Ryan Blue

I don't see a benefit to relaxing this constraint in the spec.

@ggershinsky, I'm not too concerned with the size of the cache. I'm okay with 1 day, but that seems like a long time to have unencrypted key material in memory....

> > https://github.com/apache apache deleted a comment from [rahil-c](https://github.com/rahil-c) ([#9695 (comment)](https://github.com/apache/iceberg/pull/9695#event-13139602437)) > > https://github.com/apache deleted a comment from [rahil-c](https://github.com/rahil-c) ([#9695 (comment)](https://github.com/apache/iceberg/pull/9695#event-13139602791)) > > Saw that [apache](https://github.com/apache) performed these actions above....

> The readFilteredRowGroup method provided by Parquet will detect whether there is a filter pushed down, and only return the filtered row-group when there is a push-down filter. I commented...

I brought this up at the community sync to gauge what other people thought about changing the behavior here without introducing a "strategy" option and the response was positive. Our...

Merged. Thanks, @jkolash!

@flyrain, I think your use case makes sense and that we should support some version of client-side purge. That said, I don't think that either option proposed here is the...

Here are the benchmark results: ``` ## main Benchmark Mode Cnt Score Error Units IcebergSourceFlatAvroDataReadBenchmark.readIceberg ss 5 8.883 ± 0.321 s/op IcebergSourceFlatAvroDataReadBenchmark.readWithProjectionIceberg ss 5 7.173 ± 0.254 s/op ## this...

Thanks for the updates, @cashmand! I think this looks good to go. Running the tests again.

Running tests