Search in sources :

Example 6 with BlockletBTreeIndex

use of org.apache.carbondata.format.BlockletBTreeIndex in project carbondata by apache.

the class CarbonMetadataUtil method setBlockletIndex.

private static void setBlockletIndex(FileFooter footer, List<BlockletInfoColumnar> listOfNodeInfo) {
    List<BlockletIndex> blockletIndexList = footer.getBlocklet_index_list();
    for (int i = 0; i < blockletIndexList.size(); i++) {
        BlockletBTreeIndex bTreeIndexList = blockletIndexList.get(i).getB_tree_index();
        BlockletMinMaxIndex minMaxIndexList = blockletIndexList.get(i).getMin_max_index();
        listOfNodeInfo.get(i).setStartKey(bTreeIndexList.getStart_key());
        listOfNodeInfo.get(i).setEndKey(bTreeIndexList.getEnd_key());
        byte[][] min = new byte[minMaxIndexList.getMin_values().size()][];
        byte[][] max = new byte[minMaxIndexList.getMax_values().size()][];
        for (int j = 0; j < minMaxIndexList.getMax_valuesSize(); j++) {
            min[j] = minMaxIndexList.getMin_values().get(j).array();
            max[j] = minMaxIndexList.getMax_values().get(j).array();
        }
        listOfNodeInfo.get(i).setColumnMaxData(max);
    }
}
Also used : BlockletBTreeIndex(org.apache.carbondata.format.BlockletBTreeIndex) BlockletIndex(org.apache.carbondata.format.BlockletIndex) BlockletMinMaxIndex(org.apache.carbondata.format.BlockletMinMaxIndex)

Example 7 with BlockletBTreeIndex

use of org.apache.carbondata.format.BlockletBTreeIndex in project carbondata by apache.

the class DataFileFooterConverterTest method testGetIndexInfo.

@Test
public void testGetIndexInfo() throws Exception {
    DataFileFooterConverter dataFileFooterConverter = new DataFileFooterConverter();
    final ThriftReader thriftReader = new ThriftReader("file");
    List<Encoding> encoders = new ArrayList<>();
    encoders.add(Encoding.INVERTED_INDEX);
    encoders.add(Encoding.BIT_PACKED);
    encoders.add(Encoding.DELTA);
    encoders.add(Encoding.DICTIONARY);
    encoders.add(Encoding.DIRECT_DICTIONARY);
    encoders.add(Encoding.RLE);
    ColumnSchema columnSchema = new ColumnSchema(DataType.INT, "column", "3", true, encoders, true);
    ColumnSchema columnSchema1 = new ColumnSchema(DataType.ARRAY, "column", "3", true, encoders, true);
    ColumnSchema columnSchema2 = new ColumnSchema(DataType.DECIMAL, "column", "3", true, encoders, true);
    ColumnSchema columnSchema3 = new ColumnSchema(DataType.DOUBLE, "column", "3", true, encoders, true);
    ColumnSchema columnSchema4 = new ColumnSchema(DataType.LONG, "column", "3", true, encoders, true);
    ColumnSchema columnSchema5 = new ColumnSchema(DataType.SHORT, "column", "3", true, encoders, true);
    ColumnSchema columnSchema6 = new ColumnSchema(DataType.STRUCT, "column", "3", true, encoders, true);
    ColumnSchema columnSchema7 = new ColumnSchema(DataType.STRING, "column", "3", true, encoders, true);
    final List<ColumnSchema> columnSchemas = new ArrayList<>();
    columnSchemas.add(columnSchema);
    columnSchemas.add(columnSchema1);
    columnSchemas.add(columnSchema2);
    columnSchemas.add(columnSchema3);
    columnSchemas.add(columnSchema4);
    columnSchemas.add(columnSchema5);
    columnSchemas.add(columnSchema6);
    columnSchemas.add(columnSchema7);
    final BlockIndex blockIndex = new BlockIndex();
    blockIndex.setBlock_index(new org.apache.carbondata.format.BlockletIndex());
    org.apache.carbondata.format.BlockletIndex blockletIndex1 = new org.apache.carbondata.format.BlockletIndex();
    BlockletBTreeIndex blockletBTreeIndex = new BlockletBTreeIndex();
    blockletBTreeIndex.setStart_key("1".getBytes());
    blockletBTreeIndex.setEnd_key("3".getBytes());
    blockletIndex1.setB_tree_index(blockletBTreeIndex);
    BlockletMinMaxIndex blockletMinMaxIndex = new BlockletMinMaxIndex();
    blockletMinMaxIndex.setMax_values(Arrays.asList(ByteBuffer.allocate(1).put((byte) 2)));
    blockletMinMaxIndex.setMin_values(Arrays.asList(ByteBuffer.allocate(1).put((byte) 1)));
    blockletIndex1.setMin_max_index(blockletMinMaxIndex);
    blockIndex.setBlock_index(blockletIndex1);
    List<Integer> column_cardinalities = new ArrayList<>();
    column_cardinalities.add(new Integer("1"));
    final org.apache.carbondata.format.SegmentInfo segmentInfo1 = new org.apache.carbondata.format.SegmentInfo(3, column_cardinalities);
    new MockUp<CarbonIndexFileReader>() {

        boolean mockedHasNextStatus = true;

        @SuppressWarnings("unused")
        @Mock
        public boolean hasNext() throws IOException {
            boolean temp = mockedHasNextStatus;
            mockedHasNextStatus = false;
            return temp;
        }

        @SuppressWarnings("unused")
        @Mock
        public void openThriftReader(String filePath) throws IOException {
            thriftReader.open();
        }

        @SuppressWarnings("unused")
        @Mock
        public IndexHeader readIndexHeader() throws IOException {
            return new IndexHeader(1, columnSchemas, segmentInfo1);
        }

        @SuppressWarnings("unused")
        @Mock
        public BlockIndex readBlockIndexInfo() throws IOException {
            return blockIndex;
        }

        @SuppressWarnings("unused")
        @Mock
        public void closeThriftReader() {
            thriftReader.close();
        }
    };
    new MockUp<IndexHeader>() {

        @SuppressWarnings("unused")
        @Mock
        public List<ColumnSchema> getTable_columns() {
            return columnSchemas;
        }
    };
    ByteArrayInputStream byteArrayInputStream = new ByteArrayInputStream("1".getBytes());
    final DataInputStream dataInputStream = new DataInputStream(byteArrayInputStream);
    new MockUp<FileFactory>() {

        @SuppressWarnings("unused")
        @Mock
        public DataInputStream getDataInputStream(String path, FileFactory.FileType fileType, int bufferSize) {
            return dataInputStream;
        }
    };
    String[] arr = { "a", "b", "c" };
    String fileName = "/part-0-0_batchno0-0-1495074251740.carbondata";
    TableBlockInfo tableBlockInfo = new TableBlockInfo(fileName, 3, "id", arr, 3, ColumnarFormatVersion.V1, null);
    tableBlockInfo.getBlockletInfos().setNoOfBlockLets(3);
    List<TableBlockInfo> tableBlockInfoList = new ArrayList<>();
    tableBlockInfoList.add(tableBlockInfo);
    String idxFileName = "0_batchno0-0-1495074251740.carbonindex";
    List<DataFileFooter> dataFileFooterList = dataFileFooterConverter.getIndexInfo(idxFileName, tableBlockInfoList);
    byte[] exp = dataFileFooterList.get(0).getBlockletIndex().getBtreeIndex().getStartKey();
    byte[] res = "1".getBytes();
    for (int i = 0; i < exp.length; i++) {
        assertEquals(exp[i], res[i]);
    }
}
Also used : IndexHeader(org.apache.carbondata.format.IndexHeader) TableBlockInfo(org.apache.carbondata.core.datastore.block.TableBlockInfo) ArrayList(java.util.ArrayList) ColumnSchema(org.apache.carbondata.format.ColumnSchema) MockUp(mockit.MockUp) ThriftReader(org.apache.carbondata.core.reader.ThriftReader) DataFileFooter(org.apache.carbondata.core.metadata.blocklet.DataFileFooter) BlockletMinMaxIndex(org.apache.carbondata.format.BlockletMinMaxIndex) BlockletBTreeIndex(org.apache.carbondata.format.BlockletBTreeIndex) Encoding(org.apache.carbondata.format.Encoding) DataInputStream(java.io.DataInputStream) BlockIndex(org.apache.carbondata.format.BlockIndex) ByteArrayInputStream(java.io.ByteArrayInputStream) SegmentInfo(org.apache.carbondata.core.metadata.blocklet.SegmentInfo) Test(org.junit.Test)

Aggregations

BlockletBTreeIndex (org.apache.carbondata.format.BlockletBTreeIndex)7 BlockletMinMaxIndex (org.apache.carbondata.format.BlockletMinMaxIndex)7 BlockletIndex (org.apache.carbondata.format.BlockletIndex)5 ArrayList (java.util.ArrayList)2 MockUp (mockit.MockUp)2 TableBlockInfo (org.apache.carbondata.core.datastore.block.TableBlockInfo)2 DataFileFooter (org.apache.carbondata.core.metadata.blocklet.DataFileFooter)2 SegmentInfo (org.apache.carbondata.core.metadata.blocklet.SegmentInfo)2 ColumnSchema (org.apache.carbondata.format.ColumnSchema)2 Encoding (org.apache.carbondata.format.Encoding)2 Test (org.junit.Test)2 ByteArrayInputStream (java.io.ByteArrayInputStream)1 DataInputStream (java.io.DataInputStream)1 FileFactory (org.apache.carbondata.core.datastore.impl.FileFactory)1 FileReaderImpl (org.apache.carbondata.core.datastore.impl.FileReaderImpl)1 EncodedTablePage (org.apache.carbondata.core.datastore.page.EncodedTablePage)1 TablePageStatistics (org.apache.carbondata.core.datastore.page.statistics.TablePageStatistics)1 ThriftReader (org.apache.carbondata.core.reader.ThriftReader)1 BlockIndex (org.apache.carbondata.format.BlockIndex)1 FileFooter (org.apache.carbondata.format.FileFooter)1