Re: String comparison problem.

["Scott Marlowe" <scott(dot)marlowe(at)gmail(dot)com>]

On Thu, May 29, 2008 at 9:45 PM, Theerasak Maneeneim
<theerasak(at)thns(dot)co(dot)th> wrote:
> Dear All,
>
>           I have some problem with string comparison. I use postgresql
> 8.3.1,UTF8 encoding, on CentOS 5.5. Why dose the result of  SQL statement :
>  "select '1.1' < '101';" is false but "select '1.1'::bytea < '101'::bytea;"
> is ture?

The UTF encoding isn't what's important here, it's your locale, which
is likely something like en_US (but different, whatever you'd have in
your country).  This locale will be collated (i.e. sorted etc.) by a
set of rules for the language where you live.

For instance, with a list like:

 a dc
 abc
 a bc
 adc

and an order by on it in locale en_US I get this ordering:

 abc
 a bc
 adc
 a dc

Why?  Because according to the rules of english, spaces don't count.

If you want byte code ordering, then you need to use Locale=C when you
run initdb.  Locale can only be changed at initdb time.