Обсуждение: BUG #13536: SQLParamData thows "Invalid Endian" error

Поиск
Список
Период
Сортировка

BUG #13536: SQLParamData thows "Invalid Endian" error

От
lthompson@alteryx.com
Дата:
The following bug has been logged on the website:

Bug reference:      13536
Logged by:          Linda Thompson
Email address:      lthompson@alteryx.com
PostgreSQL version: 9.4.1
Operating system:   windows
Description:

With the 9.2 ODBC driver and many previous versions my SQLParamData worked
just fine.  When we went to 9.3.3 and 9.3.4 ODBC Drivers, we can no longer
add Spatial objects.  The second call to SQLParamData to add the next
Spatial object throws an error "Invalid Endian".  However if we do 1 spatial
object at a time, we can add the exact same object with the exact same code.
 I believe there is a bug somewhere in the new ODBC driver throwing the
wrong error code.  IT should just be returning SQL_NEED_DATA or "done"

Re: BUG #13536: SQLParamData thows "Invalid Endian" error

От
Heikki Linnakangas
Дата:
(Moving to the pgsql-odbc mailing list)

On 08/03/2015 09:23 PM, lthompson@alteryx.com wrote:
> The following bug has been logged on the website:
>
> Bug reference:      13536
> Logged by:          Linda Thompson
> Email address:      lthompson@alteryx.com
> PostgreSQL version: 9.4.1
> Operating system:   windows
> Description:
>
> With the 9.2 ODBC driver and many previous versions my SQLParamData worked
> just fine.  When we went to 9.3.3 and 9.3.4 ODBC Drivers, we can no longer
> add Spatial objects.  The second call to SQLParamData to add the next
> Spatial object throws an error "Invalid Endian".  However if we do 1 spatial
> object at a time, we can add the exact same object with the exact same code.
>   I believe there is a bug somewhere in the new ODBC driver throwing the
> wrong error code.  IT should just be returning SQL_NEED_DATA or "done"

The text "Invalid Endian" doesn't appear anywhere in the psqlODBC source
code, so that must be coming from somewhere else. Could be because of
something that psqlODBC did, but it's difficult to trace without more
details. Could you write a self-contained little C program to reproduce
this, please?

- Heikki