Re: chop off non-meaningful digits

A. Kretschmer wrote:
> am  Tue, dem 14.11.2006, um  0:58:56 -0500 mailte Tom Lane folgendes:
>
>> "SunWuKung" <Balazs.Klein@t-online.hu> writes:
>>
>>> Yep, I think this is it:
>>> select trim(trailing '0.' from 1.020)
>>>
>> Um, I think not:
>>
>> regression=# select trim(trailing '0.' from 1000.000);
>>  rtrim
>> -------
>>  1
>> (1 row)
>>
> For this case:
>
> test=*> select trim(trailing '.' from trim(trailing '0' from 1000.000));
>  rtrim
> -------
>  1000
>

Perhaps the OP doesn't really care about this, but FWIW one thing I
recall from high school science (or maybe it was middle school), the
zeros to the right of the decimal place ARE meaningful, i.e., 1000.000
signifies something different than 1000, namely, a measurement three
orders of magnitude more precise.  So "chopping off non-meaningful
digits" is something you ought to do only for zeros on the left.