Search in sources :

Example 31 with TableRowParser

use of org.apache.beam.sdk.io.gcp.bigquery.BigQueryIO.TableRowParser in project beam by apache.

the class BigQueryIOStorageReadTest method testTableSourceInitialSplit_WithSelectedFieldsAndRowRestriction.

@Test
public void testTableSourceInitialSplit_WithSelectedFieldsAndRowRestriction() throws Exception {
    fakeDatasetService.createDataset("foo.com:project", "dataset", "", "", null);
    TableReference tableRef = BigQueryHelpers.parseTableSpec("foo.com:project:dataset.table");
    Table table = new Table().setTableReference(tableRef).setNumBytes(100L).setSchema(TABLE_SCHEMA);
    fakeDatasetService.createTable(table);
    CreateReadSessionRequest expectedRequest = CreateReadSessionRequest.newBuilder().setParent("projects/project-id").setReadSession(ReadSession.newBuilder().setTable("projects/foo.com:project/datasets/dataset/tables/table").setReadOptions(ReadSession.TableReadOptions.newBuilder().addSelectedFields("name").setRowRestriction("number > 5"))).setMaxStreamCount(10).build();
    ReadSession.Builder builder = ReadSession.newBuilder().setAvroSchema(AvroSchema.newBuilder().setSchema(TRIMMED_AVRO_SCHEMA_STRING)).setDataFormat(DataFormat.AVRO);
    for (int i = 0; i < 10; i++) {
        builder.addStreams(ReadStream.newBuilder().setName("stream-" + i));
    }
    StorageClient fakeStorageClient = mock(StorageClient.class);
    when(fakeStorageClient.createReadSession(expectedRequest)).thenReturn(builder.build());
    BigQueryStorageTableSource<TableRow> tableSource = BigQueryStorageTableSource.create(ValueProvider.StaticValueProvider.of(tableRef), StaticValueProvider.of(Lists.newArrayList("name")), StaticValueProvider.of("number > 5"), new TableRowParser(), TableRowJsonCoder.of(), new FakeBigQueryServices().withDatasetService(fakeDatasetService).withStorageClient(fakeStorageClient));
    List<? extends BoundedSource<TableRow>> sources = tableSource.split(10L, options);
    assertEquals(10L, sources.size());
}
Also used : TableReference(com.google.api.services.bigquery.model.TableReference) Table(com.google.api.services.bigquery.model.Table) ReadSession(com.google.cloud.bigquery.storage.v1.ReadSession) TableRow(com.google.api.services.bigquery.model.TableRow) StorageClient(org.apache.beam.sdk.io.gcp.bigquery.BigQueryServices.StorageClient) FakeBigQueryServices(org.apache.beam.sdk.io.gcp.testing.FakeBigQueryServices) CreateReadSessionRequest(com.google.cloud.bigquery.storage.v1.CreateReadSessionRequest) TableRowParser(org.apache.beam.sdk.io.gcp.bigquery.BigQueryIO.TableRowParser) Test(org.junit.Test)

Example 32 with TableRowParser

use of org.apache.beam.sdk.io.gcp.bigquery.BigQueryIO.TableRowParser in project beam by apache.

the class BigQueryIOStorageQueryTest method testQuerySourceCreateReader.

@Test
public void testQuerySourceCreateReader() throws Exception {
    BigQueryStorageQuerySource<TableRow> querySource = BigQueryStorageQuerySource.create(/* stepUuid = */
    "testStepUuid", ValueProvider.StaticValueProvider.of("SELECT * FROM `dataset.table`"), /* flattenResults = */
    false, /* useLegacySql = */
    false, /* priority = */
    QueryPriority.INTERACTIVE, /* location = */
    "asia-northeast1", /* queryTempDataset = */
    null, /* kmsKey = */
    null, null, new TableRowParser(), TableRowJsonCoder.of(), fakeBigQueryServices);
    thrown.expect(UnsupportedOperationException.class);
    thrown.expectMessage("BigQuery storage source must be split before reading");
    querySource.createReader(options);
}
Also used : TableRow(com.google.api.services.bigquery.model.TableRow) TableRowParser(org.apache.beam.sdk.io.gcp.bigquery.BigQueryIO.TableRowParser) Test(org.junit.Test)

Example 33 with TableRowParser

use of org.apache.beam.sdk.io.gcp.bigquery.BigQueryIO.TableRowParser in project beam by apache.

the class BigQueryIOStorageQueryTest method testQuerySourceInitialSplit_NoReferencedTables.

/**
 * This test simulates the scenario where the SQL text which is executed by the query job doesn't
 * by itself refer to any tables (e.g. "SELECT 17 AS value"), and thus there are no referenced
 * tables when the dry run of the query is performed.
 */
@Test
public void testQuerySourceInitialSplit_NoReferencedTables() throws Exception {
    Table queryResultTable = new Table().setSchema(new TableSchema().setFields(ImmutableList.of(new TableFieldSchema().setName("name").setType("STRING"), new TableFieldSchema().setName("number").setType("INTEGER")))).setNumBytes(1024L * 1024L);
    String encodedQuery = FakeBigQueryServices.encodeQueryResult(queryResultTable);
    fakeJobService.expectDryRunQuery(options.getProject(), encodedQuery, new JobStatistics().setQuery(new JobStatistics2().setTotalBytesProcessed(1024L * 1024L).setReferencedTables(ImmutableList.of())));
    String stepUuid = "testStepUuid";
    TableReference tempTableReference = createTempTableReference(options.getProject(), BigQueryResourceNaming.createJobIdPrefix(options.getJobName(), stepUuid, JobType.QUERY), Optional.empty());
    CreateReadSessionRequest expectedRequest = CreateReadSessionRequest.newBuilder().setParent("projects/" + options.getProject()).setReadSession(ReadSession.newBuilder().setTable(BigQueryHelpers.toTableResourceName(tempTableReference))).setMaxStreamCount(1024).build();
    Schema sessionSchema = SchemaBuilder.record("__root__").fields().name("name").type().nullable().stringType().noDefault().name("number").type().nullable().longType().noDefault().endRecord();
    ReadSession.Builder builder = ReadSession.newBuilder().setAvroSchema(AvroSchema.newBuilder().setSchema(sessionSchema.toString())).setDataFormat(DataFormat.AVRO);
    for (int i = 0; i < 1024; i++) {
        builder.addStreams(ReadStream.newBuilder().setName("stream-" + i));
    }
    StorageClient fakeStorageClient = mock(StorageClient.class);
    when(fakeStorageClient.createReadSession(expectedRequest)).thenReturn(builder.build());
    BigQueryStorageQuerySource<TableRow> querySource = BigQueryStorageQuerySource.create(stepUuid, ValueProvider.StaticValueProvider.of(encodedQuery), /* flattenResults = */
    true, /* useLegacySql = */
    true, /* priority = */
    QueryPriority.BATCH, /* location = */
    null, /* queryTempDataset = */
    null, /* kmsKey = */
    null, null, new TableRowParser(), TableRowJsonCoder.of(), new FakeBigQueryServices().withDatasetService(fakeDatasetService).withJobService(fakeJobService).withStorageClient(fakeStorageClient));
    List<? extends BoundedSource<TableRow>> sources = querySource.split(1024, options);
    assertEquals(1024, sources.size());
}
Also used : JobStatistics(com.google.api.services.bigquery.model.JobStatistics) JobStatistics2(com.google.api.services.bigquery.model.JobStatistics2) Table(com.google.api.services.bigquery.model.Table) TableSchema(com.google.api.services.bigquery.model.TableSchema) AvroSchema(com.google.cloud.bigquery.storage.v1.AvroSchema) TableSchema(com.google.api.services.bigquery.model.TableSchema) Schema(org.apache.avro.Schema) TableFieldSchema(com.google.api.services.bigquery.model.TableFieldSchema) ReadSession(com.google.cloud.bigquery.storage.v1.ReadSession) StorageClient(org.apache.beam.sdk.io.gcp.bigquery.BigQueryServices.StorageClient) ByteString(com.google.protobuf.ByteString) TableFieldSchema(com.google.api.services.bigquery.model.TableFieldSchema) TableRowParser(org.apache.beam.sdk.io.gcp.bigquery.BigQueryIO.TableRowParser) TableReference(com.google.api.services.bigquery.model.TableReference) BigQueryResourceNaming.createTempTableReference(org.apache.beam.sdk.io.gcp.bigquery.BigQueryResourceNaming.createTempTableReference) TableRow(com.google.api.services.bigquery.model.TableRow) FakeBigQueryServices(org.apache.beam.sdk.io.gcp.testing.FakeBigQueryServices) CreateReadSessionRequest(com.google.cloud.bigquery.storage.v1.CreateReadSessionRequest) Test(org.junit.Test)

Example 34 with TableRowParser

use of org.apache.beam.sdk.io.gcp.bigquery.BigQueryIO.TableRowParser in project beam by apache.

the class BigQueryIOStorageQueryTest method testQuerySourceEstimatedSize.

@Test
public void testQuerySourceEstimatedSize() throws Exception {
    String fakeQuery = "fake query text";
    fakeJobService.expectDryRunQuery(options.getProject(), fakeQuery, new JobStatistics().setQuery(new JobStatistics2().setTotalBytesProcessed(125L)));
    BigQueryStorageQuerySource<TableRow> querySource = BigQueryStorageQuerySource.create(/* stepUuid = */
    "stepUuid", ValueProvider.StaticValueProvider.of(fakeQuery), /* flattenResults = */
    true, /* useLegacySql = */
    true, /* priority = */
    QueryPriority.INTERACTIVE, /* location = */
    null, /* queryTempDataset = */
    null, /* kmsKey = */
    null, null, new TableRowParser(), TableRowJsonCoder.of(), fakeBigQueryServices);
    assertEquals(125L, querySource.getEstimatedSizeBytes(options));
}
Also used : JobStatistics(com.google.api.services.bigquery.model.JobStatistics) JobStatistics2(com.google.api.services.bigquery.model.JobStatistics2) TableRow(com.google.api.services.bigquery.model.TableRow) ByteString(com.google.protobuf.ByteString) TableRowParser(org.apache.beam.sdk.io.gcp.bigquery.BigQueryIO.TableRowParser) Test(org.junit.Test)

Aggregations

TableRow (com.google.api.services.bigquery.model.TableRow)34 TableRowParser (org.apache.beam.sdk.io.gcp.bigquery.BigQueryIO.TableRowParser)34 Test (org.junit.Test)31 FakeBigQueryServices (org.apache.beam.sdk.io.gcp.testing.FakeBigQueryServices)28 StorageClient (org.apache.beam.sdk.io.gcp.bigquery.BigQueryServices.StorageClient)22 ReadSession (com.google.cloud.bigquery.storage.v1.ReadSession)14 ReadRowsResponse (com.google.cloud.bigquery.storage.v1.ReadRowsResponse)13 ByteString (com.google.protobuf.ByteString)13 TableReference (com.google.api.services.bigquery.model.TableReference)12 Table (com.google.api.services.bigquery.model.Table)11 CreateReadSessionRequest (com.google.cloud.bigquery.storage.v1.CreateReadSessionRequest)8 ReadRowsRequest (com.google.cloud.bigquery.storage.v1.ReadRowsRequest)6 JobStatistics (com.google.api.services.bigquery.model.JobStatistics)5 JobStatistics2 (com.google.api.services.bigquery.model.JobStatistics2)5 BigQueryResourceNaming.createTempTableReference (org.apache.beam.sdk.io.gcp.bigquery.BigQueryResourceNaming.createTempTableReference)4 TableSchema (com.google.api.services.bigquery.model.TableSchema)3 GenericRecord (org.apache.avro.generic.GenericRecord)3 FailedPreconditionException (com.google.api.gax.rpc.FailedPreconditionException)2 TableFieldSchema (com.google.api.services.bigquery.model.TableFieldSchema)2 AvroSchema (com.google.cloud.bigquery.storage.v1.AvroSchema)2