tustvold commented on code in PR #6527: URL: https://github.com/apache/arrow-rs/pull/6527#discussion_r1792322244
########## arrow/src/lib.rs: ########## @@ -345,6 +345,67 @@ //! orchestrates the primitives exported by this crate into an embeddable query engine, with //! SQL and DataFrame frontends, and heavily influences this crate's roadmap. //! +//! The Rust implementation does not provide the ChunkedArray abstraction implemented by the Python +//! and C++ Arrow implementations. The recommended alternative is to use one of the following: +//! - `Vec<ArrayRef>` a simple, eager version of a `ChunkedArray` +//! - `impl Iterator<Item=ArrayRef>` a lazy version of a `ChunkedArray` +//! - `impl Stream<Item=ArrayRef>` a lazy async version of a `ChunkedArray` +//! +//! Similar patterns can be applied at the `RecordBatch` level. For example, [DataFusion] makes +//! extensive use of [RecordBatchStream]. +//! +//! This approach integrates well into the Rust ecosystem, simplifies the implementation and +//! encourages the use of performant lazy and async patterns. +//! +//! Aside from providing a slightly less convenient API, one other downside is the lack of support +//! for processing compute kernels across chunked arrays. But this use case is well-supported by +//! [DataFusion]. +//! Review Comment: ```suggestion ``` I'm not sure I agree with this statement, iterators are fairly ergonomic once you get the hang of them, and it is fairly trivial to use the processing kernels with them. e.g. ``` let arrays = todo!(); let processed = arrays.iter().map(|x| cast(&DataType::Int64, x.as_ref())).collect::<Result<Vec<_>, _>>().unwrap(); ``` -- 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. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
