Re: SQL/JSON: FOR ORDINALITY bug

On Tue, May 3, 2022 at 5:27 PM Andrew Dunstan <andrew(at)dunslane(dot)net> wrote:

>
> On 2022-05-03 Tu 11:19, Erik Rijkers wrote:
> > Hi
> >
> > I've copied some statements from the .pdf called:
> > "TECHNICAL REPORT ISO/IEC TR 19075-6 First edition 2017-03
> > Part SQL Notation support 6: (JSON) for JavaScript Object"
> > (not available anymore although there should be a similar replacement
> > file)
> >
> > In that pdf I found the data and statement (called 'table 15' in the
> > .pdf) as in the attached bash file. But the result is different: as
> > implemented by 15devel, the column rowseq is always 1. It seems to me
> > that that is wrong; it should count 1, 2, 3 as indeed the
> > example-result column in that pdf shows.
> >
> > What do you think?
> >
> >
>
> Possibly.
>
>
I don't see how rowseq can be anything but 1. Each invocation of
json_table is given a single jsonb record via the lateral reference to
bookclub.jcol. It produces one result, having a rowseq 1. It does this
for all three outer lateral reference tuples and thus produces three output
rows each with one match numbered rowseq 1.

David J.