Alvaro Herrera <alvherre@xxxxxxxxxxxxxx> writes: > On Wed, Sep 07, 2005 at 11:40:57PM -0700, Pradeepkumar, Pyatalo (IE10) wrote: >> Even if somehow if I convert wchar_t * to char * and provide it to >> Postgres, how will it differentiate between chinese characters and normal >> characters ?? > Please note that any string you pass from the client will be interpreted > using the client_encoding setting. So maybe what you have to do is just > set client_encoding to the encoding that the strings are encoded in. I think his problem is that his data is represented in UCS2 or UCS4 (or some other fixed-width representation) and we don't support those as client_encoding values AFAIK. He'll need to do an encoding conversion to some representation we do support, such as UTF8. regards, tom lane
