Examples with TableSchema com.google.api.services.bigquery.model.TableSchema used on opensource projects

Example 6 with TableSchema

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

the class FilterExamples method main.

public static void main(String[] args) throws Exception {
    Options options = PipelineOptionsFactory.fromArgs(args).withValidation().as(Options.class);
    Pipeline p = Pipeline.create(options);
    TableSchema schema = buildWeatherSchemaProjection();
    p.apply(BigQueryIO.read().from(options.getInput())).apply(ParDo.of(new ProjectionFn())).apply(new BelowGlobalMean(options.getMonthFilter())).apply(BigQueryIO.writeTableRows().to(options.getOutput()).withSchema(schema).withCreateDisposition(BigQueryIO.Write.CreateDisposition.CREATE_IF_NEEDED).withWriteDisposition(BigQueryIO.Write.WriteDisposition.WRITE_TRUNCATE));
    p.run().waitUntilFinish();
}

Example 7 with TableSchema

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

the class BigQueryAvroUtilsTest method testConvertGenericRecordToTableRow.

@Test
public void testConvertGenericRecordToTableRow() throws Exception {
    TableSchema tableSchema = new TableSchema();
    tableSchema.setFields(fields);
    Schema avroSchema = AvroCoder.of(Bird.class).getSchema();
    {
        // Test nullable fields.
        GenericRecord record = new GenericData.Record(avroSchema);
        record.put("number", 5L);
        TableRow convertedRow = BigQueryAvroUtils.convertGenericRecordToTableRow(record, tableSchema);
        TableRow row = new TableRow().set("number", "5").set("associates", new ArrayList<TableRow>());
        assertEquals(row, convertedRow);
    }
    {
        // Test type conversion for:
        // INTEGER, FLOAT, TIMESTAMP, BOOLEAN, BYTES, DATE, DATETIME, TIME.
        GenericRecord record = new GenericData.Record(avroSchema);
        byte[] soundBytes = "chirp,chirp".getBytes();
        ByteBuffer soundByteBuffer = ByteBuffer.wrap(soundBytes);
        soundByteBuffer.rewind();
        record.put("number", 5L);
        record.put("quality", 5.0);
        record.put("birthday", 5L);
        record.put("flighted", Boolean.TRUE);
        record.put("sound", soundByteBuffer);
        record.put("anniversaryDate", new Utf8("2000-01-01"));
        record.put("anniversaryDatetime", new String("2000-01-01 00:00:00.000005"));
        record.put("anniversaryTime", new Utf8("00:00:00.000005"));
        TableRow convertedRow = BigQueryAvroUtils.convertGenericRecordToTableRow(record, tableSchema);
        TableRow row = new TableRow().set("number", "5").set("birthday", "1970-01-01 00:00:00.000005 UTC").set("quality", 5.0).set("associates", new ArrayList<TableRow>()).set("flighted", Boolean.TRUE).set("sound", BaseEncoding.base64().encode(soundBytes)).set("anniversaryDate", "2000-01-01").set("anniversaryDatetime", "2000-01-01 00:00:00.000005").set("anniversaryTime", "00:00:00.000005");
        assertEquals(row, convertedRow);
    }
    {
        // Test repeated fields.
        Schema subBirdSchema = AvroCoder.of(Bird.SubBird.class).getSchema();
        GenericRecord nestedRecord = new GenericData.Record(subBirdSchema);
        nestedRecord.put("species", "other");
        GenericRecord record = new GenericData.Record(avroSchema);
        record.put("number", 5L);
        record.put("associates", Lists.<GenericRecord>newArrayList(nestedRecord));
        TableRow convertedRow = BigQueryAvroUtils.convertGenericRecordToTableRow(record, tableSchema);
        TableRow row = new TableRow().set("associates", Lists.<TableRow>newArrayList(new TableRow().set("species", "other"))).set("number", "5");
        assertEquals(row, convertedRow);
    }
}

Example 8 with TableSchema

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

the class BigQueryIOTest method testBigQueryQuerySourceInitSplit.

@Test
public void testBigQueryQuerySourceInitSplit() throws Exception {
    TableReference dryRunTable = new TableReference();
    Job queryJob = new Job();
    JobStatistics queryJobStats = new JobStatistics();
    JobStatistics2 queryStats = new JobStatistics2();
    queryStats.setReferencedTables(ImmutableList.of(dryRunTable));
    queryJobStats.setQuery(queryStats);
    queryJob.setStatus(new JobStatus()).setStatistics(queryJobStats);
    Job extractJob = new Job();
    JobStatistics extractJobStats = new JobStatistics();
    JobStatistics4 extractStats = new JobStatistics4();
    extractStats.setDestinationUriFileCounts(ImmutableList.of(1L));
    extractJobStats.setExtract(extractStats);
    extractJob.setStatus(new JobStatus()).setStatistics(extractJobStats);
    FakeJobService fakeJobService = new FakeJobService();
    FakeDatasetService fakeDatasetService = new FakeDatasetService();
    FakeBigQueryServices fakeBqServices = new FakeBigQueryServices().withJobService(fakeJobService).withDatasetService(fakeDatasetService);
    List<TableRow> expected = ImmutableList.of(new TableRow().set("name", "a").set("number", 1L), new TableRow().set("name", "b").set("number", 2L), new TableRow().set("name", "c").set("number", 3L), new TableRow().set("name", "d").set("number", 4L), new TableRow().set("name", "e").set("number", 5L), new TableRow().set("name", "f").set("number", 6L));
    PipelineOptions options = PipelineOptionsFactory.create();
    BigQueryOptions bqOptions = options.as(BigQueryOptions.class);
    bqOptions.setProject("project");
    String stepUuid = "testStepUuid";
    TableReference tempTableReference = createTempTableReference(bqOptions.getProject(), createJobIdToken(bqOptions.getJobName(), stepUuid));
    fakeDatasetService.createDataset(bqOptions.getProject(), tempTableReference.getDatasetId(), "", "");
    fakeDatasetService.createTable(new Table().setTableReference(tempTableReference).setSchema(new TableSchema().setFields(ImmutableList.of(new TableFieldSchema().setName("name").setType("STRING"), new TableFieldSchema().setName("number").setType("INTEGER")))));
    Path baseDir = Files.createTempDirectory(tempFolder, "testBigQueryQuerySourceInitSplit");
    String query = FakeBigQueryServices.encodeQuery(expected);
    BoundedSource<TableRow> bqSource = BigQueryQuerySource.create(stepUuid, StaticValueProvider.of(query), true, /* flattenResults */
    true, /* useLegacySql */
    fakeBqServices);
    options.setTempLocation(baseDir.toString());
    TableReference queryTable = new TableReference().setProjectId(bqOptions.getProject()).setDatasetId(tempTableReference.getDatasetId()).setTableId(tempTableReference.getTableId());
    fakeJobService.expectDryRunQuery(bqOptions.getProject(), query, new JobStatistics().setQuery(new JobStatistics2().setTotalBytesProcessed(100L).setReferencedTables(ImmutableList.of(queryTable))));
    List<TableRow> read = SourceTestUtils.readFromSource(bqSource, options);
    assertThat(read, containsInAnyOrder(Iterables.toArray(expected, TableRow.class)));
    SourceTestUtils.assertSplitAtFractionBehavior(bqSource, 2, 0.3, ExpectedSplitOutcome.MUST_BE_CONSISTENT_IF_SUCCEEDS, options);
    List<? extends BoundedSource<TableRow>> sources = bqSource.split(100, options);
    assertEquals(2, sources.size());
    BoundedSource<TableRow> actual = sources.get(0);
    assertThat(actual, CoreMatchers.instanceOf(TransformingSource.class));
}

Example 9 with TableSchema

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

the class BigQueryIOTest method testBigQueryTableSourceInitSplit.

@Test
public void testBigQueryTableSourceInitSplit() throws Exception {
    FakeDatasetService fakeDatasetService = new FakeDatasetService();
    FakeJobService fakeJobService = new FakeJobService();
    FakeBigQueryServices fakeBqServices = new FakeBigQueryServices().withJobService(fakeJobService).withDatasetService(fakeDatasetService);
    List<TableRow> expected = ImmutableList.of(new TableRow().set("name", "a").set("number", 1L), new TableRow().set("name", "b").set("number", 2L), new TableRow().set("name", "c").set("number", 3L), new TableRow().set("name", "d").set("number", 4L), new TableRow().set("name", "e").set("number", 5L), new TableRow().set("name", "f").set("number", 6L));
    TableReference table = BigQueryHelpers.parseTableSpec("project:data_set.table_name");
    fakeDatasetService.createDataset("project", "data_set", "", "");
    fakeDatasetService.createTable(new Table().setTableReference(table).setSchema(new TableSchema().setFields(ImmutableList.of(new TableFieldSchema().setName("name").setType("STRING"), new TableFieldSchema().setName("number").setType("INTEGER")))));
    fakeDatasetService.insertAll(table, expected, null);
    Path baseDir = Files.createTempDirectory(tempFolder, "testBigQueryTableSourceInitSplit");
    String stepUuid = "testStepUuid";
    BoundedSource<TableRow> bqSource = BigQueryTableSource.create(stepUuid, StaticValueProvider.of(table), fakeBqServices);
    PipelineOptions options = PipelineOptionsFactory.create();
    options.setTempLocation(baseDir.toString());
    BigQueryOptions bqOptions = options.as(BigQueryOptions.class);
    bqOptions.setProject("project");
    List<TableRow> read = SourceTestUtils.readFromSource(bqSource, options);
    assertThat(read, containsInAnyOrder(Iterables.toArray(expected, TableRow.class)));
    SourceTestUtils.assertSplitAtFractionBehavior(bqSource, 2, 0.3, ExpectedSplitOutcome.MUST_BE_CONSISTENT_IF_SUCCEEDS, options);
    List<? extends BoundedSource<TableRow>> sources = bqSource.split(100, options);
    assertEquals(2, sources.size());
    // Simulate a repeated call to split(), like a Dataflow worker will sometimes do.
    sources = bqSource.split(200, options);
    assertEquals(2, sources.size());
    BoundedSource<TableRow> actual = sources.get(0);
    assertThat(actual, CoreMatchers.instanceOf(TransformingSource.class));
    // A repeated call to split() should not have caused a duplicate extract job.
    assertEquals(1, fakeJobService.getNumExtractJobCalls());
}

Example 10 with TableSchema

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

the class BigQueryIOTest method testRetryPolicy.

@Test
public void testRetryPolicy() throws Exception {
    BigQueryOptions bqOptions = TestPipeline.testingPipelineOptions().as(BigQueryOptions.class);
    bqOptions.setProject("project-id");
    bqOptions.setTempLocation(testFolder.newFolder("BigQueryIOTest").getAbsolutePath());
    FakeDatasetService datasetService = new FakeDatasetService();
    FakeBigQueryServices fakeBqServices = new FakeBigQueryServices().withJobService(new FakeJobService()).withDatasetService(datasetService);
    datasetService.createDataset("project-id", "dataset-id", "", "");
    TableRow row1 = new TableRow().set("name", "a").set("number", "1");
    TableRow row2 = new TableRow().set("name", "b").set("number", "2");
    TableRow row3 = new TableRow().set("name", "c").set("number", "3");
    TableDataInsertAllResponse.InsertErrors ephemeralError = new TableDataInsertAllResponse.InsertErrors().setErrors(ImmutableList.of(new ErrorProto().setReason("timeout")));
    TableDataInsertAllResponse.InsertErrors persistentError = new TableDataInsertAllResponse.InsertErrors().setErrors(ImmutableList.of(new ErrorProto().setReason("invalidQuery")));
    datasetService.failOnInsert(ImmutableMap.<TableRow, List<TableDataInsertAllResponse.InsertErrors>>of(row1, ImmutableList.of(ephemeralError, ephemeralError), row2, ImmutableList.of(ephemeralError, ephemeralError, persistentError)));
    Pipeline p = TestPipeline.create(bqOptions);
    PCollection<TableRow> failedRows = p.apply(Create.of(row1, row2, row3)).setIsBoundedInternal(IsBounded.UNBOUNDED).apply(BigQueryIO.writeTableRows().to("project-id:dataset-id.table-id").withCreateDisposition(CreateDisposition.CREATE_IF_NEEDED).withSchema(new TableSchema().setFields(ImmutableList.of(new TableFieldSchema().setName("name").setType("STRING"), new TableFieldSchema().setName("number").setType("INTEGER")))).withFailedInsertRetryPolicy(InsertRetryPolicy.retryTransientErrors()).withTestServices(fakeBqServices).withoutValidation()).getFailedInserts();
    // row2 finally fails with a non-retryable error, so we expect to see it in the collection of
    // failed rows.
    PAssert.that(failedRows).containsInAnyOrder(row2);
    p.run();
    // Only row1 and row3 were successfully inserted.
    assertThat(datasetService.getAllRows("project-id", "dataset-id", "table-id"), containsInAnyOrder(row1, row3));
}