use of com.google.api.services.bigquery.model.JobStatistics in project beam by apache.
the class FakeJobService method runExtractJob.
private JobStatus runExtractJob(Job job, JobConfigurationExtract extract) throws InterruptedException, IOException {
TableReference sourceTable = extract.getSourceTable();
List<TableRow> rows = datasetService.getAllRows(sourceTable.getProjectId(), sourceTable.getDatasetId(), sourceTable.getTableId());
TableSchema schema = datasetService.getTable(sourceTable).getSchema();
List<Long> destinationFileCounts = Lists.newArrayList();
for (String destination : extract.getDestinationUris()) {
destinationFileCounts.add(writeRows(sourceTable.getTableId(), rows, schema, destination));
}
job.setStatistics(new JobStatistics().setExtract(new JobStatistics4().setDestinationUriFileCounts(destinationFileCounts)));
return new JobStatus().setState("DONE");
}
use of com.google.api.services.bigquery.model.JobStatistics in project beam by apache.
the class BigQueryTableRowIteratorTest method testReadFromQueryNoTables.
/**
* Verifies that queries that reference no data can be read.
*/
@Test
public void testReadFromQueryNoTables() throws IOException, InterruptedException {
// Mock job inserting.
Job dryRunJob = new Job().setStatistics(new JobStatistics().setQuery(new JobStatistics2()));
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(noTableQuerySchema());
byte[] photoBytes = "photograph".getBytes();
String photoBytesEncoded = BaseEncoding.base64().encode(photoBytes);
// Mock table data fetch.
when(mockTabledataList.execute()).thenReturn(rawDataList(rawRow("Arthur", 42, photoBytesEncoded)));
// Run query and verify
String query = String.format("SELECT \"Arthur\" as name, 42 as count, \"%s\" as photo", photoBytesEncoded);
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("count"));
assertTrue(row.containsKey("photo"));
assertEquals("Arthur", row.get("name"));
assertEquals(42, row.get("count"));
assertEquals(photoBytesEncoded, row.get("photo"));
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(2)).tables();
verify(mockTables, times(1)).get(anyString(), anyString(), anyString());
verify(mockTablesGet, times(1)).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.JobStatistics in project beam by apache.
the class BigQueryIO method getExtractFilePaths.
static List<ResourceId> getExtractFilePaths(String extractDestinationDir, Job extractJob) throws IOException {
JobStatistics jobStats = extractJob.getStatistics();
List<Long> counts = jobStats.getExtract().getDestinationUriFileCounts();
if (counts.size() != 1) {
String errorMessage = (counts.size() == 0 ? "No destination uri file count received." : String.format("More than one destination uri file count received. First two are %s, %s", counts.get(0), counts.get(1)));
throw new RuntimeException(errorMessage);
}
long filesCount = counts.get(0);
ImmutableList.Builder<ResourceId> paths = ImmutableList.builder();
ResourceId extractDestinationDirResourceId = FileSystems.matchNewResource(extractDestinationDir, true);
for (long i = 0; i < filesCount; ++i) {
ResourceId filePath = extractDestinationDirResourceId.resolve(String.format("%012d%s", i, ".avro"), ResolveOptions.StandardResolveOptions.RESOLVE_FILE);
paths.add(filePath);
}
return paths.build();
}
use of com.google.api.services.bigquery.model.JobStatistics in project beam by apache.
the class BigQueryQuerySource method dryRunQueryIfNeeded.
private synchronized JobStatistics dryRunQueryIfNeeded(BigQueryOptions bqOptions) throws InterruptedException, IOException {
if (dryRunJobStats.get() == null) {
JobStatistics jobStats = bqServices.getJobService(bqOptions).dryRunQuery(bqOptions.getProject(), createBasicQueryConfig());
dryRunJobStats.compareAndSet(null, jobStats);
}
return dryRunJobStats.get();
}
Aggregations