use of com.google.api.services.bigquery.model.JobStatistics2 in project beam by apache.
the class BigQueryTableRowIteratorTest method testReadFromQuery.
/**
* Verifies that when the query runs, the correct data is returned and the temporary dataset and
* table are both cleaned up.
*/
@Test
public void testReadFromQuery() throws IOException, InterruptedException {
// Mock job inserting.
Job dryRunJob = new Job().setStatistics(new JobStatistics().setQuery(new JobStatistics2().setReferencedTables(ImmutableList.of(new TableReference()))));
Job insertedJob = new Job().setJobReference(new JobReference());
when(mockJobsInsert.execute()).thenReturn(dryRunJob, insertedJob);
// Mock job polling.
JobStatus status = new JobStatus().setState("DONE");
JobConfigurationQuery resultQueryConfig = new JobConfigurationQuery().setDestinationTable(new TableReference().setProjectId("project").setDatasetId("tempdataset").setTableId("temptable"));
Job getJob = new Job().setJobReference(new JobReference()).setStatus(status).setConfiguration(new JobConfiguration().setQuery(resultQueryConfig));
when(mockJobsGet.execute()).thenReturn(getJob);
// Mock table schema fetch.
when(mockTablesGet.execute()).thenReturn(tableWithLocation(), tableWithBasicSchema());
byte[] photoBytes = "photograph".getBytes();
String photoBytesEncoded = BaseEncoding.base64().encode(photoBytes);
// Mock table data fetch.
when(mockTabledataList.execute()).thenReturn(rawDataList(rawRow("Arthur", 42, photoBytesEncoded, "2000-01-01", "2000-01-01 00:00:00.000005", "00:00:00.000005")));
// Run query and verify
String query = "SELECT name, count, photo, anniversary_date, " + "anniversary_datetime, anniversary_time from table";
JobConfigurationQuery queryConfig = new JobConfigurationQuery().setQuery(query);
try (BigQueryTableRowIterator iterator = BigQueryTableRowIterator.fromQuery(queryConfig, "project", mockClient)) {
iterator.open();
assertTrue(iterator.advance());
TableRow row = iterator.getCurrent();
assertTrue(row.containsKey("name"));
assertTrue(row.containsKey("answer"));
assertTrue(row.containsKey("photo"));
assertTrue(row.containsKey("anniversary_date"));
assertTrue(row.containsKey("anniversary_datetime"));
assertTrue(row.containsKey("anniversary_time"));
assertEquals("Arthur", row.get("name"));
assertEquals(42, row.get("answer"));
assertEquals(photoBytesEncoded, row.get("photo"));
assertEquals("2000-01-01", row.get("anniversary_date"));
assertEquals("2000-01-01 00:00:00.000005", row.get("anniversary_datetime"));
assertEquals("00:00:00.000005", row.get("anniversary_time"));
assertFalse(iterator.advance());
}
// Temp dataset created and later deleted.
verify(mockClient, times(2)).datasets();
verify(mockDatasets).insert(anyString(), any(Dataset.class));
verify(mockDatasetsInsert).execute();
verify(mockDatasets).delete(anyString(), anyString());
verify(mockDatasetsDelete).execute();
// Job inserted to run the query, polled once.
verify(mockClient, times(3)).jobs();
verify(mockJobs, times(2)).insert(anyString(), any(Job.class));
verify(mockJobsInsert, times(2)).execute();
verify(mockJobs).get(anyString(), anyString());
verify(mockJobsGet).execute();
// Temp table get after query finish, deleted after reading.
verify(mockClient, times(3)).tables();
verify(mockTables, times(2)).get(anyString(), anyString(), anyString());
verify(mockTablesGet, times(2)).execute();
verify(mockTables).delete(anyString(), anyString(), anyString());
verify(mockTablesDelete).execute();
// Table data read.
verify(mockClient).tabledata();
verify(mockTabledata).list("project", "tempdataset", "temptable");
verify(mockTabledataList).execute();
}
use of com.google.api.services.bigquery.model.JobStatistics2 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));
}
use of com.google.api.services.bigquery.model.JobStatistics2 in project beam by apache.
the class BigQueryIOTest method testBigQueryNoTableQuerySourceInitSplit.
@Test
public void testBigQueryNoTableQuerySourceInitSplit() 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);
FakeDatasetService datasetService = new FakeDatasetService();
FakeJobService jobService = new FakeJobService();
FakeBigQueryServices fakeBqServices = new FakeBigQueryServices().withJobService(jobService).withDatasetService(datasetService);
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));
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));
datasetService.createDataset(tempTableReference.getProjectId(), tempTableReference.getDatasetId(), "", "");
Table table = new Table().setTableReference(tempTableReference).setSchema(new TableSchema().setFields(ImmutableList.of(new TableFieldSchema().setName("name").setType("STRING"), new TableFieldSchema().setName("number").setType("INTEGER"))));
datasetService.createTable(table);
String query = FakeBigQueryServices.encodeQuery(expected);
jobService.expectDryRunQuery("project", query, new JobStatistics().setQuery(new JobStatistics2().setTotalBytesProcessed(100L).setReferencedTables(ImmutableList.of(table.getTableReference()))));
Path baseDir = Files.createTempDirectory(tempFolder, "testBigQueryNoTableQuerySourceInitSplit");
BoundedSource<TableRow> bqSource = BigQueryQuerySource.create(stepUuid, StaticValueProvider.of(query), true, /* flattenResults */
true, /* useLegacySql */
fakeBqServices);
options.setTempLocation(baseDir.toString());
List<TableRow> read = convertBigDecimaslToLong(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));
}
use of com.google.api.services.bigquery.model.JobStatistics2 in project beam by apache.
the class BigQueryIOReadTest method testBigQueryQuerySourceInitSplit_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 testBigQueryQuerySourceInitSplit_NoReferencedTables() throws Exception {
PipelineOptions options = PipelineOptionsFactory.create();
BigQueryOptions bqOptions = options.as(BigQueryOptions.class);
bqOptions.setProject("project");
Table queryResultTable = new Table().setSchema(new TableSchema().setFields(ImmutableList.of(new TableFieldSchema().setName("name").setType("STRING"), new TableFieldSchema().setName("number").setType("INTEGER"))));
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));
String encodedQuery = FakeBigQueryServices.encodeQueryResult(queryResultTable, expected);
String stepUuid = "testStepUuid";
fakeJobService.expectDryRunQuery(bqOptions.getProject(), encodedQuery, new JobStatistics().setQuery(new JobStatistics2().setTotalBytesProcessed(100L).setReferencedTables(ImmutableList.of())));
BoundedSource<TableRow> bqSource = BigQueryQuerySourceDef.create(fakeBqServices, ValueProvider.StaticValueProvider.of(encodedQuery), true, /* flattenResults */
true, /* useLegacySql */
QueryPriority.BATCH, null, null, null).toSource(stepUuid, TableRowJsonCoder.of(), BigQueryIO.TableRowParser.INSTANCE, false);
options.setTempLocation(testFolder.getRoot().getAbsolutePath());
List<TableRow> read = convertStringsToLong(SourceTestUtils.readFromSplitsOfSource(bqSource, 0L, /* ignored */
options));
assertThat(read, containsInAnyOrder(Iterables.toArray(expected, TableRow.class)));
List<? extends BoundedSource<TableRow>> sources = bqSource.split(100, options);
assertEquals(2, sources.size());
}
use of com.google.api.services.bigquery.model.JobStatistics2 in project beam by apache.
the class BigQueryIOReadTest method testBigQueryQuerySourceInitSplit.
@Test
public void testBigQueryQuerySourceInitSplit() throws Exception {
PipelineOptions options = PipelineOptionsFactory.create();
BigQueryOptions bqOptions = options.as(BigQueryOptions.class);
bqOptions.setProject("project");
TableReference sourceTableRef = BigQueryHelpers.parseTableSpec("project:dataset.table");
fakeDatasetService.createDataset(sourceTableRef.getProjectId(), sourceTableRef.getDatasetId(), "asia-northeast1", "Fake plastic tree^H^H^H^Htables", null);
fakeDatasetService.createTable(new Table().setTableReference(sourceTableRef).setLocation("asia-northeast1"));
Table queryResultTable = new Table().setSchema(new TableSchema().setFields(ImmutableList.of(new TableFieldSchema().setName("name").setType("STRING"), new TableFieldSchema().setName("number").setType("INTEGER"))));
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));
String encodedQuery = FakeBigQueryServices.encodeQueryResult(queryResultTable, expected);
String stepUuid = "testStepUuid";
TableReference tempTableReference = createTempTableReference(bqOptions.getProject(), BigQueryResourceNaming.createJobIdPrefix(options.getJobName(), stepUuid, JobType.QUERY), Optional.empty());
fakeJobService.expectDryRunQuery(bqOptions.getProject(), encodedQuery, new JobStatistics().setQuery(new JobStatistics2().setTotalBytesProcessed(100L).setReferencedTables(ImmutableList.of(sourceTableRef, tempTableReference))));
BoundedSource<TableRow> bqSource = BigQueryQuerySourceDef.create(fakeBqServices, ValueProvider.StaticValueProvider.of(encodedQuery), true, /* flattenResults */
true, /* useLegacySql */
QueryPriority.BATCH, null, null, null).toSource(stepUuid, TableRowJsonCoder.of(), BigQueryIO.TableRowParser.INSTANCE, false);
options.setTempLocation(testFolder.getRoot().getAbsolutePath());
List<TableRow> read = convertStringsToLong(SourceTestUtils.readFromSplitsOfSource(bqSource, 0L, /* ignored */
options));
assertThat(read, containsInAnyOrder(Iterables.toArray(expected, TableRow.class)));
List<? extends BoundedSource<TableRow>> sources = bqSource.split(100, options);
assertEquals(2, sources.size());
}
Aggregations