Examples with Namespace com.bakdata.conquery.models.worker.Namespace used on opensource projects

Example 26 with Namespace

use of com.bakdata.conquery.models.worker.Namespace in project conquery by bakdata.

the class ImportDeletionTest method execute.

@Override
public void execute(String name, TestConquery testConquery) throws Exception {
    final StandaloneSupport conquery = testConquery.getSupport(name);
    MetaStorage storage = conquery.getMetaStorage();
    final String testJson = In.resource("/tests/query/DELETE_IMPORT_TESTS/SIMPLE_TREECONCEPT_Query.test.json").withUTF8().readAll();
    final Dataset dataset = conquery.getDataset();
    final Namespace namespace = conquery.getNamespace();
    final ImportId importId = ImportId.Parser.INSTANCE.parse(dataset.getName(), "test_table2", "test_table2");
    final QueryTest test = (QueryTest) JsonIntegrationTest.readJson(dataset, testJson);
    // Manually import data, so we can do our own work.
    {
        ValidatorHelper.failOnError(log, conquery.getValidator().validate(test));
        importSecondaryIds(conquery, test.getContent().getSecondaryIds());
        conquery.waitUntilWorkDone();
        LoadingUtil.importTables(conquery, test.getContent().getTables());
        conquery.waitUntilWorkDone();
        LoadingUtil.importConcepts(conquery, test.getRawConcepts());
        conquery.waitUntilWorkDone();
        LoadingUtil.importTableContents(conquery, test.getContent().getTables());
        conquery.waitUntilWorkDone();
    }
    final Query query = IntegrationUtils.parseQuery(conquery, test.getRawQuery());
    final int nImports = namespace.getStorage().getAllImports().size();
    // State before deletion.
    {
        log.info("Checking state before deletion");
        // Must contain the import.
        assertThat(namespace.getStorage().getAllImports()).filteredOn(imp -> imp.getId().equals(importId)).isNotEmpty();
        assertThat(namespace.getStorage().getCentralRegistry().getOptional(importId)).isNotEmpty();
        for (ShardNode node : conquery.getShardNodes()) {
            for (Worker worker : node.getWorkers().getWorkers().values()) {
                if (!worker.getInfo().getDataset().equals(dataset.getId())) {
                    continue;
                }
                final ModificationShieldedWorkerStorage workerStorage = worker.getStorage();
                assertThat(workerStorage.getAllCBlocks()).describedAs("CBlocks for Worker %s", worker.getInfo().getId()).filteredOn(block -> block.getBucket().getId().getDataset().equals(dataset.getId())).isNotEmpty();
                assertThat(workerStorage.getAllBuckets()).filteredOn(bucket -> bucket.getId().getDataset().equals(dataset.getId())).describedAs("Buckets for Worker %s", worker.getInfo().getId()).isNotEmpty();
                // Must contain the import.
                assertThat(workerStorage.getImport(importId)).isNotNull();
            }
        }
        log.info("Executing query before deletion");
        IntegrationUtils.assertQueryResult(conquery, query, 2L, ExecutionState.DONE, conquery.getTestUser(), 201);
    }
    // Delete the import.
    {
        log.info("Issuing deletion of import {}", importId);
        final URI deleteImportUri = HierarchyHelper.hierarchicalPath(conquery.defaultAdminURIBuilder(), AdminTablesResource.class, "deleteImport").buildFromMap(Map.of(ResourceConstants.DATASET, conquery.getDataset().getId(), ResourceConstants.TABLE, importId.getTable(), ResourceConstants.IMPORT_ID, importId));
        final Response delete = conquery.getClient().target(deleteImportUri).request(MediaType.APPLICATION_JSON).delete();
        assertThat(delete.getStatusInfo().getFamily()).isEqualTo(Response.Status.Family.SUCCESSFUL);
        conquery.waitUntilWorkDone();
    }
    // State after deletion.
    {
        log.info("Checking state after deletion");
        // We have deleted an import now there should be one less!
        assertThat(namespace.getStorage().getAllImports().size()).isEqualTo(nImports - 1);
        // The deleted import should not be found.
        assertThat(namespace.getStorage().getAllImports()).filteredOn(imp -> imp.getId().equals(importId)).isEmpty();
        for (ShardNode node : conquery.getShardNodes()) {
            for (Worker worker : node.getWorkers().getWorkers().values()) {
                if (!worker.getInfo().getDataset().equals(dataset.getId())) {
                    continue;
                }
                final ModificationShieldedWorkerStorage workerStorage = worker.getStorage();
                // No bucket should be found referencing the import.
                assertThat(workerStorage.getAllBuckets()).describedAs("Buckets for Worker %s", worker.getInfo().getId()).filteredOn(bucket -> bucket.getImp().getId().equals(importId)).isEmpty();
                // No CBlock associated with import may exist
                assertThat(workerStorage.getAllCBlocks()).describedAs("CBlocks for Worker %s", worker.getInfo().getId()).filteredOn(cBlock -> cBlock.getBucket().getId().getImp().equals(importId)).isEmpty();
                // Import should not exists anymore
                assertThat(workerStorage.getImport(importId)).describedAs("Import for Worker %s", worker.getInfo().getId()).isNull();
            }
        }
        log.info("Executing query after deletion");
        // Issue a query and assert that it has less content.
        IntegrationUtils.assertQueryResult(conquery, query, 1L, ExecutionState.DONE, conquery.getTestUser(), 201);
    }
    conquery.waitUntilWorkDone();
    // Load more data under the same name into the same table, with only the deleted import/table
    {
        // only import the deleted import/table
        final RequiredTable import2Table = test.getContent().getTables().stream().filter(table -> table.getName().equalsIgnoreCase(importId.getTable().getTable())).findFirst().orElseThrow();
        final ResourceFile csv = import2Table.getCsv();
        final String path = csv.getPath();
        // copy csv to tmp folder
        // Content 2.2 contains an extra entry of a value that hasn't been seen before.
        FileUtils.copyInputStreamToFile(In.resource(path.substring(0, path.lastIndexOf('/')) + "/" + "content2.2.csv").asStream(), new File(conquery.getTmpDir(), csv.getName()));
        File descriptionFile = new File(conquery.getTmpDir(), import2Table.getName() + ConqueryConstants.EXTENSION_DESCRIPTION);
        File preprocessedFile = new File(conquery.getTmpDir(), import2Table.getName() + ConqueryConstants.EXTENSION_PREPROCESSED);
        // create import descriptor
        TableImportDescriptor desc = new TableImportDescriptor();
        desc.setName(import2Table.getName());
        desc.setTable(import2Table.getName());
        TableInputDescriptor input = new TableInputDescriptor();
        {
            input.setPrimary(import2Table.getPrimaryColumn().createOutput());
            input.setSourceFile(import2Table.getCsv().getName());
            input.setOutput(new OutputDescription[import2Table.getColumns().length]);
            for (int i = 0; i < import2Table.getColumns().length; i++) {
                input.getOutput()[i] = import2Table.getColumns()[i].createOutput();
            }
        }
        desc.setInputs(new TableInputDescriptor[] { input });
        Jackson.MAPPER.writeValue(descriptionFile, desc);
        // preprocess
        conquery.preprocessTmp(conquery.getTmpDir(), List.of(descriptionFile));
        // import preprocessedFiles
        conquery.getDatasetsProcessor().addImport(conquery.getNamespace(), new GZIPInputStream(new FileInputStream(preprocessedFile)));
        conquery.waitUntilWorkDone();
    }
    // State after reimport.
    {
        log.info("Checking state after re-import");
        assertThat(namespace.getStorage().getAllImports().size()).isEqualTo(nImports);
        for (ShardNode node : conquery.getShardNodes()) {
            for (Worker worker : node.getWorkers().getWorkers().values()) {
                if (!worker.getInfo().getDataset().equals(dataset.getId())) {
                    continue;
                }
                final ModificationShieldedWorkerStorage workerStorage = worker.getStorage();
                assertThat(workerStorage.getAllBuckets()).describedAs("Buckets for Worker %s", worker.getInfo().getId()).filteredOn(bucket -> bucket.getImp().getId().equals(importId)).filteredOn(bucket -> bucket.getId().getDataset().equals(dataset.getId())).isNotEmpty();
            }
        }
        log.info("Executing query after re-import");
        // Issue a query and assert that it has the same content as the first time around.
        IntegrationUtils.assertQueryResult(conquery, query, 2L, ExecutionState.DONE, conquery.getTestUser(), 201);
    }
    // Finally, restart conquery and assert again, that the data is correct.
    {
        testConquery.shutdown();
        // restart
        testConquery.beforeAll();
        StandaloneSupport conquery2 = testConquery.openDataset(dataset.getId());
        log.info("Checking state after re-start");
        {
            assertThat(namespace.getStorage().getAllImports().size()).isEqualTo(2);
            for (ShardNode node : conquery2.getShardNodes()) {
                for (Worker worker : node.getWorkers().getWorkers().values()) {
                    if (!worker.getInfo().getDataset().equals(dataset.getId()))
                        continue;
                    final ModificationShieldedWorkerStorage workerStorage = worker.getStorage();
                    assertThat(workerStorage.getAllBuckets()).describedAs("Buckets for Worker %s", worker.getInfo().getId()).filteredOn(bucket -> bucket.getId().getDataset().equals(dataset.getId())).filteredOn(bucket -> bucket.getImp().getId().equals(importId)).isNotEmpty();
                }
            }
            log.info("Executing query after re-import");
            // Issue a query and assert that it has the same content as the first time around.
            IntegrationUtils.assertQueryResult(conquery2, query, 2L, ExecutionState.DONE, conquery.getTestUser(), 201);
        }
    }
}

Example 27 with Namespace

use of com.bakdata.conquery.models.worker.Namespace in project conquery by bakdata.

the class TableDeletionTest method execute.

@Override
public void execute(String name, TestConquery testConquery) throws Exception {
    final StandaloneSupport conquery = testConquery.getSupport(name);
    final MetaStorage storage = conquery.getMetaStorage();
    final String testJson = In.resource("/tests/query/DELETE_IMPORT_TESTS/SIMPLE_TREECONCEPT_Query.test.json").withUTF8().readAll();
    final Dataset dataset = conquery.getDataset();
    final Namespace namespace = conquery.getNamespace();
    final TableId tableId = TableId.Parser.INSTANCE.parse(dataset.getName(), "test_table2");
    final QueryTest test = (QueryTest) JsonIntegrationTest.readJson(dataset, testJson);
    // Manually import data, so we can do our own work.
    {
        ValidatorHelper.failOnError(log, conquery.getValidator().validate(test));
        importSecondaryIds(conquery, test.getContent().getSecondaryIds());
        conquery.waitUntilWorkDone();
        LoadingUtil.importTables(conquery, test.getContent().getTables());
        conquery.waitUntilWorkDone();
        LoadingUtil.importConcepts(conquery, test.getRawConcepts());
        conquery.waitUntilWorkDone();
        LoadingUtil.importTableContents(conquery, test.getContent().getTables());
        conquery.waitUntilWorkDone();
    }
    final Query query = IntegrationUtils.parseQuery(conquery, test.getRawQuery());
    final int nImports = namespace.getStorage().getAllImports().size();
    // State before deletion.
    {
        log.info("Checking state before deletion");
        // Must contain the import.
        assertThat(namespace.getStorage().getCentralRegistry().getOptional(tableId)).isNotEmpty();
        for (ShardNode node : conquery.getShardNodes()) {
            for (Worker value : node.getWorkers().getWorkers().values()) {
                if (!value.getInfo().getDataset().equals(dataset.getId())) {
                    continue;
                }
                final ModificationShieldedWorkerStorage workerStorage = value.getStorage();
                assertThat(workerStorage.getAllCBlocks()).describedAs("CBlocks for Worker %s", value.getInfo().getId()).isNotEmpty();
                assertThat(workerStorage.getAllBuckets()).describedAs("Buckets for Worker %s", value.getInfo().getId()).isNotEmpty();
            }
        }
        log.info("Executing query before deletion");
        IntegrationUtils.assertQueryResult(conquery, query, 2L, ExecutionState.DONE, conquery.getTestUser(), 201);
    }
    // Delete the import.
    {
        log.info("Issuing deletion of import {}", tableId);
        // Delete the import via API.
        // But, we do not allow deletion of tables with associated connectors, so this should throw!
        final URI deleteTable = HierarchyHelper.hierarchicalPath(conquery.defaultAdminURIBuilder(), AdminTablesResource.class, "remove").buildFromMap(Map.of(ResourceConstants.DATASET, conquery.getDataset().getName(), ResourceConstants.TABLE, tableId.toString()));
        final Response failed = conquery.getClient().target(deleteTable).request().delete();
        assertThat(failed.getStatusInfo().getFamily()).isEqualTo(Response.Status.Family.CLIENT_ERROR);
        conquery.getDatasetsProcessor().deleteConcept(conquery.getNamespace().getStorage().getAllConcepts().iterator().next());
        Thread.sleep(100);
        conquery.waitUntilWorkDone();
        final Response success = conquery.getClient().target(deleteTable).request().delete();
        assertThat(success.getStatusInfo().getStatusCode()).isEqualTo(Response.Status.OK.getStatusCode());
        Thread.sleep(100);
        conquery.waitUntilWorkDone();
    }
    // State after deletion.
    {
        log.info("Checking state after deletion");
        // We have deleted an import now there should be two less!
        assertThat(namespace.getStorage().getAllImports().size()).isEqualTo(nImports - 1);
        // The deleted import should not be found.
        assertThat(namespace.getStorage().getAllImports()).filteredOn(imp -> imp.getId().getTable().equals(tableId)).isEmpty();
        for (ShardNode node : conquery.getShardNodes()) {
            for (Worker value : node.getWorkers().getWorkers().values()) {
                if (!value.getInfo().getDataset().equals(dataset.getId())) {
                    continue;
                }
                final ModificationShieldedWorkerStorage workerStorage = value.getStorage();
                // No bucket should be found referencing the import.
                assertThat(workerStorage.getAllBuckets()).describedAs("Buckets for Worker %s", value.getInfo().getId()).filteredOn(bucket -> bucket.getImp().getTable().getId().equals(tableId)).isEmpty();
                // No CBlock associated with import may exist
                assertThat(workerStorage.getAllCBlocks()).describedAs("CBlocks for Worker %s", value.getInfo().getId()).filteredOn(cBlock -> cBlock.getBucket().getImp().getTable().getId().equals(tableId)).isEmpty();
            }
        }
        log.info("Executing query after deletion");
        // Issue a query and asseert that it has less content.
        IntegrationUtils.assertQueryResult(conquery, query, 0L, ExecutionState.FAILED, conquery.getTestUser(), 400);
    }
    conquery.waitUntilWorkDone();
    // Load the same import into the same table, with only the deleted import/table
    {
        // only import the deleted import/table
        LoadingUtil.importTables(conquery, test.getContent().getTables().stream().filter(table -> table.getName().equalsIgnoreCase(tableId.getTable())).collect(Collectors.toList()));
        conquery.waitUntilWorkDone();
        LoadingUtil.importTableContents(conquery, test.getContent().getTables().stream().filter(table -> table.getName().equalsIgnoreCase(tableId.getTable())).collect(Collectors.toList()));
        conquery.waitUntilWorkDone();
        LoadingUtil.importConcepts(conquery, test.getRawConcepts());
        conquery.waitUntilWorkDone();
        assertThat(namespace.getStorage().getTable(tableId)).describedAs("Table after re-import.").isNotNull();
        for (ShardNode node : conquery.getShardNodes()) {
            for (Worker value : node.getWorkers().getWorkers().values()) {
                if (!value.getInfo().getDataset().equals(dataset.getId())) {
                    continue;
                }
                assertThat(value.getStorage().getCentralRegistry().resolve(tableId)).describedAs("Table in worker storage.").isNotNull();
            }
        }
    }
    // Test state after reimport.
    {
        log.info("Checking state after re-import");
        assertThat(namespace.getStorage().getAllImports().size()).isEqualTo(nImports);
        for (ShardNode node : conquery.getShardNodes()) {
            for (Worker value : node.getWorkers().getWorkers().values()) {
                if (!value.getInfo().getDataset().equals(dataset.getId())) {
                    continue;
                }
                final ModificationShieldedWorkerStorage workerStorage = value.getStorage();
                assertThat(workerStorage.getAllBuckets().stream().filter(bucket -> bucket.getImp().getTable().getId().equals(tableId))).describedAs("Buckets for Worker %s", value.getInfo().getId()).isNotEmpty();
            }
        }
        log.info("Executing query after re-import");
        // Issue a query and assert that it has the same content as the first time around.
        IntegrationUtils.assertQueryResult(conquery, query, 2L, ExecutionState.DONE, conquery.getTestUser(), 201);
    }
    // Finally, restart conquery and assert again, that the data is correct.
    {
        testConquery.shutdown();
        // restart
        testConquery.beforeAll();
        StandaloneSupport conquery2 = testConquery.openDataset(dataset.getId());
        log.info("Checking state after re-start");
        {
            assertThat(namespace.getStorage().getAllImports().size()).isEqualTo(2);
            for (ShardNode node : conquery2.getShardNodes()) {
                for (Worker value : node.getWorkers().getWorkers().values()) {
                    if (!value.getInfo().getDataset().equals(dataset.getId())) {
                        continue;
                    }
                    final ModificationShieldedWorkerStorage workerStorage = value.getStorage();
                    assertThat(workerStorage.getAllBuckets().stream().filter(bucket -> bucket.getImp().getTable().getId().equals(tableId))).describedAs("Buckets for Worker %s", value.getInfo().getId()).isNotEmpty();
                }
            }
            log.info("Executing query after re-import");
            // Issue a query and assert that it has the same content as the first time around.
            IntegrationUtils.assertQueryResult(conquery2, query, 2L, ExecutionState.DONE, conquery.getTestUser(), 201);
        }
    }
}

Example 28 with Namespace

use of com.bakdata.conquery.models.worker.Namespace in project conquery by bakdata.

the class TestConquery method createSupport.

private synchronized StandaloneSupport createSupport(DatasetId datasetId, String name) {
    DatasetRegistry datasets = standaloneCommand.getManager().getDatasetRegistry();
    Namespace ns = datasets.get(datasetId);
    assertThat(datasets.getShardNodes()).hasSize(2);
    // make tmp subdir and change cfg accordingly
    File localTmpDir = new File(tmpDir, "tmp_" + name);
    if (!localTmpDir.exists()) {
        if (!localTmpDir.mkdir()) {
            throw new IllegalStateException("Could not create directory for Support");
        }
    } else {
        log.info("Reusing existing folder {} for Support", localTmpDir.getPath());
    }
    ConqueryConfig localCfg = Cloner.clone(config, Map.of(Validator.class, standaloneCommand.getManager().getEnvironment().getValidator()), IntegrationTests.MAPPER);
    StandaloneSupport support = new StandaloneSupport(this, ns, ns.getStorage().getDataset(), localTmpDir, localCfg, standaloneCommand.getManager().getAdmin().getAdminProcessor(), standaloneCommand.getManager().getAdmin().getAdminDatasetProcessor(), // Getting the User from AuthorizationConfig
    testUser);
    Wait.builder().total(Duration.ofSeconds(5)).stepTime(Duration.ofMillis(5)).build().until(() -> ns.getWorkers().size() == ns.getNamespaces().getShardNodes().size());
    support.waitUntilWorkDone();
    openSupports.add(support);
    return support;
}