Github user rmetzger commented on a diff in the pull request:
https://github.com/apache/flink/pull/411#discussion_r24833839
--- Diff:
flink-staging/flink-hcatalog/src/main/java/org/apache/flink/hcatalog/HCatInputFormatBase.java
---
@@ -0,0 +1,413 @@
+/*
+ * Licensed to the Apache Software Foundation (ASF) under one
+ * or more contributor license agreements. See the NOTICE file
+ * distributed with this work for additional information
+ * regarding copyright ownership. The ASF licenses this file
+ * to you under the Apache License, Version 2.0 (the
+ * "License"); you may not use this file except in compliance
+ * with the License. You may obtain a copy of the License at
+ *
+ * http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+package org.apache.flink.hcatalog;
+
+import org.apache.flink.api.common.io.InputFormat;
+import org.apache.flink.api.common.io.LocatableInputSplitAssigner;
+import org.apache.flink.api.common.io.statistics.BaseStatistics;
+import org.apache.flink.api.common.typeinfo.BasicTypeInfo;
+import org.apache.flink.api.common.typeinfo.PrimitiveArrayTypeInfo;
+import org.apache.flink.api.common.typeinfo.TypeInformation;
+import org.apache.flink.api.java.hadoop.mapreduce.utils.HadoopUtils;
+import org.apache.flink.api.java.hadoop.mapreduce.wrapper.HadoopInputSplit;
+import org.apache.flink.api.java.typeutils.ResultTypeQueryable;
+import org.apache.flink.api.java.typeutils.TupleTypeInfo;
+import org.apache.flink.api.java.typeutils.WritableTypeInfo;
+import org.apache.flink.core.io.InputSplitAssigner;
+import org.apache.hadoop.conf.Configuration;
+import org.apache.hadoop.io.WritableComparable;
+import org.apache.hadoop.mapreduce.InputSplit;
+import org.apache.hadoop.mapreduce.JobContext;
+import org.apache.hadoop.mapreduce.JobID;
+import org.apache.hadoop.mapreduce.RecordReader;
+import org.apache.hadoop.mapreduce.TaskAttemptContext;
+import org.apache.hadoop.mapreduce.TaskAttemptID;
+import org.apache.hive.hcatalog.common.HCatException;
+import org.apache.hive.hcatalog.common.HCatUtil;
+import org.apache.hive.hcatalog.data.DefaultHCatRecord;
+import org.apache.hive.hcatalog.data.HCatRecord;
+import org.apache.hive.hcatalog.data.schema.HCatFieldSchema;
+import org.apache.hive.hcatalog.data.schema.HCatSchema;
+
+import java.io.IOException;
+import java.io.ObjectInputStream;
+import java.io.ObjectOutputStream;
+import java.util.ArrayList;
+import java.util.List;
+
+/**
+ * A InputFormat to read from HCatalog tables.
+ * The InputFormat supports projection (selection and order of fields) and
partition filters.
+ *
+ * Data can be returned as {@link
org.apache.hive.hcatalog.data.HCatRecord} or Flink {@link
org.apache.flink.api.java.tuple.Tuple}.
+ * Flink Tuples are only supported for primitive type fields
+ * (no STRUCT, ARRAY, or MAP data types) and have a size limitation.
+ *
+ * @param <T>
+ */
+public abstract class HCatInputFormatBase<T> implements InputFormat<T,
HadoopInputSplit>, ResultTypeQueryable<T> {
+
+ private static final long serialVersionUID = 1L;
+
+ private Configuration configuration;
+
+ private org.apache.hive.hcatalog.mapreduce.HCatInputFormat
hCatInputFormat;
+ private RecordReader<WritableComparable, HCatRecord> recordReader;
+ private boolean fetched = false;
+ private boolean hasNext;
+
+ protected String[] fieldNames = new String[0];
+ protected HCatSchema outputSchema;
+
+ private TypeInformation<T> resultType;
+
+ public HCatInputFormatBase() { }
+
+ /**
+ * Creates a HCatInputFormat for the given database and table.
+ * By default, the InputFormat returns {@link
org.apache.hive.hcatalog.data.HCatRecord}.
+ * The return type of the InputFormat can be changed to Flink {@link
org.apache.flink.api.java.tuple.Tuple} by calling
+ * {@link HCatInputFormatBase#asFlinkTuples()}.
+ *
+ * @param database The name of the database to read from.
+ * @param table The name of the table to read.
+ * @throws java.io.IOException
+ */
+ public HCatInputFormatBase(String database, String table) throws
IOException {
+ this(database, table, new Configuration());
+ }
+
+ /**
+ * Creates a HCatInputFormat for the given database, table, and
+ * {@link org.apache.hadoop.conf.Configuration}.
+ * By default, the InputFormat returns {@link
org.apache.hive.hcatalog.data.HCatRecord}.
+ * The return type of the InputFormat can be changed to Flink {@link
org.apache.flink.api.java.tuple.Tuple} by calling
+ * {@link HCatInputFormatBase#asFlinkTuples()}.
+ *
+ * @param database The name of the database to read from.
+ * @param table The name of the table to read.
+ * @param config The Configuration for the InputFormat.
+ * @throws java.io.IOException
+ */
+ public HCatInputFormatBase(String database, String table, Configuration
config) throws IOException {
+ super();
+ this.configuration = config;
+ HadoopUtils.mergeHadoopConf(this.configuration);
+
+ this.hCatInputFormat =
org.apache.hive.hcatalog.mapreduce.HCatInputFormat.setInput(this.configuration,
database, table);
+ this.outputSchema =
org.apache.hive.hcatalog.mapreduce.HCatInputFormat.getTableSchema(this.configuration);
+
+ // configure output schema of HCatFormat
+ configuration.set("mapreduce.lib.hcat.output.schema",
HCatUtil.serialize(outputSchema));
+ // set type information
+ this.resultType = new WritableTypeInfo(DefaultHCatRecord.class);
+ }
+
+ /**
+ * Specifies the fields which are returned by the InputFormat and their
order.
+ *
+ * @param fields The fields and their order which are returned by the
InputFormat.
+ * @return This InputFormat with specified return fields.
+ * @throws java.io.IOException
+ */
+ public HCatInputFormatBase<T> getFields(String... fields) throws
IOException {
+
+ // build output schema
+ ArrayList<HCatFieldSchema> fieldSchemas = new
ArrayList<HCatFieldSchema>(fields.length);
+ for(String field : fields) {
+ fieldSchemas.add(this.outputSchema.get(field));
+ }
+ this.outputSchema = new HCatSchema(fieldSchemas);
+
+ // update output schema configuration
+ configuration.set("mapreduce.lib.hcat.output.schema",
HCatUtil.serialize(outputSchema));
+
+ return this;
+ }
+
+ /**
+ * Specifies a SQL-like filter condition on the table's partition
columns.
+ * Filter conditions on non-partition columns are invalid.
+ * A partition filter can significantly reduce the amount of data to be
read.
+ *
+ * @param filter A SQL-like filter condition on the table's partition
columns.
+ * @return This InputFormat with specified partition filter.
+ * @throws java.io.IOException
+ */
+ public HCatInputFormatBase<T> withFilter(String filter) throws
IOException {
+
+ // set filter
+ this.hCatInputFormat.setFilter(filter);
+
+ return this;
+ }
+
+ /**
+ * Specifies that the InputFormat returns Flink {@link
org.apache.flink.api.java.tuple.Tuple}
+ * instead of {@link org.apache.hive.hcatalog.data.HCatRecord}.
+ * At the moment, the following restrictions apply for returning Flink
tuples:
+ *
+ * <ul>
+ * <li>Only primitive type fields can be returned in Flink Tuples
+ * (no STRUCT, MAP, ARRAY data types).</li>
+ * <li>Only a limited number of fields can be returned as Flink
Tuple.</li>
+ * </ul>
+ *
+ * @return This InputFormat.
+ * @throws org.apache.hive.hcatalog.common.HCatException
+ */
+ public HCatInputFormatBase<T> asFlinkTuples() throws HCatException {
+
+ // build type information
+ int numFields = outputSchema.getFields().size();
+ if(numFields > this.getMaxFlinkTupleSize()) {
+ throw new IllegalArgumentException("Only up to
"+this.getMaxFlinkTupleSize()+
+ " fields can be returned as Flink
tuples.");
+ }
+
+ TypeInformation[] fieldTypes = new TypeInformation[numFields];
+ fieldNames = new String[numFields];
+ for (String fieldName : outputSchema.getFieldNames()) {
+ HCatFieldSchema field = outputSchema.get(fieldName);
+
+ int fieldPos = outputSchema.getPosition(fieldName);
+ TypeInformation fieldType = getFieldType(field);
+
+ fieldTypes[fieldPos] = fieldType;
+ fieldNames[fieldPos] = fieldName;
+
+ }
+ this.resultType = new TupleTypeInfo(fieldTypes);
+
+ return this;
+ }
+
+ protected abstract int getMaxFlinkTupleSize();
+
+ private TypeInformation getFieldType(HCatFieldSchema fieldSchema) {
+
+ switch(fieldSchema.getType()) {
+ case INT:
+ return BasicTypeInfo.INT_TYPE_INFO;
+ case TINYINT:
+ return BasicTypeInfo.BYTE_TYPE_INFO;
+ case SMALLINT:
+ return BasicTypeInfo.SHORT_TYPE_INFO;
+ case BIGINT:
+ return BasicTypeInfo.LONG_TYPE_INFO;
+ case BOOLEAN:
+ return BasicTypeInfo.BOOLEAN_TYPE_INFO;
+ case FLOAT:
+ return BasicTypeInfo.FLOAT_TYPE_INFO;
+ case DOUBLE:
+ return BasicTypeInfo.DOUBLE_TYPE_INFO;
+ case STRING:
+ return BasicTypeInfo.STRING_TYPE_INFO;
+ case BINARY:
+ return
PrimitiveArrayTypeInfo.BYTE_PRIMITIVE_ARRAY_TYPE_INFO;
+ case ARRAY:
+ throw new UnsupportedOperationException("ARRAY
type is not supported in Flink tuples, yet.");
--- End diff --
Why can't we use genericTypes here for arrays, hashmps, lists (see
https://cwiki.apache.org/confluence/display/Hive/HCatalog+InputOutput)
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
