Search in sources :

Example 36 with TableReference

use of com.google.api.services.bigquery.model.TableReference in project beam by apache.

the class BigQueryIOWriteTest method testWriteValidatesDataset.

@Test
public void testWriteValidatesDataset() throws Exception {
    TableReference tableRef = new TableReference();
    tableRef.setDatasetId("somedataset");
    tableRef.setTableId("sometable");
    PCollection<TableRow> tableRows;
    if (useStreaming) {
        tableRows = p.apply(GenerateSequence.from(0)).apply(MapElements.via(new SimpleFunction<Long, TableRow>() {

            @Override
            public TableRow apply(Long input) {
                return null;
            }
        })).setCoder(TableRowJsonCoder.of());
    } else {
        tableRows = p.apply(Create.empty(TableRowJsonCoder.of()));
    }
    thrown.expect(RuntimeException.class);
    // Message will be one of following depending on the execution environment.
    thrown.expectMessage(Matchers.either(Matchers.containsString("Unable to confirm BigQuery dataset presence")).or(Matchers.containsString("BigQuery dataset not found for table")));
    tableRows.apply(BigQueryIO.writeTableRows().to(tableRef).withCreateDisposition(BigQueryIO.Write.CreateDisposition.CREATE_IF_NEEDED).withSchema(new TableSchema()).withTestServices(fakeBqServices));
    p.run();
}
Also used : TableReference(com.google.api.services.bigquery.model.TableReference) TableSchema(com.google.api.services.bigquery.model.TableSchema) TableRow(com.google.api.services.bigquery.model.TableRow) SimpleFunction(org.apache.beam.sdk.transforms.SimpleFunction) Test(org.junit.Test)

Example 37 with TableReference

use of com.google.api.services.bigquery.model.TableReference in project beam by apache.

the class BigQueryIOWriteTest method testBuildWriteWithTableReference.

@Test
public void testBuildWriteWithTableReference() {
    TableReference table = new TableReference().setProjectId("foo.com:project").setDatasetId("somedataset").setTableId("sometable");
    BigQueryIO.Write<TableRow> write = BigQueryIO.writeTableRows().to(table);
    assertEquals("foo.com:project", write.getTable().get().getProjectId());
    assertEquals("somedataset", write.getTable().get().getDatasetId());
    assertEquals("sometable", write.getTable().get().getTableId());
}
Also used : TableReference(com.google.api.services.bigquery.model.TableReference) TableRow(com.google.api.services.bigquery.model.TableRow) Test(org.junit.Test)

Example 38 with TableReference

use of com.google.api.services.bigquery.model.TableReference in project beam by apache.

the class BigQueryIOStorageReadTest method testReadFromBigQueryIO.

@Test
public void testReadFromBigQueryIO() 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(10L).setSchema(TABLE_SCHEMA);
    fakeDatasetService.createTable(table);
    CreateReadSessionRequest expectedCreateReadSessionRequest = CreateReadSessionRequest.newBuilder().setParent("projects/project-id").setReadSession(ReadSession.newBuilder().setTable("projects/foo.com:project/datasets/dataset/tables/table").setDataFormat(DataFormat.AVRO)).setMaxStreamCount(10).build();
    ReadSession readSession = ReadSession.newBuilder().setName("readSessionName").setAvroSchema(AvroSchema.newBuilder().setSchema(AVRO_SCHEMA_STRING)).addStreams(ReadStream.newBuilder().setName("streamName")).setDataFormat(DataFormat.AVRO).build();
    ReadRowsRequest expectedReadRowsRequest = ReadRowsRequest.newBuilder().setReadStream("streamName").build();
    List<GenericRecord> records = Lists.newArrayList(createRecord("A", 1, AVRO_SCHEMA), createRecord("B", 2, AVRO_SCHEMA), createRecord("C", 3, AVRO_SCHEMA), createRecord("D", 4, AVRO_SCHEMA));
    List<ReadRowsResponse> readRowsResponses = Lists.newArrayList(createResponse(AVRO_SCHEMA, records.subList(0, 2), 0.0, 0.50), createResponse(AVRO_SCHEMA, records.subList(2, 4), 0.5, 0.75));
    StorageClient fakeStorageClient = mock(StorageClient.class, withSettings().serializable());
    when(fakeStorageClient.createReadSession(expectedCreateReadSessionRequest)).thenReturn(readSession);
    when(fakeStorageClient.readRows(expectedReadRowsRequest, "")).thenReturn(new FakeBigQueryServerStream<>(readRowsResponses));
    PCollection<KV<String, Long>> output = p.apply(BigQueryIO.read(new ParseKeyValue()).from("foo.com:project:dataset.table").withMethod(Method.DIRECT_READ).withFormat(DataFormat.AVRO).withTestServices(new FakeBigQueryServices().withDatasetService(fakeDatasetService).withStorageClient(fakeStorageClient)));
    PAssert.that(output).containsInAnyOrder(ImmutableList.of(KV.of("A", 1L), KV.of("B", 2L), KV.of("C", 3L), KV.of("D", 4L)));
    p.run();
}
Also used : Table(com.google.api.services.bigquery.model.Table) ReadSession(com.google.cloud.bigquery.storage.v1.ReadSession) ReadRowsRequest(com.google.cloud.bigquery.storage.v1.ReadRowsRequest) StorageClient(org.apache.beam.sdk.io.gcp.bigquery.BigQueryServices.StorageClient) KV(org.apache.beam.sdk.values.KV) TableReference(com.google.api.services.bigquery.model.TableReference) ReadRowsResponse(com.google.cloud.bigquery.storage.v1.ReadRowsResponse) FakeBigQueryServices(org.apache.beam.sdk.io.gcp.testing.FakeBigQueryServices) GenericRecord(org.apache.avro.generic.GenericRecord) CreateReadSessionRequest(com.google.cloud.bigquery.storage.v1.CreateReadSessionRequest) Test(org.junit.Test)

Example 39 with TableReference

use of com.google.api.services.bigquery.model.TableReference in project beam by apache.

the class BigQueryIOStorageReadTest method testTableSourceInitialSplit_WithDefaultProject.

@Test
public void testTableSourceInitialSplit_WithDefaultProject() throws Exception {
    fakeDatasetService.createDataset("project-id", "dataset", "", "", null);
    TableReference tableRef = BigQueryHelpers.parseTableSpec("project-id:dataset.table");
    Table table = new Table().setTableReference(tableRef).setNumBytes(1024L * 1024L).setSchema(TABLE_SCHEMA);
    fakeDatasetService.createTable(table);
    CreateReadSessionRequest expectedRequest = CreateReadSessionRequest.newBuilder().setParent("projects/project-id").setReadSession(ReadSession.newBuilder().setTable("projects/project-id/datasets/dataset/tables/table")).setMaxStreamCount(1024).build();
    ReadSession.Builder builder = ReadSession.newBuilder().setAvroSchema(AvroSchema.newBuilder().setSchema(AVRO_SCHEMA_STRING)).setDataFormat(DataFormat.AVRO);
    for (int i = 0; i < 50; 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(BigQueryHelpers.parseTableSpec("dataset.table")), null, null, new TableRowParser(), TableRowJsonCoder.of(), new FakeBigQueryServices().withDatasetService(fakeDatasetService).withStorageClient(fakeStorageClient));
    List<? extends BoundedSource<TableRow>> sources = tableSource.split(1024L, options);
    assertEquals(50L, 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 40 with TableReference

use of com.google.api.services.bigquery.model.TableReference in project beam by apache.

the class BigQueryIOStorageReadTest method testTableSourceInitialSplit_EmptyTable.

@Test
public void testTableSourceInitialSplit_EmptyTable() 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(1024L * 1024L).setSchema(new TableSchema());
    fakeDatasetService.createTable(table);
    CreateReadSessionRequest expectedRequest = CreateReadSessionRequest.newBuilder().setParent("projects/project-id").setReadSession(ReadSession.newBuilder().setTable("projects/foo.com:project/datasets/dataset/tables/table")).setMaxStreamCount(1024).build();
    ReadSession emptyReadSession = ReadSession.newBuilder().build();
    StorageClient fakeStorageClient = mock(StorageClient.class);
    when(fakeStorageClient.createReadSession(expectedRequest)).thenReturn(emptyReadSession);
    BigQueryStorageTableSource<TableRow> tableSource = BigQueryStorageTableSource.create(ValueProvider.StaticValueProvider.of(tableRef), null, null, new TableRowParser(), TableRowJsonCoder.of(), new FakeBigQueryServices().withDatasetService(fakeDatasetService).withStorageClient(fakeStorageClient));
    List<? extends BoundedSource<TableRow>> sources = tableSource.split(1024L, options);
    assertTrue(sources.isEmpty());
}
Also used : TableReference(com.google.api.services.bigquery.model.TableReference) Table(com.google.api.services.bigquery.model.Table) TableSchema(com.google.api.services.bigquery.model.TableSchema) 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)

Aggregations

TableReference (com.google.api.services.bigquery.model.TableReference)139 Test (org.junit.Test)75 TableRow (com.google.api.services.bigquery.model.TableRow)68 Table (com.google.api.services.bigquery.model.Table)61 TableSchema (com.google.api.services.bigquery.model.TableSchema)36 DatasetServiceImpl (org.apache.beam.sdk.io.gcp.bigquery.BigQueryServicesImpl.DatasetServiceImpl)29 TableFieldSchema (com.google.api.services.bigquery.model.TableFieldSchema)22 TableDataInsertAllResponse (com.google.api.services.bigquery.model.TableDataInsertAllResponse)19 FakeBigQueryServices (org.apache.beam.sdk.io.gcp.testing.FakeBigQueryServices)19 MockSleeper (com.google.api.client.testing.util.MockSleeper)17 BigQueryHelpers.createTempTableReference (org.apache.beam.sdk.io.gcp.bigquery.BigQueryHelpers.createTempTableReference)17 FailsafeValueInSingleWindow (org.apache.beam.sdk.values.FailsafeValueInSingleWindow)16 JobStatus (com.google.api.services.bigquery.model.JobStatus)15 ReadSession (com.google.cloud.bigquery.storage.v1.ReadSession)15 BigQueryResourceNaming.createTempTableReference (org.apache.beam.sdk.io.gcp.bigquery.BigQueryResourceNaming.createTempTableReference)15 ErrorProto (com.google.api.services.bigquery.model.ErrorProto)14 JobStatistics (com.google.api.services.bigquery.model.JobStatistics)14 CreateReadSessionRequest (com.google.cloud.bigquery.storage.v1.CreateReadSessionRequest)14 ByteString (com.google.protobuf.ByteString)14 IOException (java.io.IOException)14