Re: BUG #1517: SQL interval syntax is accepted by the parser,

Roy Badami wrote:
> Roy> The 'constraint' (interval type descriptor or whatever it's
> Roy> really called) is mandatory in standard SQL, I think, so
> Roy> there's no ambiguity anyway, unless anyone is using this
> Roy> undocumented syntax at the moment...
>
> Incidentally, this was the ratinale behind my earlier suggestion, that:
>
> * if the interval type descriptor is absent, parse the interval as a
> traditional postgres interval
>
> * if the interval type descriptor is present, parse the interval according
> to the standard syntax
>
> I have no objection to allowing things like
>
> '1 hour 10 minutes' DAY TO SECOND
>
> but I'm just wondering whether the hybrid syntax is an unnecessary
> complication.

Added to TODO:

o For syntax that isn't uniquely ISO or PG syntax, like '1:30' or
'1', treat as ISO if there is a range specification clause,
and as PG if there no clause is present, e.g. interpret '1:30'
MINUTE TO SECOND as '1 minute 30 seconds', and interpret '1:30'
as '1 hour, 30 minutes'

This brings up the issue of how we are to output interval values.
Currently we do:

test=> select interval '1 hour 1 minute';
interval
----------
01:01:00
(1 row)

and

test=> select interval '1 hour 1 second' hour to minute;
interval
----------
01:00:00
(1 row)

so I think we are OK because we don't output ambiguous syntax.

--
Bruce Momjian | http://candle.pha.pa.us
pgman(at)candle(dot)pha(dot)pa(dot)us | (610) 359-1001
+ If your life is a hard drive, | 13 Roberts Road
+ Christ can be your backup. | Newtown Square, Pennsylvania 19073