paleolimbot commented on PR #61: URL: https://github.com/apache/arrow-nanoarrow/pull/61#issuecomment-1380980847
I think this PR is at a good point for initial review! What it does is: - Establishes a home for the IPC extension (the extensions/nanoarrow_ipc folder) - Implements a minimum viable build configuration for the flatcc buffer configuration + unit testing with Arrow C++ - Implements the minimum number of type definitions to implement decoding and verifying a schema message: `ArrowIpcError`, `ArrowIpcBufferView`, `ArrowIpcMessageType`, `ArrowIpcEndianness`, and `ArrowIpcReader`. - Implements three functions: `ArrowIpcReaderPeek()` (checks the first few bytes of the message and advances the data pointer past the end of the message), `ArrowIpcReaderVerify()` (runs flatbuffers verification and advances the data pointer past the end of the message on success), and `ArrowIpcReaderDecode()` (places information from the message into the `ArrowIpcReader` struct). After this PR, there are a number of improvements that could be worked on more or less independently of one another: - Make the build configuration more flexible (some of David's suggestions included ways to vendor the flatcc runtime and/or link to a shared flatcc) - Add doxygen documentation + add it to the nanoarrow documentation build - Implement the rest of the schema types - Add the requisite GitHub actions (ctest, coverage, valgrind) - Add R bindings so that actual humans can test this - Implement RecordBatch message decoding -- 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]
