Search in sources :

Example 16 with RecordResponse

use of com.amazonaws.athena.connector.lambda.records.RecordResponse in project aws-athena-query-federation by awslabs.

the class DynamoDBRecordHandlerTest method testReadScanSplit.

@Test
public void testReadScanSplit() throws Exception {
    Split split = Split.newBuilder(SPILL_LOCATION, keyFactory.create()).add(TABLE_METADATA, TEST_TABLE).add(SEGMENT_ID_PROPERTY, "0").add(SEGMENT_COUNT_METADATA, "1").build();
    ReadRecordsRequest request = new ReadRecordsRequest(TEST_IDENTITY, TEST_CATALOG_NAME, TEST_QUERY_ID, TEST_TABLE_NAME, schema, split, new Constraints(ImmutableMap.of()), // too big to spill
    100_000_000_000L, 100_000_000_000L);
    RecordResponse rawResponse = handler.doReadRecords(allocator, request);
    assertTrue(rawResponse instanceof ReadRecordsResponse);
    ReadRecordsResponse response = (ReadRecordsResponse) rawResponse;
    logger.info("testReadScanSplit: rows[{}]", response.getRecordCount());
    assertEquals(1000, response.getRecords().getRowCount());
    logger.info("testReadScanSplit: {}", BlockUtils.rowToString(response.getRecords(), 0));
}
Also used : ReadRecordsRequest(com.amazonaws.athena.connector.lambda.records.ReadRecordsRequest) Constraints(com.amazonaws.athena.connector.lambda.domain.predicate.Constraints) ReadRecordsResponse(com.amazonaws.athena.connector.lambda.records.ReadRecordsResponse) RecordResponse(com.amazonaws.athena.connector.lambda.records.RecordResponse) Split(com.amazonaws.athena.connector.lambda.domain.Split) Test(org.junit.Test)

Example 17 with RecordResponse

use of com.amazonaws.athena.connector.lambda.records.RecordResponse in project aws-athena-query-federation by awslabs.

the class ElasticsearchRecordHandlerTest method doReadRecordsNoSpill.

@Test
public void doReadRecordsNoSpill() throws Exception {
    logger.info("doReadRecordsNoSpill: enter");
    SearchHit[] searchHit = new SearchHit[2];
    searchHit[0] = new SearchHit(1);
    searchHit[1] = new SearchHit(2);
    SearchHits searchHits = new SearchHits(searchHit, new TotalHits(2, TotalHits.Relation.EQUAL_TO), 4);
    when(mockResponse.getHits()).thenReturn(searchHits);
    Map<String, ValueSet> constraintsMap = new HashMap<>();
    constraintsMap.put("myshort", SortedRangeSet.copyOf(Types.MinorType.SMALLINT.getType(), ImmutableList.of(Range.range(allocator, Types.MinorType.SMALLINT.getType(), (short) 1955, false, (short) 1972, true)), false));
    List<String> expectedProjection = new ArrayList<>();
    mapping.getFields().forEach(field -> expectedProjection.add(field.getName()));
    String expectedPredicate = "(_exists_:myshort) AND myshort:({1955 TO 1972])";
    ReadRecordsRequest request = new ReadRecordsRequest(fakeIdentity(), "elasticsearch", "queryId-" + System.currentTimeMillis(), new TableName("movies", "mishmash"), mapping, split, new Constraints(constraintsMap), // 100GB don't expect this to spill
    100_000_000_000L, 100_000_000_000L);
    RecordResponse rawResponse = handler.doReadRecords(allocator, request);
    // Capture the SearchRequest object from the call to client.getDocuments().
    // The former contains information such as the projection and predicate.
    ArgumentCaptor<SearchRequest> argumentCaptor = ArgumentCaptor.forClass(SearchRequest.class);
    verify(mockClient).getDocuments(argumentCaptor.capture());
    SearchRequest searchRequest = argumentCaptor.getValue();
    // Get the actual projection and compare to the expected one.
    List<String> actualProjection = ImmutableList.copyOf(searchRequest.source().fetchSource().includes());
    assertEquals("Projections do not match", expectedProjection, actualProjection);
    // Get the actual predicate and compare to the expected one.
    String actualPredicate = searchRequest.source().query().queryName();
    assertEquals("Predicates do not match", expectedPredicate, actualPredicate);
    assertTrue(rawResponse instanceof ReadRecordsResponse);
    ReadRecordsResponse response = (ReadRecordsResponse) rawResponse;
    logger.info("doReadRecordsNoSpill: rows[{}]", response.getRecordCount());
    assertEquals(2, response.getRecords().getRowCount());
    for (int i = 0; i < response.getRecords().getRowCount(); ++i) {
        logger.info("doReadRecordsNoSpill - Row: {}, {}", i, BlockUtils.rowToString(response.getRecords(), i));
    }
    logger.info("doReadRecordsNoSpill: exit");
}
Also used : TotalHits(org.apache.lucene.search.TotalHits) SearchRequest(org.elasticsearch.action.search.SearchRequest) SearchHit(org.elasticsearch.search.SearchHit) HashMap(java.util.HashMap) ReadRecordsResponse(com.amazonaws.athena.connector.lambda.records.ReadRecordsResponse) RemoteReadRecordsResponse(com.amazonaws.athena.connector.lambda.records.RemoteReadRecordsResponse) ArrayList(java.util.ArrayList) Mockito.anyString(org.mockito.Mockito.anyString) RecordResponse(com.amazonaws.athena.connector.lambda.records.RecordResponse) TableName(com.amazonaws.athena.connector.lambda.domain.TableName) ReadRecordsRequest(com.amazonaws.athena.connector.lambda.records.ReadRecordsRequest) Constraints(com.amazonaws.athena.connector.lambda.domain.predicate.Constraints) SearchHits(org.elasticsearch.search.SearchHits) ValueSet(com.amazonaws.athena.connector.lambda.domain.predicate.ValueSet) Test(org.junit.Test)

Example 18 with RecordResponse

use of com.amazonaws.athena.connector.lambda.records.RecordResponse in project aws-athena-query-federation by awslabs.

the class DynamoDBRecordHandlerTest method testReadQuerySplit.

@Test
public void testReadQuerySplit() throws Exception {
    Map<String, String> expressionNames = ImmutableMap.of("#col_1", "col_1");
    Map<String, AttributeValue> expressionValues = ImmutableMap.of(":v0", toAttributeValue(1));
    Split split = Split.newBuilder(SPILL_LOCATION, keyFactory.create()).add(TABLE_METADATA, TEST_TABLE).add(HASH_KEY_NAME_METADATA, "col_0").add("col_0", toJsonString(toAttributeValue("test_str_0"))).add(RANGE_KEY_FILTER_METADATA, "#col_1 >= :v0").add(EXPRESSION_NAMES_METADATA, toJsonString(expressionNames)).add(EXPRESSION_VALUES_METADATA, toJsonString(expressionValues)).build();
    ReadRecordsRequest request = new ReadRecordsRequest(TEST_IDENTITY, TEST_CATALOG_NAME, TEST_QUERY_ID, TEST_TABLE_NAME, schema, split, new Constraints(ImmutableMap.of()), // too big to spill
    100_000_000_000L, 100_000_000_000L);
    RecordResponse rawResponse = handler.doReadRecords(allocator, request);
    assertTrue(rawResponse instanceof ReadRecordsResponse);
    ReadRecordsResponse response = (ReadRecordsResponse) rawResponse;
    logger.info("testReadQuerySplit: rows[{}]", response.getRecordCount());
    assertEquals(2, response.getRecords().getRowCount());
    logger.info("testReadQuerySplit: {}", BlockUtils.rowToString(response.getRecords(), 0));
}
Also used : ItemUtils.toAttributeValue(com.amazonaws.services.dynamodbv2.document.ItemUtils.toAttributeValue) AttributeValue(com.amazonaws.services.dynamodbv2.model.AttributeValue) ReadRecordsRequest(com.amazonaws.athena.connector.lambda.records.ReadRecordsRequest) Constraints(com.amazonaws.athena.connector.lambda.domain.predicate.Constraints) ReadRecordsResponse(com.amazonaws.athena.connector.lambda.records.ReadRecordsResponse) Jackson.toJsonString(com.amazonaws.util.json.Jackson.toJsonString) RecordResponse(com.amazonaws.athena.connector.lambda.records.RecordResponse) Split(com.amazonaws.athena.connector.lambda.domain.Split) Test(org.junit.Test)

Example 19 with RecordResponse

use of com.amazonaws.athena.connector.lambda.records.RecordResponse in project aws-athena-query-federation by awslabs.

the class RecordHandler method doHandleRequest.

protected final void doHandleRequest(BlockAllocator allocator, ObjectMapper objectMapper, RecordRequest req, OutputStream outputStream) throws Exception {
    logger.info("doHandleRequest: request[{}]", req);
    RecordRequestType type = req.getRequestType();
    switch(type) {
        case READ_RECORDS:
            try (RecordResponse response = doReadRecords(allocator, (ReadRecordsRequest) req)) {
                logger.info("doHandleRequest: response[{}]", response);
                assertNotNull(response);
                objectMapper.writeValue(outputStream, response);
            }
            return;
        default:
            throw new IllegalArgumentException("Unknown request type " + type);
    }
}
Also used : RecordRequestType(com.amazonaws.athena.connector.lambda.records.RecordRequestType) RecordResponse(com.amazonaws.athena.connector.lambda.records.RecordResponse)

Example 20 with RecordResponse

use of com.amazonaws.athena.connector.lambda.records.RecordResponse in project foundry-athena-query-federation-connector by palantir.

the class FoundryRecordHandlerTest method testReadRecordsLargeBlock.

@Test
void testReadRecordsLargeBlock() throws Exception {
    mockS3();
    when(readRecordsRequest.getCatalogName()).thenReturn("catalog");
    List<List<Object>> data = ImmutableList.of(ImmutableList.of(1, 2, 3, 4, 5), ImmutableList.of(1.1, 2.2, 3.3, 4.4, 5.5));
    InputStream stream = ArrowUtils.writeToStream(TestConstants.SCHEMA, ImmutableList.of(ArrowUtils.createBatch(TestConstants.SCHEMA, data)));
    when(recordService.fetchSlice(any(), any())).thenReturn(stream);
    RecordResponse response = handler.doReadRecords(allocator, readRecordsRequest);
    assetRoundTripSerializable(response);
    assertThat(response).isInstanceOf(RemoteReadRecordsResponse.class);
    assertThat(s3).hasSize(1);
    List<byte[]> blobs = new ArrayList<>(s3.values());
    assertThatSpillLocationContainsData(blobs.get(0), data);
}
Also used : InputStream(java.io.InputStream) ArrayList(java.util.ArrayList) ArrayList(java.util.ArrayList) ImmutableList(com.google.common.collect.ImmutableList) List(java.util.List) RecordResponse(com.amazonaws.athena.connector.lambda.records.RecordResponse) Test(org.junit.jupiter.api.Test)

Aggregations

RecordResponse (com.amazonaws.athena.connector.lambda.records.RecordResponse)37 Constraints (com.amazonaws.athena.connector.lambda.domain.predicate.Constraints)33 ReadRecordsRequest (com.amazonaws.athena.connector.lambda.records.ReadRecordsRequest)33 Test (org.junit.Test)32 ReadRecordsResponse (com.amazonaws.athena.connector.lambda.records.ReadRecordsResponse)27 ValueSet (com.amazonaws.athena.connector.lambda.domain.predicate.ValueSet)24 HashMap (java.util.HashMap)24 Matchers.anyString (org.mockito.Matchers.anyString)22 RemoteReadRecordsResponse (com.amazonaws.athena.connector.lambda.records.RemoteReadRecordsResponse)21 S3SpillLocation (com.amazonaws.athena.connector.lambda.domain.spill.S3SpillLocation)20 TableName (com.amazonaws.athena.connector.lambda.domain.TableName)19 Split (com.amazonaws.athena.connector.lambda.domain.Split)18 ArrayList (java.util.ArrayList)18 InvocationOnMock (org.mockito.invocation.InvocationOnMock)14 Block (com.amazonaws.athena.connector.lambda.data.Block)13 SpillLocation (com.amazonaws.athena.connector.lambda.domain.spill.SpillLocation)11 Schema (org.apache.arrow.vector.types.pojo.Schema)10 EquatableValueSet (com.amazonaws.athena.connector.lambda.domain.predicate.EquatableValueSet)9 InputStream (java.io.InputStream)7 List (java.util.List)7