Github user eyalfa commented on the issue:
https://github.com/apache/spark/pull/14444
Yes,as I wrote earlier:
When creating the named struct,we extract name from the values by pattern
matching named expressions. We can add a special case for star and treat it
as unnamed. This will produce a 'col1->star's pair which will later be
resolved to a collection of the actual fields once the analyzer gets to it.
The analyzer rule that expands star in named struct,ignores the column name
and replaces the name-value pair with a list of the resolved fields.
On Sep 15, 2016 2:51 PM, "Wenchen Fan" <[email protected]> wrote:
> although I don't think struct should respect the explicit aliased names,
> I agree we shouldn't change it to avoid breaking existing query. Do you
> have an idea to fix #14444 (comment)
> <https://github.com/apache/spark/pull/14444#r78928177>?
>
> â
> You are receiving this because you were mentioned.
> Reply to this email directly, view it on GitHub
> <https://github.com/apache/spark/pull/14444#issuecomment-247307042>, or
mute
> the thread
>
<https://github.com/notifications/unsubscribe-auth/ABFFOddwjbtqJarD7w3cmwyZr0R66rlLks5qqTE6gaJpZM4JaDJr>
> .
>
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
