On Thu, Sep 24, 2020 at 03:05:24PM -0400, Tom Lane wrote:
> Bruce Momjian <bruce@momjian.us> writes:
> > Yes, this is very helpful. I was afraid the JSON would overwhelm the
> > example, but this looks good. I wrote the attached doc patch which I
> > think improves this. I plan to apply it to all supported versions.
>
> Couple thoughts:
>
> * Taking the initial ... out of the syntax synopsis is not an improvement.
> It makes it look like you can only apply AS to the first function of a
> ROWS FROM.
Oh, so the dots represent optional non-column_definition function calls.
I can't think if a cleaner way to show that, so I guess "..." will have
to do.
> * I think the ORDER BY adds nothing to the example except complication
> and confusion.
I wanted to highlight that the column_definition specifies the data type
of the column in other parts of the query.
> * Maybe the other sentence of explanation would read better as
>
> json_to_recordset() is instructed to return two columns,
> the first integer and the second text. The result of
> generate_series() is used directly.
OK, better. New patch attached.
--
Bruce Momjian <bruce@momjian.us> https://momjian.us
EnterpriseDB https://enterprisedb.com
The usefulness of a cup is in its emptiness, Bruce Lee
