jorisvandenbossche commented on code in PR #37797:
URL: https://github.com/apache/arrow/pull/37797#discussion_r1354859901
##########
python/pyarrow/table.pxi:
##########
@@ -2983,6 +2985,100 @@ cdef class RecordBatch(_Tabular):
<ArrowArray*> c_ptr, c_schema))
return pyarrow_wrap_batch(c_batch)
+ def __arrow_c_array__(self, requested_schema=None):
+ """
+ Get a pair of PyCapsules containing a C ArrowArray representation of
the object.
+
+ Parameters
+ ----------
+ requested_schema : PyCapsule | None
+ A PyCapsule containing a C ArrowSchema representation of a
requested
+ schema. PyArrow will attempt to cast the batch to this schema.
+ If None, the schema will be returned as-is, with a schema matching
the
+ one returned by :meth:`__arrow_c_schema__()`.
+
+ Returns
+ -------
+ Tuple[PyCapsule, PyCapsule]
+ A pair of PyCapsules containing a C ArrowSchema and ArrowArray,
+ respectively.
+ """
+ cdef:
+ ArrowArray* c_array
+ ArrowSchema* c_schema
+
+ if requested_schema is not None:
+ target_schema = Schema._import_from_c_capsule(requested_schema)
+
+ if target_schema != self.schema:
+ try:
+ # We don't expose .cast() on RecordBatch, only on Table.
+ casted_batch = Table.from_batches([self]).cast(
+ target_schema, safe=True).to_batches()[0]
+ inner_batch = pyarrow_unwrap_batch(casted_batch)
+ except ArrowInvalid as e:
+ raise ValueError(
+ f"Could not cast {self.schema} to requested schema
{target_schema}"
Review Comment:
My suggestion was mostly for improving clarity (I personally think it will
be easier to understand if the final error message has the information about
why the cast failed, than having to look somewhere inside the traceback, which
can be quite long in IPython, in contrast to the concise example in the python
docs), and for consistency, as the same is done for the Array version.
--
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]
