kbendick commented on a change in pull request #1573: URL: https://github.com/apache/iceberg/pull/1573#discussion_r502760999
########## File path: aws/src/main/java/org/apache/iceberg/aws/s3/S3InputStream.java ########## @@ -0,0 +1,146 @@ +/* + * Licensed to the Apache Software Foundation (ASF) under one + * or more contributor license agreements. See the NOTICE file + * distributed with this work for additional information + * regarding copyright ownership. The ASF licenses this file + * to you under the Apache License, Version 2.0 (the + * "License"); you may not use this file except in compliance + * with the License. You may obtain a copy of the License at + * + * http://www.apache.org/licenses/LICENSE-2.0 + * + * Unless required by applicable law or agreed to in writing, + * software distributed under the License is distributed on an + * "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY + * KIND, either express or implied. See the License for the + * specific language governing permissions and limitations + * under the License. + */ + +package org.apache.iceberg.aws.s3; + +import com.amazonaws.services.s3.AmazonS3; +import com.amazonaws.services.s3.AmazonS3URI; +import com.amazonaws.services.s3.model.GetObjectRequest; +import com.amazonaws.services.s3.model.S3ObjectInputStream; +import java.io.IOException; +import java.util.Arrays; +import org.apache.commons.compress.utils.IOUtils; +import org.apache.iceberg.io.SeekableInputStream; +import org.apache.iceberg.relocated.com.google.common.base.Joiner; +import org.apache.iceberg.relocated.com.google.common.base.Preconditions; +import org.slf4j.Logger; +import org.slf4j.LoggerFactory; + +public class S3InputStream extends SeekableInputStream { + private static final Logger LOG = LoggerFactory.getLogger(S3InputStream.class); + + private final StackTraceElement[] createStack; + private final AmazonS3 s3; + private final AmazonS3URI location; + + private S3ObjectInputStream stream; + private long pos = 0; + private long next = 0; + private boolean closed = false; + + private int skipSize = 1024 * 1024; + + public S3InputStream(AmazonS3 s3, AmazonS3URI location) { + this.s3 = s3; + this.location = location; + + createStack = Thread.currentThread().getStackTrace(); + } + + @Override + public long getPos() { + return pos; + } + + @Override + public void seek(long newPos) { + Preconditions.checkState(!closed, "already closed"); + Preconditions.checkArgument(newPos >= 0, "position is negative: %s", newPos); + + // this allows a seek beyond the end of the stream but the next read will fail + next = newPos; + } + + @Override + public int read() { + // This stream is wrapped with BufferedInputStream, so this method should never be called + throw new UnsupportedOperationException(); Review comment: I'm in agreement on getting a working initial implementation in and then optimizing it later. We use similar optimizations when reading / writing directly to S3 from plain old java programs. This might somewhat go against the grain of how `FileIO` works, but if there were a filesystem concept like exists in some of the python code, we could make use of the multipart file upload manager and then upload based on the configured block size, which makes reads from S3 (when reading along the same block size) significantly more performant. It also means that all file uploads have a final commit operation (such that they all succeed or don't). I guess with Iceberg that final commit operation isn't really necessary though, as it's ok for tables to have files in them that don't belong as they will be filtered out by the metadata. But I'm still hugely in favor of making multipart uploads work for us. As well as a more performant way to download only the portion of the file thats needed (e.g. the footer, etc). ---------------------------------------------------------------- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
