Re: BUG #5869: postgresql corrupts unquoted non-ascii chars in column aliases

On Thu, Mar 3, 2011 at 11:51 AM, David Schmitt <david(at)dasz(dot)at> wrote:
> On 03.03.2011 16:08, Robert Haas wrote:
>>
>> On Tue, Feb 8, 2011 at 2:19 AM, David Schmitt<david(at)dasz(dot)at>  wrote:
>>>
>>> Example query:
>>>
>>>  SELECT column as zurück FROM table;
>>>
>>> results in corruption of the "ü" (umlaut u). This causes Npgsql to fail
>>> to
>>> match up the columns in the result set (see
>>>
>>> http://pgfoundry.org/tracker/?func=detail&aid=1010988&group_id=1000140&atid=
>>> 590) and pgadminIII doesn't display the alias in the result pane.
>>
>> That link doesn't show anything interesting when I look at it.  What
>> exactly do you mean by corrupted?
>
> There are some garbage bytes instead of the umlaut.

I'm not really an encoding guy, but I would think that someone who is
would want to know exactly what bytes you were ending up with.

--
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company