Re: Add generate_series(date,date) and generate_series(date,date,integer)

Christoph Berg <myon(at)debian(dot)org> writes:
> Re: David Fetter 2016-01-26 <20160126180011(dot)GA16903(at)fetter(dot)org>
>> +1 for back-patching. There's literally no case where an infinite
>> input could be correct as the start or end of an interval for
>> generate_series.

> select * from generate_series(now(), 'infinity', '1 day') limit 10;
> ... seems pretty legit to me. If limit pushdown into SRFs happened to
> work some day, it'd be a pity if the above query raised an error.

Oooh ... actually, that works today if you consider the SRF-in-targetlist
case:

regression=# select generate_series(now(), 'infinity', '1 day') limit 10;
generate_series
-------------------------------
2016-02-21 16:51:03.303064-05
2016-02-22 16:51:03.303064-05
2016-02-23 16:51:03.303064-05
2016-02-24 16:51:03.303064-05
2016-02-25 16:51:03.303064-05
2016-02-26 16:51:03.303064-05
2016-02-27 16:51:03.303064-05
2016-02-28 16:51:03.303064-05
2016-02-29 16:51:03.303064-05
2016-03-01 16:51:03.303064-05
(10 rows)

Time: 8.457 ms

Given that counterexample, I think we not only shouldn't back-patch such a
change but should reject it altogether.

regards, tom lane