Обсуждение: Comma Comma Comma 8601

Hackers,

According to [Wikipedia](https://en.wikipedia.org/wiki/ISO_8601#Times):

> Decimal fractions may also be added to any of the three time elements. A decimal mark, either a comma or a dot
(withoutany preference as stated in resolution 10 of the 22nd General Conference CGPM in 2003,[11] but with a
preferencefor a comma according to ISO 8601:2004)[12] is used as a separator between the time element and its fraction.
Afraction may only be added to the lowest order time element in the representation. To denote "14 hours, 30 and one
halfminutes", do not include a seconds figure. Represent it as "14:30,5", "1430,5", "14:30.5", or "1430.5". There is no
limiton the number of decimal places for the decimal fraction. However, the number of decimal places needs to be agreed
toby the communicating parties. 

I assume that the Postgres project has no interest in supporting the input of whack times like “14:30.5”, “1430.5”,
“14:30.5”,or “1430.5”, right? I mean, that’s just bizarre, amirite? 

But I do wonder if the comma should be allowed for fractional seconds, since the spec says it is preferred (and often
usedin Javaland, I’m told). As in “14:30:50,232”. Thoughts? 

Best,

David

PS: Apologies if you somehow ended up with a bad 80s pop song in your head.

"David E. Wheeler" <david@justatheory.com> writes:
> But I do wonder if the comma should be allowed for fractional seconds,
> since the spec says it is preferred (and often used in Javaland, I'm
> told). As in "14:30:50,232". Thoughts?

Does that create any ambiguities against formats we already support?
I'm worried about examples like this one:

select 'monday, july 22, 22:30 2013'::timestamptz;     timestamptz       
------------------------2013-07-22 22:30:00-04
(1 row)

Right now I'm pretty sure that the datetime parser treats comma as a
noise symbol.  If that stops being true, we're likely to break some
applications out there (admittedly, possibly rather strange ones,
but still ...)
        regards, tom lane

On Jul 23, 2013, at 1:17 AM, Tom Lane <tgl@sss.pgh.pa.us> wrote:

> Does that create any ambiguities against formats we already support?
> I'm worried about examples like this one:
>
> select 'monday, july 22, 22:30 2013'::timestamptz;
>      timestamptz
> ------------------------
> 2013-07-22 22:30:00-04
> (1 row)
>
> Right now I'm pretty sure that the datetime parser treats comma as a
> noise symbol.  If that stops being true, we're likely to break some
> applications out there (admittedly, possibly rather strange ones,
> but still ...)

I kind of suspect not, since this fails:

david=# select '12:24:53 654'::time;
ERROR:  invalid input syntax for type time: "12:24:53 654"
LINE 1: select '12:24:53 654'::time;              ^

I would have guessed that the time parser gets to a state where it knows it is dealing with a ISO-8601-style time. But
Ihave not looked at the code, of course. 

David

On Jul 23, 2013, at 6:24 PM, David E. Wheeler <david@justatheory.com> wrote:

> I kind of suspect not, since this fails:
>
> david=# select '12:24:53 654'::time;
> ERROR:  invalid input syntax for type time: "12:24:53 654"
> LINE 1: select '12:24:53 654'::time;
>               ^
>
> I would have guessed that the time parser gets to a state where it knows it is dealing with a ISO-8601-style time.
ButI have not looked at the code, of course. 

I added this to the To Dos so it won’t get lost.
 https://wiki.postgresql.org/wiki/Todo#Dates_and_Times

Best,

David