rdblue edited a comment on issue #569: Use broadcast variables in IcebergSource URL: https://github.com/apache/incubator-iceberg/pull/569#issuecomment-561897938 Here's a slightly cleaner way. I updated `HadoopFileIO` to use a `Supplier<Configuration>` instead of a `SerializableConfiguration`: ```java public class HadoopFileIO implements FileIO { private final Supplier<Configuration> hadoopConf; public HadoopFileIO(Configuration hadoopConf) { this(new SerializableConfiguration(hadoopConf)::get); } public HadoopFileIO(Supplier<Configuration> hadoopConf) { this.hadoopConf = hadoopConf; } public Configuration conf() { return hadoopConf.get(); } ... } ``` Then we just need to pass Spark's `SerializableConfiguration` in Spark: ```java if (table.io() instanceof HadoopFileIO) { this.fileIo = new HadoopFileIO(new SerializableConfiguration(((HadoopFileIO) table.io()).conf())::value); } else { this.fileIo = table.io(); } ``` Here's a branch: https://github.com/rdblue/iceberg/commit/84aa3ba434898b5c8b9ca71c9754c437ec3cff51 It shouldn't be too difficult to add the broadcast to that.
---------------------------------------------------------------- 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] With regards, Apache Git Services --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
