[V4RB] RegEx search and UTF-8

[Peter De Berdt]

On 22 Jul 2004, at 11:33, Ruslan Zasukhin wrote:

> On 7/22/04 11:46 AM, "Eric Ferrer" <wonderfef at noos.fr> wrote:
>
> Hi Eric,
>
>> Thanks to the power of Valentina and RegEx, I was able to add a very=20=

>> cool
>> "find anything" tool to my application, i.e:
>>
>> type "borea" and Valentina will find "Bor=8Eales" using
>> Select Name from Clients where name like 'b[o=9A=99=98]r[e=90=91=8E=8F]=
[a=89=8A=88]'=20
>> no_case
>>
>> But now that I'm using RB 5.5, "Bor=C3=A9ales" is stored in my record=20=

>> instead of
>> "Bor=8Eales", things are getting tricky...
>
> I think you should in RB convert from UTF8 to single byte enoding=20
> before you
> store into Valentina.
>
> This was mentioned on list few times
>
Sadly enough, this doesn't work at all. You have to keep it UTF-8, but=20=

make sure you use DefineEncoding(string,encodings.UTF8) for every query=20=

sent and record retrieved from the database.

Best regards

Peter De Berdt=