yihua opened a new pull request, #716:
URL: https://github.com/apache/hudi-rs/pull/716
## Description
Follow-up to #709, fixing the stale claims it did not cover, found while
verifying every README example against real tables ahead of the 0.5.0 release:
- The Rust snapshot example never compiled: `println!("{}: {}", col_name,
batch.column(idx))` fails E0277 because `Arc<dyn Array>` has no `Display` impl.
Switched to `{:?}` (and the inline `{col_name}` arg).
- `compute_table_stats` was described as aggregating from changed file
slices for incremental queries; it returns `None` for incremental. Reworded to
match the code and the reader spec, including the conditions under which the
estimate is unavailable.
- "Both args are optional" on `create_file_group_reader_with_options` holds
in Python only; in Rust the second arg is a required iterator.
- The Daft integration link now points at its current canonical page instead
of redirecting twice.
- The reader spec still said file slices without a base file are not
supported; log-only groups are readable under file group reader version 2 since
#660/#706.
- The `read_file_slice_stream` docstring in `_internal.pyi` still described
the version 1 single-merged-batch fallback as current behavior; version 2 (the
default) streams merged chunks.
- The AGENTS.md layout block predates `crates/jvm-ffi` and
`benchmark/filegroup` and listed only part of hudi-core's modules; the
config-family list gains `HudiPlanConfig`; python/AGENTS.md's testing example
called a `read_snapshot` method that does not exist in the Python API.
## How are the changes test-covered
- [ ] N/A
- [ ] Automated tests (unit and/or integration tests)
- [x] Manual tests
- [x] Details are described below
Compiled and ran the corrected Rust example (and the original, which fails
E0277) against a Spark-written quick-start table; ran the corrected Python
example pattern likewise. `ruff format --check` and `ruff check` pass on the
python directory.
--
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]
