Tzafrir Cohen <[EMAIL PROTECTED]> wrote on 31/7/02 17:25:
>On Wed, 31 Jul 2002, Oded
>Arbel wrote:
>
>> BTW: I prefer storing
>textual data in the database
>as unicode
>> (preferably utf8 to
>facilitate easier display to
>the web) encoded data
>> in binary fields - it gives
>predictable enough sorting,
>and you neednot
>> worry about character
>sets, especially when aiming
>for multi-lingual
>> applications (and I do
>consider english/hebrew
>being multi-lingual
>> enough to warrant
>unicode).
>>
>
>What if there is some English
>text? case insensitive search
>is quite
>problematic in some cases
>('aaa' comes after 'ZZZ').
true, but if you are willing to accept that (and I don't think its such a problem for
most uses), then you are in the clear - no matter what encoding you choose, as long as
you are being consistent about it.
>Also, if the UTF8 text
>includes nikud it has to be
>ignored-away during the
>sorting.
why ? though have never tried sorting on text with nikud, but I fail to see the
problem : why should it matter if alef with 'segol' sorts after alef with 'kamatz'?
>IIRC proper utf collating
>would do the above two (this
>costs some extra
>cpu cycles even when this
>collating is not used, I
>believe).
Does mysql know about utf-8 ?
--
Oded
=================================================================
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word "unsubscribe" in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]
