Joe Conway <[EMAIL PROTECTED]> writes:
> A first shot at documentation for generate_series() is available here in
> html form:
> http://www.joeconway.com/functions-srf.html
> Feedback welcome.
This bit seems unnecessarily vague:
Depending on the requested combination of start, stop, and step,
Joe Conway <[EMAIL PROTECTED]> writes:
> Tom Lane wrote:
>> folklore has it that Mariner II was lost to exactly such a bug).
> Ouch -- got the point.
BTW, I think I was beating you over the head with an urban legend.
Some idle googling revealed the true facts of the Mariner failure:
http://www.rc
Joe Conway <[EMAIL PROTECTED]> writes:
> The one corner case not discussed is a step size of zero. Currently it
> returns zero rows, but I considered having it generate an ERROR.
I'd go for ERROR --- can't think of any reason to do otherwise, nor
any standard programming language that wouldn't co
Joe Conway <[EMAIL PROTECTED]> writes:
> Tom Lane wrote:
>> Joe Conway <[EMAIL PROTECTED]> writes:
>>> regression=# select * from pg_generate_sequence(8, 4);
>>> ERROR: finish is less than start
>>
>> Hm, would it be better just to return an empty set? Certainly I'd
>> expect pg_generate_sequenc
Joe Conway <[EMAIL PROTECTED]> writes:
> regression=# select * from pg_generate_sequence(8, 4);
> ERROR: finish is less than start
Hm, would it be better just to return an empty set? Certainly I'd
expect pg_generate_sequence(1,0) to return an empty set with no error.
> regression=# select * fro
