geographybuff opened a new pull request, #11034:
URL: https://github.com/apache/arrow-rs/pull/11034
# Which issue does this PR close?
N/A — small standalone cleanup, following the precedent of recent
comment-typo PRs merged without a linked issue:
- #10956 `chore: Fix typos` (47 files, +108/-107)
- #10943 `chore: fix typos in code comments` (11 files, +16/-16)
- #10530 `chore: fix typo in release email template`
# Rationale for this change
Several comments contain accidentally doubled words — `run-encoded encoded
data`, `and and an offset`, `typed typed on`, `are used used`. Most are rustdoc
on public items, so they render on docs.rs.
These survive automated checking because the CI spell check
(`.github/ci/scripts/typos_check.sh`, the `typos` crate) matches misspelled
*words* and does not detect a correctly spelled word repeated twice in a row.
`typos --config typos.toml` passes on both the unmodified tree and this branch.
One case is more than cosmetic. In
`parquet/src/file/metadata/thrift/mod.rs`, the `PageHeader` doc comment is a
transcription of the Parquet Thrift IDL, and reads `to be be calculated as
follows`. Upstream `parquet-format`'s `parquet.thrift` reads `to be
calculated`, so this is a transcription slip and the fix restores agreement
with the spec.
Found by searching the tree for `\b(\w+)\s+\1\b` restricted to comment and
prose lines. The remaining matches are ASCII-art layout diagrams in
`list_array.rs` / `list_view_array.rs` (`Logical Logical` column headings)
and correct English like `100-row row group`; those are left alone.
# What changes are included in this PR?
Removes 13 doubled words across 11 Rust files and 1 Markdown file. Comment
and prose text only — no code changes.
| File | Fix |
| --- | --- |
| `arrow-buffer/src/lib.rs` | `Run-ends used in run-encoded encoded data` |
| `arrow-buffer/src/util/bit_chunk_iterator.rs` | `and and an offset and
length in bits` |
| `arrow-array/src/array/run_array.rs` | `A RunArray typed typed on its
child values array` |
| `arrow-data/src/data.rs` | `2. the the null count is correct` |
| `arrow-data/src/transform/mod.rs` | `Note this this is not stored in
_MutableArrayData` |
| `parquet/src/file/properties.rs` | `default values for ndv and fpp value
are used used` |
| `parquet/src/file/metadata/mod.rs` (x2) | `Takes ownership of the the
column metadata`; `Returns a reference to the the histogram's values` |
| `parquet/src/file/metadata/thrift/mod.rs` | `CRC checksum for the page, to
be be calculated` |
| `parquet/src/arrow/mod.rs` | `the AES encryption keys required required
for decrypting` |
| `parquet/src/arrow/arrow_reader/mod.rs` | `maximum size of value to pass
to to value generator` |
| `parquet/src/arrow/arrow_reader/statistics.rs` | `extracting i128 values
from from an iterator` |
| `parquet_derive/README.md` | `You must must enable the chrono feature` |
# Are these changes tested?
No new tests — the change is confined to comments and prose, so there is no
behaviour to cover.
Checks run locally on Linux x86_64 with the pinned 1.98.1 toolchain:
- `typos --config typos.toml` (v1.37.0, the version pinned in `dev.yml`) —
pass
- `cargo fmt --all --check` — pass
- `cargo test --doc -p arrow-buffer -p arrow-array -p arrow-data -p parquet
--all-features` — pass, 376 doctests, 0 failed (arrow-array 206, parquet 96,
arrow-buffer 60, arrow-data 14)
This first failed with 5 errors, all of them `PARQUET_TEST_DATA is
undefined` from an uninitialised `parquet-testing` submodule rather than
anything in this PR. The numbers above are from the re-run after `git submodule
update --init`.
`parquet_derive/README.md` is outside the prettier glob in `dev.yml`
(`{arrow,arrow-flight,dev,arrow-integration-testing,parquet}/**/*.md`), so the
docs formatting job is unaffected.
I did not run the full test suite. One edited line sits inside a rustdoc
example (`parquet/src/arrow/mod.rs`, the encryption example), and that doctest
is covered by the run above; every other edited line is an ordinary comment.
# Are there any user-facing changes?
No API or behaviour changes. The rendered rustdoc is corrected for
`RunEndBuffer`, `UnalignedBitChunk::new`, `Int64RunArray`, `ArrayData`,
`WriterProperties`, `ColumnChunkMetaDataBuilder::take_column_metadata`,
`Histogram::values`, `PageHeader`, and two `parquet::arrow` reader items.
---
## AI disclosure
This contribution was AI-assisted. Claude Code (Claude Opus 5) performed the
tree-wide doubled-word search, made the 13 line edits, checked the `PageHeader`
comment against upstream `parquet-format`, and ran every check listed above.
The contributor reviewed the diff and the check output before this PR was
opened; the check commands themselves were executed by the AI and were not
independently re-run by hand.
--
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]
