westonpace opened a new pull request #9626:
URL: https://github.com/apache/arrow/pull/9626
When converting a struct with chunks to python the ownership of the arrow
arrays was not being properly tracked and the deletion of the resulting pandas
dataframe would leave some buffers behind. I believe it was something like
this (pseudo-code)...
```
array_tracker** refs[num_fields];
for chunk in chunks:
*refs[chunk.index] = convert(...)
for row in rows:
row.owns.append(refs[chunk.index])
```
----------------------------------------------------------------
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.
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
