Use Case
my specific use case is to override the basename_template param because at times when I'm bombarding S3 with thousands of concurrent tasks, hitting the parquet file directly without using the txn log to find the files yields performance benefits. hence, explicitly naming the parquet file allows me to statically deterministically specify the filename (as opposed to querying for it at runtime)
Furthermore, this would future proof and expose any additional kwargs / enhancements to the pyarrow datasets api
I'm happy to propose the change and submit PR if this is an acceptable enhancement.
Description
allow passing
ds.write_dataset
kwarg overrides intowrite_deltalake
function https://github.com/delta-io/delta-rs/blob/fcfd1bfa08b8ddda2a606d7f1f58b75e89206d45/python/deltalake/writer.py#L327-L344Use Case my specific use case is to override the basename_template param because at times when I'm bombarding S3 with thousands of concurrent tasks, hitting the parquet file directly without using the txn log to find the files yields performance benefits. hence, explicitly naming the parquet file allows me to statically deterministically specify the filename (as opposed to querying for it at runtime)
Furthermore, this would future proof and expose any additional kwargs / enhancements to the pyarrow datasets api
I'm happy to propose the change and submit PR if this is an acceptable enhancement.