Do case insensitive comparison between dereferenced fields and intern…

…al ORC field names
trinodb · Apr 9, 2021 · 8bac015 · 8bac015
1 parent 987b1dc
commit 8bac015
Show file tree

Hide file tree

Showing 2 changed files with 47 additions and 4 deletions.
diff --git a/plugin/trino-hive/src/main/java/io/trino/plugin/hive/orc/OrcPageSourceFactory.java b/plugin/trino-hive/src/main/java/io/trino/plugin/hive/orc/OrcPageSourceFactory.java
@@ -298,16 +298,14 @@ private static ConnectorPageSource createOrcPageSource(
                         .collect(Collectors.groupingBy(
                                 HiveColumnHandle::getBaseColumnName,
                                 mapping(
-                                        column -> column.getHiveColumnProjectionInfo().map(HiveColumnProjectionInfo::getDereferenceNames).orElse(ImmutableList.<String>of()),
-                                        toList())));
+                                        OrcPageSourceFactory::getDereferencesAsList, toList())));
             }
             else {
                 projectionsByColumnIndex = projections.stream()
                         .collect(Collectors.groupingBy(
                                 HiveColumnHandle::getBaseHiveColumnIndex,
                                 mapping(
-                                        column -> column.getHiveColumnProjectionInfo().map(HiveColumnProjectionInfo::getDereferenceNames).orElse(ImmutableList.<String>of()),
-                                        toList())));
+                                        OrcPageSourceFactory::getDereferencesAsList, toList())));
             }
 
             TupleDomainOrcPredicateBuilder predicateBuilder = TupleDomainOrcPredicate.builder()
@@ -540,4 +538,13 @@ private static OrcColumn getNestedColumn(OrcColumn baseColumn, Optional<HiveColu
         }
         return current;
     }
+
+    private static List<String> getDereferencesAsList(HiveColumnHandle column)
+    {
+        return column.getHiveColumnProjectionInfo()
+                .map(info -> info.getDereferenceNames().stream()
+                        .map(dereference -> dereference.toLowerCase(ENGLISH))
+                        .collect(toImmutableList()))
+                .orElse(ImmutableList.of());
+    }
 }
diff --git a/testing/trino-product-tests/src/main/java/io/trino/tests/hive/TestHiveStorageFormats.java b/testing/trino-product-tests/src/main/java/io/trino/tests/hive/TestHiveStorageFormats.java
@@ -480,6 +480,42 @@ public void testSnappyCompressedParquetTableCreatedInHive()
         onHive().executeQuery("DROP TABLE " + tableName);
     }
 
+    @Test
+    public void testOrcStructsWithNonLowercaseFields()
+            throws SQLException
+    {
+        String tableName = "orc_structs_with_non_lowercase";
+
+        ensureDummyExists();
+        onHive().executeQuery("DROP TABLE IF EXISTS " + tableName);
+
+        onHive().executeQuery(format(
+                "CREATE TABLE %s (" +
+                        "   c_bigint BIGINT," +
+                        "   c_struct struct<testCustId:string, requestDate:string>)" +
+                        "STORED AS ORC ",
+                tableName));
+
+        onHive().executeQuery(format(
+                "INSERT INTO %s"
+                        // insert with SELECT because hive does not support array/map/struct functions in VALUES
+                        + " SELECT"
+                        + "   1,"
+                        + "   named_struct('testCustId', '1234', 'requestDate', 'some day')"
+                        // some hive versions don't allow INSERT from SELECT without FROM
+                        + " FROM dummy",
+                tableName));
+
+        setSessionProperty(onTrino().getConnection(), "hive.projection_pushdown_enabled", "true");
+        assertThat(onTrino().executeQuery("SELECT c_struct.testCustId FROM " + tableName)).containsOnly(row("1234"));
+        assertThat(onTrino().executeQuery("SELECT c_struct.testcustid FROM " + tableName)).containsOnly(row("1234"));
+        assertThat(onTrino().executeQuery("SELECT c_struct.requestDate FROM " + tableName)).containsOnly(row("some day"));
+        setSessionProperty(onTrino().getConnection(), "hive.projection_pushdown_enabled", "false");
+        assertThat(onTrino().executeQuery("SELECT c_struct.testCustId FROM " + tableName)).containsOnly(row("1234"));
+        assertThat(onTrino().executeQuery("SELECT c_struct.testcustid FROM " + tableName)).containsOnly(row("1234"));
+        assertThat(onTrino().executeQuery("SELECT c_struct.requestDate FROM " + tableName)).containsOnly(row("some day"));
+    }
+
     @Test(dataProvider = "storageFormatsWithNanosecondPrecision")
     public void testTimestampCreatedFromHive(StorageFormat storageFormat)
     {