David Fetter wrote:
> That depends on what you mean by "dynamic columns." The approach
> taken in the tablefunc extension is to use functions which return
> SETOF RECORD, which in turn need to be cast at runtime.
For me, "PIVOT with dynamic columns" would be a pivot query
whose output columns are not enumerated as input in the
SQL query itself, in any form.
> The second, more on point, is to specify a serialization for the rows
> in the "dynamic columns" case. Their syntax is "PIVOT XML", but I
> would rather do something more like "PIVOT (SERIALIZATION XML)".
The SERIALIZATION looks interesting, but I believe these days JSON
would make more sense than XML, both as easier for the client-side and
because of all the json_* functions we now have to mix json with
relational structures.
Best regards,
--
Daniel Vérité
PostgreSQL-powered mailer: http://www.manitou-mail.org
Twitter: @DanielVerite