asfimport

[Micah Kornfield](https://issues.apache.org/jira/browse/ARROW-12099?focusedCommentId=17309732) / @emkornfield: Note there is a flatten o[n list array](https://arrow.apache.org/docs/python/generated/pyarrow.ListArray.html#pyarrow.ListArray.flatten) which sounds very similar to explode?

[Malthe Borch](https://issues.apache.org/jira/browse/ARROW-12099?focusedCommentId=17309744): @emkornfield that looks helpful but what's very useful about Spark's explode is that it works directly on the table level. I'm not sure how easy it would be...

[Micah Kornfield](https://issues.apache.org/jira/browse/ARROW-12099?focusedCommentId=17309766) / @emkornfield: OK, does explode do so recursively or only for top-level types.  It would be good to formally specify this. 

[Malthe Borch](https://issues.apache.org/jira/browse/ARROW-12099?focusedCommentId=17309767): @emkornfield added; there is no support for recursively exploding a structure. I think that is a reasonable position to take here as well.

[Joris Van den Bossche](https://issues.apache.org/jira/browse/ARROW-12099?focusedCommentId=17310717) / @jorisvandenbossche: In the `pyarrow.compute` module, we also have `list_parent_indices` which can be used to join the flattened list column with repeated rows of the original...

[Malthe Borch](https://issues.apache.org/jira/browse/ARROW-12099?focusedCommentId=17310753): @jorisvandenbossche in Spark, explode does not "zip" arrays in different columns actually – it just copies the entire row for each value in the exploded column (which is...

[Joris Van den Bossche](https://issues.apache.org/jira/browse/ARROW-12099?focusedCommentId=17310772) / @jorisvandenbossche: Your description to me sounds more or less what I showed. Can you eg show an example of input / expected result? (as you...

[Malthe Borch](https://issues.apache.org/jira/browse/ARROW-12099?focusedCommentId=17310781):   So Spark cannot actually explode (or "generate") more than one expression per select statement (that is simply not allowed), but I suppose sometimes you want to "zip" the...

[Joris Van den Bossche](https://issues.apache.org/jira/browse/ARROW-12099?focusedCommentId=17310784) / @jorisvandenbossche: Thanks. Could you also post what the FROM clause looks like? (`SELECT explode(sequence(0, 2)) a, sequence(4, 6) b`) (to be sure how the input...

[Malthe Borch](https://issues.apache.org/jira/browse/ARROW-12099?focusedCommentId=17310789): What I can perhaps add is that I imagine an API such as: ```python table.explode("a") # This would unroll the "a" array, but leave any other array as...