kylebarron commented on PR #35780:
URL: https://github.com/apache/arrow/pull/35780#issuecomment-1843288624

   > Is that a problem? Looking at the C++ docs at 
https://arrow.apache.org/docs/cpp/api/array.html#classarrow_1_1_chunked_array 
it seems like it should be okay if chunk lengths are not consistent.
   
   It's not the ChunkedArray you want to look at. That indeed allows each 
`Data` instance within a _single_ `Vector` to have different lengths. But if 
you look at the [doc for 
`RecordBatch`](https://arrow.apache.org/docs/cpp/api/table.html#_CPPv4N5arrow11RecordBatchE)
 it indeed says
   
   > Collection of **equal-length arrays** matching a particular 
[Schema](https://arrow.apache.org/docs/cpp/api/datatype.html#classarrow_1_1_schema)
   
   (emphasis mine). So two columns can't have different lengths within a single 
`Batch`. 
   
   Maybe you're right that this is effectively the same as what pyarrow is 
doing. It's hard to tell if those batches are views on the existing buffers, 
like the slicing Arrow JS's `batches` have on the existing buffers.


-- 
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]

Reply via email to