Repository navigation
Conversation
github-actions
Bot
added
logical-expr
This was referenced Oct 27, 2026
Closed
alamb
pushed a commit
to Rich-T-kid/arrow-rs
that referenced
this pull request
Oct 30, 2026
…et-variant` (apache#12005) # Which issue does this PR close? - Closes apache#11952. # Rationale for this change Using the `arrow.parquet.variant` extension type currently requires depending on `parquet-variant-compute`, which pulls in the full `arrow` crate and `parquet-variant-json`. Schema-only consumers such as `datafusion-common` and `datafusion-expr` (apache/datafusion#26822) only need the type itself, which depends on nothing beyond `arrow-schema`. # What changes are included in this PR? `VariantType` moves unchanged from `parquet-variant-compute/src/variant_array.rs` to a new `parquet-variant/src/extension.rs` and is exported as `parquet_variant::VariantType`. `parquet-variant-compute` re-exports it, so `parquet_variant_compute::VariantType` and `parquet::variant::VariantType` keep working and refer to the same type. # Are these changes tested? Yes. Existing uses through the old paths (the `VariantArray` doctests, `variant_get`, and the `parquet` Variant schema tests) are unchanged and still pass. A new unit test in `parquet-variant` covers the type on its own: a struct field accepts the extension and a non-struct field is rejected with the existing error message. # Are there any significant user-facing changes? `VariantType` is now also available from `parquet-variant`. The existing import path is kept, so this is not a breaking change. 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-authored-by: Claude Opus 5.5 <[email protected]>
Contributor
|
I think apart from moving From what I can gather, the display implementation is mostly done in datafusion-variant. Should we add this code to this PR or do a follow-up? |
Member
Author
I thought about doing it separately to make the review process easier 😅 |
This was referenced Nov 13, 2026
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Which issue does this PR close?
Closes #21144. Follows #21291.
Rationale for this change
Add
arrow.parquet.variantto DataFusion's canonical extension type registry, following the thin wrappers used for JSON and other canonical types.Draft: waiting on moving Arrow's
VariantTypefromparquet-variant-computeintoparquet-variant, and adopting an Arrow release containing that move. The current dependency pulls in the compute and JSON crates just to register the extension. Before merging, this PR should depend only onparquet-variantfor that type.What changes are included in this PR?
Add
DFVariantand its canonical registration. The wrapper retains the storage datatype and delegates validation and metadata handling to Arrow'sVariantType. Values use ordinary storage-type formatting.Arrow's incomplete storage-field validation is tracked separately in apache/arrow-rs#11938.
What is the testing strategy for this PR?
No new unit tests for the thin delegation wrapper, consistent with the unformatted extensions in #21291.
Formatting, strict workspace Clippy (
--all-targets --all-features),cargo check -p datafusion-common -p datafusion-expr --offline, and all 1,471 core integration tests passed. The required extended workspace run stopped after one sort memory-limit failure (154 extended tests passed); that test passed when rerun alone. Later workspace test binaries did not run.Are there any user-facing changes?
Yes. The canonical registry recognizes
arrow.parquet.variant, andDFVariantis available throughdatafusion_common::types.