Examples with HIVE_BAD_DATA com.facebook.presto.hive.HiveErrorCode.HIVE_BAD_DATA used on opensource projects

Example 1 with HIVE_BAD_DATA

use of com.facebook.presto.hive.HiveErrorCode.HIVE_BAD_DATA in project presto by prestodb.

the class RcFilePageSourceFactory method createPageSource.

@Override
public Optional<? extends ConnectorPageSource> createPageSource(Configuration configuration, ConnectorSession session, Path path, long start, long length, long fileSize, Storage storage, SchemaTableName tableName, Map<String, String> tableParameters, List<HiveColumnHandle> columns, TupleDomain<HiveColumnHandle> effectivePredicate, DateTimeZone hiveStorageTimeZone, HiveFileContext hiveFileContext, Optional<EncryptionInformation> encryptionInformation) {
    if (!columns.isEmpty() && columns.stream().allMatch(hiveColumnHandle -> hiveColumnHandle.getColumnType() == AGGREGATED)) {
        throw new UnsupportedOperationException("Partial aggregation pushdown only supported for ORC/Parquet files. " + "Table " + tableName.toString() + " has file (" + path.toString() + ") of format " + storage.getStorageFormat().getOutputFormat() + ". Set session property hive.pushdown_partial_aggregations_into_scan=false and execute query again");
    }
    RcFileEncoding rcFileEncoding;
    if (LazyBinaryColumnarSerDe.class.getName().equals(storage.getStorageFormat().getSerDe())) {
        rcFileEncoding = new BinaryRcFileEncoding();
    } else if (ColumnarSerDe.class.getName().equals(storage.getStorageFormat().getSerDe())) {
        rcFileEncoding = createTextVectorEncoding(getHiveSchema(storage.getSerdeParameters(), tableParameters), hiveStorageTimeZone);
    } else {
        return Optional.empty();
    }
    if (fileSize == 0) {
        throw new PrestoException(HIVE_BAD_DATA, "RCFile is empty: " + path);
    }
    FSDataInputStream inputStream;
    try {
        inputStream = hdfsEnvironment.getFileSystem(session.getUser(), path, configuration).openFile(path, hiveFileContext);
    } catch (Exception e) {
        if (nullToEmpty(e.getMessage()).trim().equals("Filesystem closed") || e instanceof FileNotFoundException) {
            throw new PrestoException(HIVE_CANNOT_OPEN_SPLIT, e);
        }
        throw new PrestoException(HIVE_CANNOT_OPEN_SPLIT, splitError(e, path, start, length), e);
    }
    try {
        ImmutableMap.Builder<Integer, Type> readColumns = ImmutableMap.builder();
        for (HiveColumnHandle column : columns) {
            readColumns.put(column.getHiveColumnIndex(), column.getHiveType().getType(typeManager));
        }
        RcFileReader rcFileReader = new RcFileReader(new HdfsRcFileDataSource(path.toString(), inputStream, fileSize, stats), rcFileEncoding, readColumns.build(), new AircompressorCodecFactory(new HadoopCodecFactory(configuration.getClassLoader())), start, length, new DataSize(8, Unit.MEGABYTE));
        return Optional.of(new RcFilePageSource(rcFileReader, columns, typeManager));
    } catch (Throwable e) {
        try {
            inputStream.close();
        } catch (IOException ignored) {
        }
        if (e instanceof PrestoException) {
            throw (PrestoException) e;
        }
        String message = splitError(e, path, start, length);
        if (e instanceof RcFileCorruptionException) {
            throw new PrestoException(HIVE_BAD_DATA, message, e);
        }
        if (e.getClass().getSimpleName().equals("BlockMissingException")) {
            throw new PrestoException(HIVE_MISSING_DATA, message, e);
        }
        throw new PrestoException(HIVE_CANNOT_OPEN_SPLIT, message, e);
    }
}

Example 2 with HIVE_BAD_DATA

use of com.facebook.presto.hive.HiveErrorCode.HIVE_BAD_DATA in project presto by prestodb.

the class ParquetPageSourceFactory method createParquetPageSource.

public static ConnectorPageSource createParquetPageSource(HdfsEnvironment hdfsEnvironment, String user, Configuration configuration, Path path, long start, long length, long fileSize, List<HiveColumnHandle> columns, SchemaTableName tableName, boolean useParquetColumnNames, DataSize maxReadBlockSize, boolean batchReaderEnabled, boolean verificationEnabled, TypeManager typeManager, StandardFunctionResolution functionResolution, TupleDomain<HiveColumnHandle> effectivePredicate, FileFormatDataSourceStats stats, HiveFileContext hiveFileContext, ParquetMetadataSource parquetMetadataSource, boolean columnIndexFilterEnabled) {
    AggregatedMemoryContext systemMemoryContext = newSimpleAggregatedMemoryContext();
    ParquetDataSource dataSource = null;
    try {
        FSDataInputStream inputStream = hdfsEnvironment.getFileSystem(user, path, configuration).openFile(path, hiveFileContext);
        dataSource = buildHdfsParquetDataSource(inputStream, path, stats);
        ParquetMetadata parquetMetadata = parquetMetadataSource.getParquetMetadata(dataSource, fileSize, hiveFileContext.isCacheable()).getParquetMetadata();
        if (!columns.isEmpty() && columns.stream().allMatch(hiveColumnHandle -> hiveColumnHandle.getColumnType() == AGGREGATED)) {
            return new AggregatedParquetPageSource(columns, parquetMetadata, typeManager, functionResolution);
        }
        FileMetaData fileMetaData = parquetMetadata.getFileMetaData();
        MessageType fileSchema = fileMetaData.getSchema();
        Optional<MessageType> message = columns.stream().filter(column -> column.getColumnType() == REGULAR || isPushedDownSubfield(column)).map(column -> getColumnType(typeManager.getType(column.getTypeSignature()), fileSchema, useParquetColumnNames, column, tableName, path)).filter(Optional::isPresent).map(Optional::get).map(type -> new MessageType(fileSchema.getName(), type)).reduce(MessageType::union);
        MessageType requestedSchema = message.orElse(new MessageType(fileSchema.getName(), ImmutableList.of()));
        ImmutableList.Builder<BlockMetaData> footerBlocks = ImmutableList.builder();
        for (BlockMetaData block : parquetMetadata.getBlocks()) {
            long firstDataPage = block.getColumns().get(0).getFirstDataPageOffset();
            if (firstDataPage >= start && firstDataPage < start + length) {
                footerBlocks.add(block);
            }
        }
        Map<List<String>, RichColumnDescriptor> descriptorsByPath = getDescriptors(fileSchema, requestedSchema);
        TupleDomain<ColumnDescriptor> parquetTupleDomain = getParquetTupleDomain(descriptorsByPath, effectivePredicate);
        Predicate parquetPredicate = buildPredicate(requestedSchema, parquetTupleDomain, descriptorsByPath);
        final ParquetDataSource finalDataSource = dataSource;
        ImmutableList.Builder<BlockMetaData> blocks = ImmutableList.builder();
        List<ColumnIndexStore> blockIndexStores = new ArrayList<>();
        for (BlockMetaData block : footerBlocks.build()) {
            Optional<ColumnIndexStore> columnIndexStore = ColumnIndexFilterUtils.getColumnIndexStore(parquetPredicate, finalDataSource, block, descriptorsByPath, columnIndexFilterEnabled);
            if (predicateMatches(parquetPredicate, block, finalDataSource, descriptorsByPath, parquetTupleDomain, columnIndexStore, columnIndexFilterEnabled)) {
                blocks.add(block);
                blockIndexStores.add(columnIndexStore.orElse(null));
                hiveFileContext.incrementCounter("parquet.blocksRead", 1);
                hiveFileContext.incrementCounter("parquet.rowsRead", block.getRowCount());
                hiveFileContext.incrementCounter("parquet.totalBytesRead", block.getTotalByteSize());
            } else {
                hiveFileContext.incrementCounter("parquet.blocksSkipped", 1);
                hiveFileContext.incrementCounter("parquet.rowsSkipped", block.getRowCount());
                hiveFileContext.incrementCounter("parquet.totalBytesSkipped", block.getTotalByteSize());
            }
        }
        MessageColumnIO messageColumnIO = getColumnIO(fileSchema, requestedSchema);
        ParquetReader parquetReader = new ParquetReader(messageColumnIO, blocks.build(), dataSource, systemMemoryContext, maxReadBlockSize, batchReaderEnabled, verificationEnabled, parquetPredicate, blockIndexStores, columnIndexFilterEnabled);
        ImmutableList.Builder<String> namesBuilder = ImmutableList.builder();
        ImmutableList.Builder<Type> typesBuilder = ImmutableList.builder();
        ImmutableList.Builder<Optional<Field>> fieldsBuilder = ImmutableList.builder();
        for (HiveColumnHandle column : columns) {
            checkArgument(column.getColumnType() == REGULAR || column.getColumnType() == SYNTHESIZED, "column type must be regular or synthesized column");
            String name = column.getName();
            Type type = typeManager.getType(column.getTypeSignature());
            namesBuilder.add(name);
            typesBuilder.add(type);
            if (column.getColumnType() == SYNTHESIZED) {
                Subfield pushedDownSubfield = getPushedDownSubfield(column);
                List<String> nestedColumnPath = nestedColumnPath(pushedDownSubfield);
                Optional<ColumnIO> columnIO = findNestedColumnIO(lookupColumnByName(messageColumnIO, pushedDownSubfield.getRootName()), nestedColumnPath);
                if (columnIO.isPresent()) {
                    fieldsBuilder.add(constructField(type, columnIO.get()));
                } else {
                    fieldsBuilder.add(Optional.empty());
                }
            } else if (getParquetType(type, fileSchema, useParquetColumnNames, column, tableName, path).isPresent()) {
                String columnName = useParquetColumnNames ? name : fileSchema.getFields().get(column.getHiveColumnIndex()).getName();
                fieldsBuilder.add(constructField(type, lookupColumnByName(messageColumnIO, columnName)));
            } else {
                fieldsBuilder.add(Optional.empty());
            }
        }
        return new ParquetPageSource(parquetReader, typesBuilder.build(), fieldsBuilder.build(), namesBuilder.build(), hiveFileContext.getStats());
    } catch (Exception e) {
        try {
            if (dataSource != null) {
                dataSource.close();
            }
        } catch (IOException ignored) {
        }
        if (e instanceof PrestoException) {
            throw (PrestoException) e;
        }
        if (e instanceof ParquetCorruptionException) {
            throw new PrestoException(HIVE_BAD_DATA, e);
        }
        if (e instanceof AccessControlException) {
            throw new PrestoException(PERMISSION_DENIED, e.getMessage(), e);
        }
        if (nullToEmpty(e.getMessage()).trim().equals("Filesystem closed") || e instanceof FileNotFoundException) {
            throw new PrestoException(HIVE_CANNOT_OPEN_SPLIT, e);
        }
        String message = format("Error opening Hive split %s (offset=%s, length=%s): %s", path, start, length, e.getMessage());
        if (e.getClass().getSimpleName().equals("BlockMissingException")) {
            throw new PrestoException(HIVE_MISSING_DATA, message, e);
        }
        throw new PrestoException(HIVE_CANNOT_OPEN_SPLIT, message, e);
    }
}

Example 3 with HIVE_BAD_DATA

use of com.facebook.presto.hive.HiveErrorCode.HIVE_BAD_DATA in project presto by prestodb.

the class HiveUtil method createRecordReader.

public static RecordReader<?, ?> createRecordReader(Configuration configuration, Path path, long start, long length, Properties schema, List<HiveColumnHandle> columns, Map<String, String> customSplitInfo) {
    // determine which hive columns we will read
    List<HiveColumnHandle> readColumns = ImmutableList.copyOf(filter(columns, column -> column.getColumnType() == REGULAR));
    List<Integer> readHiveColumnIndexes = ImmutableList.copyOf(transform(readColumns, HiveColumnHandle::getHiveColumnIndex));
    // Tell hive the columns we would like to read, this lets hive optimize reading column oriented files
    setReadColumns(configuration, readHiveColumnIndexes);
    // Only propagate serialization schema configs by default
    Predicate<String> schemaFilter = schemaProperty -> schemaProperty.startsWith("serialization.");
    InputFormat<?, ?> inputFormat = getInputFormat(configuration, getInputFormatName(schema), true);
    JobConf jobConf = toJobConf(configuration);
    FileSplit fileSplit = new FileSplit(path, start, length, (String[]) null);
    if (!customSplitInfo.isEmpty() && isHudiRealtimeSplit(customSplitInfo)) {
        fileSplit = recreateSplitWithCustomInfo(fileSplit, customSplitInfo);
        // Add additional column information for record reader
        List<String> readHiveColumnNames = ImmutableList.copyOf(transform(readColumns, HiveColumnHandle::getName));
        jobConf.set(READ_COLUMN_NAMES_CONF_STR, Joiner.on(',').join(readHiveColumnNames));
        // Remove filter when using customSplitInfo as the record reader requires complete schema configs
        schemaFilter = schemaProperty -> true;
    }
    schema.stringPropertyNames().stream().filter(schemaFilter).forEach(name -> jobConf.set(name, schema.getProperty(name)));
    // add Airlift LZO and LZOP to head of codecs list so as to not override existing entries
    List<String> codecs = newArrayList(Splitter.on(",").trimResults().omitEmptyStrings().split(jobConf.get("io.compression.codecs", "")));
    if (!codecs.contains(LzoCodec.class.getName())) {
        codecs.add(0, LzoCodec.class.getName());
    }
    if (!codecs.contains(LzopCodec.class.getName())) {
        codecs.add(0, LzopCodec.class.getName());
    }
    jobConf.set("io.compression.codecs", codecs.stream().collect(joining(",")));
    try {
        RecordReader<WritableComparable, Writable> recordReader = (RecordReader<WritableComparable, Writable>) inputFormat.getRecordReader(fileSplit, jobConf, Reporter.NULL);
        int headerCount = getHeaderCount(schema);
        // Only skip header rows when the split is at the beginning of the file
        if (start == 0 && headerCount > 0) {
            Utilities.skipHeader(recordReader, headerCount, recordReader.createKey(), recordReader.createValue());
        }
        int footerCount = getFooterCount(schema);
        if (footerCount > 0) {
            recordReader = new FooterAwareRecordReader<>(recordReader, footerCount, jobConf);
        }
        return recordReader;
    } catch (IOException e) {
        if (e instanceof TextLineLengthLimitExceededException) {
            throw new PrestoException(HIVE_BAD_DATA, "Line too long in text file: " + path, e);
        }
        throw new PrestoException(HIVE_CANNOT_OPEN_SPLIT, format("Error opening Hive split %s (offset=%s, length=%s) using %s: %s", path, start, length, getInputFormatName(schema), firstNonNull(e.getMessage(), e.getClass().getName())), e);
    }
}