Examples with FileMetaData org.apache.parquet.format.FileMetaData used on opensource projects

Example 6 with FileMetaData

use of org.apache.parquet.format.FileMetaData in project drill by axbaretto.

the class ParquetReaderUtility method getColNameToSchemaElementMapping.

public static Map<String, SchemaElement> getColNameToSchemaElementMapping(ParquetMetadata footer) {
    HashMap<String, SchemaElement> schemaElements = new HashMap<>();
    FileMetaData fileMetaData = new ParquetMetadataConverter().toParquetMetadata(ParquetFileWriter.CURRENT_VERSION, footer);
    for (SchemaElement se : fileMetaData.getSchema()) {
        schemaElements.put(se.getName(), se);
    }
    return schemaElements;
}

Example 7 with FileMetaData

use of org.apache.parquet.format.FileMetaData in project presto by prestodb.

the class ParquetWriter method getFooter.

static Slice getFooter(List<RowGroup> rowGroups, MessageType messageType) throws IOException {
    FileMetaData fileMetaData = new FileMetaData();
    fileMetaData.setVersion(1);
    fileMetaData.setSchema(MessageTypeConverter.toParquetSchema(messageType));
    long totalRows = rowGroups.stream().mapToLong(RowGroup::getNum_rows).sum();
    fileMetaData.setNum_rows(totalRows);
    fileMetaData.setRow_groups(ImmutableList.copyOf(rowGroups));
    DynamicSliceOutput dynamicSliceOutput = new DynamicSliceOutput(40);
    Util.writeFileMetaData(fileMetaData, dynamicSliceOutput);
    return dynamicSliceOutput.slice();
}

Example 8 with FileMetaData

use of org.apache.parquet.format.FileMetaData in project presto by prestodb.

the class MetadataReader method readFooter.

public static ParquetFileMetadata readFooter(ParquetDataSource parquetDataSource, long fileSize) throws IOException {
    // Parquet File Layout:
    // 
    // MAGIC
    // variable: Data
    // variable: Metadata
    // 4 bytes: MetadataLength
    // MAGIC
    validateParquet(fileSize >= MAGIC.length() + POST_SCRIPT_SIZE, "%s is not a valid Parquet File", parquetDataSource.getId());
    // EXPECTED_FOOTER_SIZE is an int, so this will never fail
    byte[] buffer = new byte[toIntExact(min(fileSize, EXPECTED_FOOTER_SIZE))];
    parquetDataSource.readFully(fileSize - buffer.length, buffer);
    Slice tailSlice = wrappedBuffer(buffer);
    Slice magic = tailSlice.slice(tailSlice.length() - MAGIC.length(), MAGIC.length());
    if (!MAGIC.equals(magic)) {
        throw new ParquetCorruptionException(format("Not valid Parquet file: %s expected magic number: %s got: %s", parquetDataSource.getId(), Arrays.toString(MAGIC.getBytes()), Arrays.toString(magic.getBytes())));
    }
    int metadataLength = tailSlice.getInt(tailSlice.length() - POST_SCRIPT_SIZE);
    int completeFooterSize = metadataLength + POST_SCRIPT_SIZE;
    long metadataFileOffset = fileSize - completeFooterSize;
    validateParquet(metadataFileOffset >= MAGIC.length() && metadataFileOffset + POST_SCRIPT_SIZE < fileSize, "Corrupted Parquet file: %s metadata index: %s out of range", parquetDataSource.getId(), metadataFileOffset);
    // Ensure the slice covers the entire metadata range
    if (tailSlice.length() < completeFooterSize) {
        byte[] footerBuffer = new byte[completeFooterSize];
        parquetDataSource.readFully(metadataFileOffset, footerBuffer, 0, footerBuffer.length - tailSlice.length());
        // Copy the previous slice contents into the new buffer
        tailSlice.getBytes(0, footerBuffer, footerBuffer.length - tailSlice.length(), tailSlice.length());
        tailSlice = wrappedBuffer(footerBuffer, 0, footerBuffer.length);
    }
    FileMetaData fileMetaData = readFileMetaData(tailSlice.slice(tailSlice.length() - completeFooterSize, metadataLength).getInput());
    List<SchemaElement> schema = fileMetaData.getSchema();
    validateParquet(!schema.isEmpty(), "Empty Parquet schema in file: %s", parquetDataSource.getId());
    MessageType messageType = readParquetSchema(schema);
    List<BlockMetaData> blocks = new ArrayList<>();
    List<RowGroup> rowGroups = fileMetaData.getRow_groups();
    if (rowGroups != null) {
        for (RowGroup rowGroup : rowGroups) {
            BlockMetaData blockMetaData = new BlockMetaData();
            blockMetaData.setRowCount(rowGroup.getNum_rows());
            blockMetaData.setTotalByteSize(rowGroup.getTotal_byte_size());
            List<ColumnChunk> columns = rowGroup.getColumns();
            validateParquet(!columns.isEmpty(), "No columns in row group: %s", rowGroup);
            String filePath = columns.get(0).getFile_path();
            for (ColumnChunk columnChunk : columns) {
                validateParquet((filePath == null && columnChunk.getFile_path() == null) || (filePath != null && filePath.equals(columnChunk.getFile_path())), "all column chunks of the same row group must be in the same file");
                ColumnMetaData metaData = columnChunk.meta_data;
                String[] path = metaData.path_in_schema.stream().map(value -> value.toLowerCase(Locale.ENGLISH)).toArray(String[]::new);
                ColumnPath columnPath = ColumnPath.get(path);
                PrimitiveType primitiveType = messageType.getType(columnPath.toArray()).asPrimitiveType();
                PrimitiveTypeName primitiveTypeName = primitiveType.getPrimitiveTypeName();
                ColumnChunkMetaData column = ColumnChunkMetaData.get(columnPath, primitiveType, CompressionCodecName.fromParquet(metaData.codec), PARQUET_METADATA_CONVERTER.convertEncodingStats(metaData.encoding_stats), readEncodings(metaData.encodings), readStats(metaData.statistics, primitiveTypeName), metaData.data_page_offset, metaData.dictionary_page_offset, metaData.num_values, metaData.total_compressed_size, metaData.total_uncompressed_size);
                column.setColumnIndexReference(toColumnIndexReference(columnChunk));
                column.setOffsetIndexReference(toOffsetIndexReference(columnChunk));
                blockMetaData.addColumn(column);
            }
            blockMetaData.setPath(filePath);
            blocks.add(blockMetaData);
        }
    }
    Map<String, String> keyValueMetaData = new HashMap<>();
    List<KeyValue> keyValueList = fileMetaData.getKey_value_metadata();
    if (keyValueList != null) {
        for (KeyValue keyValue : keyValueList) {
            keyValueMetaData.put(keyValue.key, keyValue.value);
        }
    }
    ParquetMetadata parquetMetadata = new ParquetMetadata(new org.apache.parquet.hadoop.metadata.FileMetaData(messageType, keyValueMetaData, fileMetaData.getCreated_by()), blocks);
    return new ParquetFileMetadata(parquetMetadata, toIntExact(metadataLength));
}

Example 9 with FileMetaData

use of org.apache.parquet.format.FileMetaData in project drill by apache.

the class ParquetReaderUtility method getColNameToSchemaElementMapping.

/**
 * Map full schema paths in format `a`.`b`.`c` to respective SchemaElement objects.
 *
 * @param footer Parquet file metadata
 * @return       schema full path to SchemaElement map
 */
public static Map<String, SchemaElement> getColNameToSchemaElementMapping(ParquetMetadata footer) {
    Map<String, SchemaElement> schemaElements = new HashMap<>();
    FileMetaData fileMetaData = new ParquetMetadataConverter().toParquetMetadata(ParquetFileWriter.CURRENT_VERSION, footer);
    Iterator<SchemaElement> iter = fileMetaData.getSchema().iterator();
    // and thus to avoid the need to cut it out again when comparing with SchemaPath string representation
    if (iter.hasNext()) {
        iter.next();
    }
    while (iter.hasNext()) {
        addSchemaElementMapping(iter, new StringBuilder(), schemaElements);
    }
    return schemaElements;
}