Read Avro parquet file from inside JAR

Question

I'm trying to read a parquet file bundled as a resource inside a JAR, ideally as a stream.

Does anyone have a working example that doesn't involve writing the resource out as a temporary file first?

Here is the code I'm using to read the files which works fine in the IDE before bundling as a JAR:

import org.apache.hadoop.conf.Configuration;
import org.apache.hadoop.fs.Path;
import org.apache.parquet.avro.AvroParquetReader;

                        try {
                            Path path = new Path(classLoader.getResource(pattern_id).toURI());

                            Configuration conf = new Configuration();

                            try (ParquetReader<GenericRecord> r = AvroParquetReader.<GenericRecord>builder(
                                                                             HadoopInputFile.fromPath(path, conf))
                                                                             .disableCompatibility()
                                                                             .build()) {
                                patternsFound.add(pattern_id);

                                GenericRecord record;
                                while ((record = r.read()) != null) {
                                        // Do some work

                                }


                            } catch (IOException e) {
                                e.printStackTrace();
                            }
                        } catch (NullPointerException | URISyntaxException e) {

                            e.printStackTrace();
                        }

When running this code from a JAR file, I get this error:

org.apache.hadoop.fs.UnsupportedFileSystemException: No FileSystem for scheme "jar"

Which I figured I could get around by using:

InputStream inputFile = classLoader.getResourceAsStream(pattern_id);

But don't know how to get AvroParquetReader to work with Input Streams.

Answer 1

I winded up being able to read a parquet file as a resource stream by tweaking the solution here: https://stackoverflow.com/a/58261488/3112960

import org.apache.commons.io.IOUtils;
import org.apache.parquet.io.DelegatingSeekableInputStream;
import org.apache.parquet.io.InputFile;
import org.apache.parquet.io.SeekableInputStream;

import java.io.ByteArrayInputStream;
import java.io.IOException;
import java.io.InputStream;

public class ParquetStream implements InputFile {
    private final String streamId;
    private final byte[] data;

    private static class SeekableByteArrayInputStream extends ByteArrayInputStream {
        public SeekableByteArrayInputStream(byte[] buf) {
            super(buf);
        }

        public void setPos(int pos) {
            this.pos = pos;
        }

        public int getPos() {
            return this.pos;
        }
    }

    public ParquetStream(String streamId, InputStream stream) throws IOException {
        this.streamId = streamId;

        this.data = IOUtils.toByteArray(stream);
    }

    @Override
    public long getLength()  {
        return this.data.length;
    }

    @Override
    public SeekableInputStream newStream() throws IOException {
        return new DelegatingSeekableInputStream(new SeekableByteArrayInputStream(this.data)) {
            @Override
            public void seek(long newPos) {
                ((SeekableByteArrayInputStream) this.getStream()).setPos((int) newPos);
            }

            @Override
            public long getPos() {
                return ((SeekableByteArrayInputStream) this.getStream()).getPos();
            }
        };
    }

    @Override
    public String toString() {
        return "ParquetStream[" + streamId + "]";
    }
}

Then I can just do this:

                        InputStream in = classLoader.getResourceAsStream(pattern_id);

                        try {
                            ParquetStream parquetStream = new ParquetStream(pattern_id, in);

                            ParquetReader<GenericRecord> r = AvroParquetReader.<GenericRecord>builder(parquetStream)
                                                                             .disableCompatibility()
                                                                             .build();

                            GenericRecord record;
                            while ((record = r.read()) != null) {
                                // do some work
                            }
                        } catch (IOException e) {
                             e.printStackTrace();
                        }

Maybe this will help someone in the future since I couldn't find any straightforward answers to this.