Examples with PageHeader parquet.format.PageHeader used on opensource projects

Example 1 with PageHeader

use of parquet.format.PageHeader in project presto by prestodb.

the class ParquetColumnChunk method readAllPages.

public ParquetPageReader readAllPages() throws IOException {
    List<ParquetDataPage> pages = new ArrayList<>();
    ParquetDictionaryPage dictionaryPage = null;
    long valueCount = 0;
    while (valueCount < descriptor.getColumnChunkMetaData().getValueCount()) {
        PageHeader pageHeader = readPageHeader();
        int uncompressedPageSize = pageHeader.getUncompressed_page_size();
        int compressedPageSize = pageHeader.getCompressed_page_size();
        switch(pageHeader.type) {
            case DICTIONARY_PAGE:
                if (dictionaryPage != null) {
                    throw new ParquetCorruptionException("%s has more than one dictionary page in column chunk", descriptor.getColumnDescriptor());
                }
                dictionaryPage = readDictionaryPage(pageHeader, uncompressedPageSize, compressedPageSize);
                break;
            case DATA_PAGE:
                valueCount += readDataPageV1(pageHeader, uncompressedPageSize, compressedPageSize, pages);
                break;
            case DATA_PAGE_V2:
                valueCount += readDataPageV2(pageHeader, uncompressedPageSize, compressedPageSize, pages);
                break;
            default:
                skip(compressedPageSize);
                break;
        }
    }
    return new ParquetPageReader(descriptor.getColumnChunkMetaData().getCodec(), pages, dictionaryPage);
}

Example 2 with PageHeader

use of parquet.format.PageHeader in project presto by prestodb.

the class ParquetPredicateUtils method readDictionaryPage.

private static Optional<ParquetDictionaryPage> readDictionaryPage(byte[] data, CompressionCodecName codecName) {
    try {
        ByteArrayInputStream inputStream = new ByteArrayInputStream(data);
        PageHeader pageHeader = Util.readPageHeader(inputStream);
        if (pageHeader.type != PageType.DICTIONARY_PAGE) {
            return Optional.empty();
        }
        Slice compressedData = wrappedBuffer(data, data.length - inputStream.available(), pageHeader.getCompressed_page_size());
        DictionaryPageHeader dicHeader = pageHeader.getDictionary_page_header();
        ParquetEncoding encoding = getParquetEncoding(Encoding.valueOf(dicHeader.getEncoding().name()));
        int dictionarySize = dicHeader.getNum_values();
        return Optional.of(new ParquetDictionaryPage(decompress(codecName, compressedData, pageHeader.getUncompressed_page_size()), dictionarySize, encoding));
    } catch (IOException ignored) {
        return Optional.empty();
    }
}