Re: unnest on multi-dimensional arrays

From: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
To: David Johnston <polobo(at)yahoo(dot)com>
Cc: pgsql-general(at)postgresql(dot)org
Subject: Re: unnest on multi-dimensional arrays
Date: 2013-11-28 15:03:47
Message-ID: 27706.1385651027@sss.pgh.pa.us
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-general

David Johnston <polobo(at)yahoo(dot)com> writes:
> Zev Benjamin wrote
>> It appears that unnest, when called on a multi-dimensional array,
>> effectively flattens the array first. For example: ...

> Multidimensional arrays do have shortcomings in the current implementation
> of which this is one. I'm not sure, though, if there is anything
> substantial and centralized in the docs so pertaining.

It might be worth explaining that this is a consequence of the fact that
Postgres treats all arrays over the same element type as being of the
same data type --- that is, 1-D and 2-D arrays are not distinguished
by the type system. Thus, when the polymorphic function "unnest(anyarray)
returns setof anyelement" is applied to an integer array, it must return
a series of integers; not a series of lower-dimensional arrays.

There have been some discussions over whether this could be changed
without a backwards-compatibility disaster, but nobody sees how.

regards, tom lane

In response to

Responses

Browse pgsql-general by date

  From Date Subject
Next Message Bruce Momjian 2013-11-28 15:39:18 Re: [GENERAL] pg_upgrade ?deficiency
Previous Message Pavel Stehule 2013-11-28 11:16:45 Re: Will be Wildspeed inside Postgresql contrib?