Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[WIP] Pushdown dereference expression to Parquet reader #187

Closed
wants to merge 1 commit into from
Closed

[WIP] Pushdown dereference expression to Parquet reader #187

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
23 changes: 17 additions & 6 deletions presto-hive/src/main/java/io/prestosql/plugin/hive/HiveColumnHandle.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -15,6 +15,7 @@

import com.fasterxml.jackson.annotation.JsonCreator;
import com.fasterxml.jackson.annotation.JsonProperty;
import io.prestosql.spi.NestedColumn;
import io.prestosql.spi.connector.ColumnHandle;
import io.prestosql.spi.connector.ColumnMetadata;
import io.prestosql.spi.type.TypeManager;
Expand Down Expand Up @@ -60,6 +61,7 @@ public enum ColumnType
private final int hiveColumnIndex;
private final ColumnType columnType;
private final Optional<String> comment;
private final Optional<NestedColumn> nestedColumn;

@JsonCreator
public HiveColumnHandle(
Expand All @@ -68,7 +70,8 @@ public HiveColumnHandle(
@JsonProperty("typeSignature") TypeSignature typeSignature,
@JsonProperty("hiveColumnIndex") int hiveColumnIndex,
@JsonProperty("columnType") ColumnType columnType,
@JsonProperty("comment") Optional<String> comment)
@JsonProperty("comment") Optional<String> comment,
@JsonProperty("nestedColumn") Optional<NestedColumn> nestedColumn)
{
this.name = requireNonNull(name, "name is null");
checkArgument(hiveColumnIndex >= 0 || columnType == PARTITION_KEY || columnType == SYNTHESIZED, "hiveColumnIndex is negative");
Expand All @@ -77,6 +80,7 @@ public HiveColumnHandle(
this.typeName = requireNonNull(typeSignature, "type is null");
this.columnType = requireNonNull(columnType, "columnType is null");
this.comment = requireNonNull(comment, "comment is null");
this.nestedColumn = requireNonNull(nestedColumn, "nestedColumn is null");
}

@JsonProperty
Expand All @@ -97,6 +101,12 @@ public int getHiveColumnIndex()
return hiveColumnIndex;
}

@JsonProperty
public Optional<NestedColumn> getNestedColumn()
{
return nestedColumn;
}

public boolean isPartitionKey()
{
return columnType == PARTITION_KEY;
Expand Down Expand Up @@ -133,7 +143,7 @@ public ColumnType getColumnType()
@Override
public int hashCode()
{
return Objects.hash(name, hiveColumnIndex, hiveType, columnType, comment);
return Objects.hash(name, hiveColumnIndex, hiveType, columnType, comment, nestedColumn);
}

@Override
Expand All @@ -150,7 +160,8 @@ public boolean equals(Object obj)
Objects.equals(this.hiveColumnIndex, other.hiveColumnIndex) &&
Objects.equals(this.hiveType, other.hiveType) &&
Objects.equals(this.columnType, other.columnType) &&
Objects.equals(this.comment, other.comment);
Objects.equals(this.comment, other.comment) &&
Objects.equals(this.nestedColumn, other.nestedColumn);
}

@Override
Expand All @@ -167,12 +178,12 @@ public static HiveColumnHandle updateRowIdHandle()
// plan-time support for row-by-row delete so that planning doesn't fail. This is why we need
// rowid handle. Note that in Hive connector, rowid handle is not implemented beyond plan-time.

return new HiveColumnHandle(UPDATE_ROW_ID_COLUMN_NAME, HIVE_LONG, BIGINT.getTypeSignature(), -1, SYNTHESIZED, Optional.empty());
return new HiveColumnHandle(UPDATE_ROW_ID_COLUMN_NAME, HIVE_LONG, BIGINT.getTypeSignature(), -1, SYNTHESIZED, Optional.empty(), Optional.empty());
}

public static HiveColumnHandle pathColumnHandle()
{
return new HiveColumnHandle(PATH_COLUMN_NAME, PATH_HIVE_TYPE, PATH_TYPE_SIGNATURE, PATH_COLUMN_INDEX, SYNTHESIZED, Optional.empty());
return new HiveColumnHandle(PATH_COLUMN_NAME, PATH_HIVE_TYPE, PATH_TYPE_SIGNATURE, PATH_COLUMN_INDEX, SYNTHESIZED, Optional.empty(), Optional.empty());
}

/**
Expand All @@ -182,7 +193,7 @@ public static HiveColumnHandle pathColumnHandle()
*/
public static HiveColumnHandle bucketColumnHandle()
{
return new HiveColumnHandle(BUCKET_COLUMN_NAME, BUCKET_HIVE_TYPE, BUCKET_TYPE_SIGNATURE, BUCKET_COLUMN_INDEX, SYNTHESIZED, Optional.empty());
return new HiveColumnHandle(BUCKET_COLUMN_NAME, BUCKET_HIVE_TYPE, BUCKET_TYPE_SIGNATURE, BUCKET_COLUMN_INDEX, SYNTHESIZED, Optional.empty(), Optional.empty());
}

public static boolean isPathColumnHandle(HiveColumnHandle column)
Expand Down
39 changes: 38 additions & 1 deletion presto-hive/src/main/java/io/prestosql/plugin/hive/HiveMetadata.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -15,6 +15,7 @@

import com.google.common.annotations.VisibleForTesting;
import com.google.common.base.Joiner;
import com.google.common.base.Preconditions;
import com.google.common.base.Splitter;
import com.google.common.base.Suppliers;
import com.google.common.base.Verify;
Expand Down Expand Up @@ -44,6 +45,7 @@
import io.prestosql.plugin.hive.metastore.Table;
import io.prestosql.plugin.hive.metastore.thrift.ThriftMetastoreUtil;
import io.prestosql.plugin.hive.statistics.HiveStatisticsProvider;
import io.prestosql.spi.NestedColumn;
import io.prestosql.spi.PrestoException;
import io.prestosql.spi.StandardErrorCode;
import io.prestosql.spi.block.Block;
Expand Down Expand Up @@ -172,6 +174,7 @@
import static io.prestosql.plugin.hive.HiveUtil.decodeViewData;
import static io.prestosql.plugin.hive.HiveUtil.encodeViewData;
import static io.prestosql.plugin.hive.HiveUtil.getPartitionKeyColumnHandles;
import static io.prestosql.plugin.hive.HiveUtil.getRegularColumnHandles;
import static io.prestosql.plugin.hive.HiveUtil.hiveColumnHandles;
import static io.prestosql.plugin.hive.HiveUtil.schemaTableName;
import static io.prestosql.plugin.hive.HiveUtil.toPartitionValues;
Expand Down Expand Up @@ -616,6 +619,39 @@ public ColumnMetadata getColumnMetadata(ConnectorSession session, ConnectorTable
return ((HiveColumnHandle) columnHandle).getColumnMetadata(typeManager);
}

@Override
public Map<NestedColumn, ColumnHandle> getNestedColumnHandles(ConnectorSession session, ConnectorTableHandle tableHandle, Collection<NestedColumn> nestedColumns)
{
if (!HiveSessionProperties.isStatisticsEnabled(session)) {
return ImmutableMap.of();
}

SchemaTableName tableName = schemaTableName(tableHandle);
Optional<Table> table = metastore.getTable(tableName.getSchemaName(), tableName.getTableName());

if (!table.isPresent()) {
throw new TableNotFoundException(tableName);
}

// Only pushdown nested column for parquet table for now
if (!extractHiveStorageFormat(table.get()).equals(HiveStorageFormat.PARQUET)) {
return ImmutableMap.of();
}

List<HiveColumnHandle> regularColumnHandles = getRegularColumnHandles(table.get());
Map<String, HiveColumnHandle> regularHiveColumnHandles = regularColumnHandles.stream().collect(Collectors.toMap(HiveColumnHandle::getName, identity()));
ImmutableMap.Builder<NestedColumn, ColumnHandle> columnHandles = ImmutableMap.builder();
for (NestedColumn nestedColumn : nestedColumns) {
HiveColumnHandle hiveColumnHandle = regularHiveColumnHandles.get(nestedColumn.getBase());
Optional<HiveType> childType = hiveColumnHandle.getHiveType().findChildType(nestedColumn);
Preconditions.checkArgument(childType.isPresent(), "%s doesn't exist in parent type %s", nestedColumn, hiveColumnHandle.getHiveType());
if (hiveColumnHandle != null) {
columnHandles.put(nestedColumn, new HiveColumnHandle(nestedColumn.getName(), childType.get(), childType.get().getTypeSignature(), hiveColumnHandle.getHiveColumnIndex(), hiveColumnHandle.getColumnType(), hiveColumnHandle.getComment(), Optional.of(nestedColumn)));
}
}
return columnHandles.build();
}

@Override
public void createSchema(ConnectorSession session, String schemaName, Map<String, Object> properties)
{
Expand Down Expand Up @@ -2124,7 +2160,8 @@ else if (column.isHidden()) {
column.getType().getTypeSignature(),
ordinal,
columnType,
Optional.ofNullable(column.getComment())));
Optional.ofNullable(column.getComment()),
Optional.empty()));
ordinal++;
}

Expand Down
18 changes: 15 additions & 3 deletions presto-hive/src/main/java/io/prestosql/plugin/hive/HivePageSourceProvider.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -17,6 +17,7 @@
import com.google.common.collect.ImmutableSet;
import io.prestosql.plugin.hive.HdfsEnvironment.HdfsContext;
import io.prestosql.plugin.hive.HiveSplit.BucketConversion;
import io.prestosql.spi.NestedColumn;
import io.prestosql.spi.connector.ColumnHandle;
import io.prestosql.spi.connector.ConnectorPageSource;
import io.prestosql.spi.connector.ConnectorPageSourceProvider;
Expand Down Expand Up @@ -319,8 +320,13 @@ public static List<ColumnMapping> buildColumnMappings(
for (HiveColumnHandle column : columns) {
Optional<HiveType> coercionFrom = Optional.ofNullable(columnCoercions.get(column.getHiveColumnIndex()));
if (column.getColumnType() == REGULAR) {
checkArgument(regularColumnIndices.add(column.getHiveColumnIndex()), "duplicate hiveColumnIndex in columns list");
columnMappings.add(regular(column, regularIndex, coercionFrom));
if (column.getNestedColumn().isPresent()) {
columnMappings.add(regular(column, regularIndex, getHiveType(coercionFrom, column.getNestedColumn().get())));
}
else {
checkArgument(regularColumnIndices.add(column.getHiveColumnIndex()), "duplicate hiveColumnIndex in columns list");
columnMappings.add(regular(column, regularIndex, coercionFrom));
}
regularIndex++;
}
else {
Expand All @@ -344,6 +350,11 @@ public static List<ColumnMapping> buildColumnMappings(
return columnMappings.build();
}

private static Optional<HiveType> getHiveType(Optional<HiveType> baseType, NestedColumn nestedColumn)
{
return baseType.flatMap(type -> type.findChildType(nestedColumn));
}

public static List<ColumnMapping> extractRegularAndInterimColumnMappings(List<ColumnMapping> columnMappings)
{
return columnMappings.stream()
Expand All @@ -365,7 +376,8 @@ public static List<HiveColumnHandle> toColumnHandles(List<ColumnMapping> regular
columnMapping.getCoercionFrom().get().getTypeSignature(),
columnHandle.getHiveColumnIndex(),
columnHandle.getColumnType(),
Optional.empty());
Optional.empty(),
columnHandle.getNestedColumn());
})
.collect(toList());
}
Expand Down
18 changes: 18 additions & 0 deletions presto-hive/src/main/java/io/prestosql/plugin/hive/HiveType.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -15,7 +15,9 @@

import com.fasterxml.jackson.annotation.JsonCreator;
import com.fasterxml.jackson.annotation.JsonValue;
import com.google.common.base.Preconditions;
import com.google.common.collect.ImmutableList;
import io.prestosql.spi.NestedColumn;
import io.prestosql.spi.PrestoException;
import io.prestosql.spi.type.NamedTypeSignature;
import io.prestosql.spi.type.RowFieldName;
Expand Down Expand Up @@ -175,6 +177,22 @@ public static boolean isSupportedType(TypeInfo typeInfo)
return false;
}

public Optional<HiveType> findChildType(NestedColumn nestedColumn)
{
TypeInfo typeInfo = getTypeInfo();
for (String part : nestedColumn.getRest()) {
Preconditions.checkArgument(typeInfo instanceof StructTypeInfo, "typeinfo is not struct type", typeInfo);
StructTypeInfo structTypeInfo = (StructTypeInfo) typeInfo;
try {
typeInfo = structTypeInfo.getStructFieldTypeInfo(part);
}
catch (RuntimeException e) {
return Optional.empty();
}
}
return Optional.of(toHiveType(typeInfo));
}

@JsonCreator
public static HiveType valueOf(String hiveTypeName)
{
Expand Down
4 changes: 2 additions & 2 deletions presto-hive/src/main/java/io/prestosql/plugin/hive/HiveUtil.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -825,7 +825,7 @@ public static List<HiveColumnHandle> getRegularColumnHandles(Table table)
// ignore unsupported types rather than failing
HiveType hiveType = field.getType();
if (hiveType.isSupportedType()) {
columns.add(new HiveColumnHandle(field.getName(), hiveType, hiveType.getTypeSignature(), hiveColumnIndex, REGULAR, field.getComment()));
columns.add(new HiveColumnHandle(field.getName(), hiveType, hiveType.getTypeSignature(), hiveColumnIndex, REGULAR, field.getComment(), Optional.empty()));
}
hiveColumnIndex++;
}
Expand All @@ -843,7 +843,7 @@ public static List<HiveColumnHandle> getPartitionKeyColumnHandles(Table table)
if (!hiveType.isSupportedType()) {
throw new PrestoException(NOT_SUPPORTED, format("Unsupported Hive type %s found in partition keys of table %s.%s", hiveType, table.getDatabaseName(), table.getTableName()));
}
columns.add(new HiveColumnHandle(field.getName(), hiveType, hiveType.getTypeSignature(), -1, PARTITION_KEY, field.getComment()));
columns.add(new HiveColumnHandle(field.getName(), hiveType, hiveType.getTypeSignature(), -1, PARTITION_KEY, field.getComment(), Optional.empty()));
}

return columns.build();
Expand Down
2 changes: 1 addition & 1 deletion presto-hive/src/main/java/io/prestosql/plugin/hive/orc/OrcPageSourceFactory.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -264,7 +264,7 @@ private static List<HiveColumnHandle> getPhysicalHiveColumnHandles(List<HiveColu
physicalOrdinal = nextMissingColumnIndex;
nextMissingColumnIndex++;
}
physicalColumns.add(new HiveColumnHandle(column.getName(), column.getHiveType(), column.getTypeSignature(), physicalOrdinal, column.getColumnType(), column.getComment()));
physicalColumns.add(new HiveColumnHandle(column.getName(), column.getHiveType(), column.getTypeSignature(), physicalOrdinal, column.getColumnType(), column.getComment(), column.getNestedColumn()));
}
return physicalColumns.build();
}
Expand Down
Loading