v0y4g3r opened a new pull request, #2855:
URL: https://github.com/apache/arrow-datafusion/pull/2855
# Which issue does this PR close?
Closes #2709.
# Rationale for this change
Current branch `arrow2` is still using arrow2 version `v0.10` and falls far
behind the latest version `v0.12`
# What changes are included in this PR?
Dependencies upgraded:
- arrow2: `v0.10` -> `v0.12`
- parquet2: `v0.12` -> `v0.13`
- arrow-format: `0.4`->`0.6`
- prost: `0.9` -> `0.10 `
- prost-types: `0.9` -> `0.10`
- tonic: 0.6 -> `0.7`
- topic-build: `0.6` -> `0.7`
API change:
- `arrow2::error::ArrowError` -> `arrow2::error::Error`
- arrow2 FileWriter now accepts `Vec<Vec<Encoding>>` instead of
`Vec<Encoding>` and `transverse` can be used to create encodings from schema
with a customized mapping.
- arrow2 `RowGroupMetadata` no longer as a `column(usize)` method to fetch
column metadata at given index, instead, it provides a `columns()` method that
returns a column slice.
- `datafusion::dataframe::DataFrame::write_parquet` method now accepts
`arrow::io::parquet::write::WriteOptions` instead of
`parquet::write::WriteOptions`
- some other minor changes.
# Are there any user-facing changes?
Yes, please refer to previous section for full API change list.
But I think it's rather easy for users of `arrow2` branch to adapt to these
changes.
# About tests
Some tests are failling in current arrow2 branch, so I'm not going to fix
all failing tests in this PR, instead I'll make sure no more unit test fails.
I'd be happy to fix all failing tests after this PR is merged so I don't have
to fix them twice :)
--
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]
