nssalian opened a new pull request, #859: URL: https://github.com/apache/arrow-go/pull/859
### Rationale for this change A shredded variant array reads back in the shredded storage layout, which shares the `parquet.variant` extension name with the non-shredded layout but has a different storage type. There was no way to collapse it short of decoding row by row. iceberg-go needs this at scan-projection time to read a shredded variant column identically to a non-shredded one (apache/iceberg-go#986). ### What changes are included in this PR? - Adds `extensions.UnshredVariant(arr, mem)`, returning an equivalent non-shredded `VariantArray`. An already-non-shredded array is returned unchanged with an added reference. - Fixed a typo in `UntypedValues()`'s doc string too ### Are these changes tested? Yes. `TestUnshredVariant` covers scalar, list, nested object, residual fields, null vs present variant-null, and passthrough. ### Are there any user-facing changes? One new exported function. No change to existing behavior. -- 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]
