDirectory(tableName);

testing/trino-testing/src/main/java/io/trino/testing/QueryAssertions.java

@@ -336,7 +336,7 @@ private static void assertDistributedQuery(
         List<MaterializedRow> actualRows = actualResults.getMaterializedRows();
         List<MaterializedRow> expectedRows = expectedResults.getMaterializedRows();
 
-        if (compareUpdate) {
+        if (compareUpdate && !actualResults.getUpdateType().equals(Optional.of("ALTER TABLE EXECUTE"))) {
             if (actualResults.getUpdateType().isEmpty()) {
                 fail("update type not present for query " + queryId + ": \n" + actual);
             }