Hi Patrick,
versions are as follows:
OpenSUSE 10.2
Virtuoso 5.0
pyodbc 2.0.35 with your patch
iODBC 3.52.5 with your patch
Python 2.5
Also, I had to patch pyodbc/cursor.cpp line 575 and
611 changing SQLINTEGER to SQLLEN to build it with
iODBC.
Regards,
Anton
--- Patrick van Kleef wrote:
HI Anton,
thanks for your fixes, especially for doing extra work
finding the problem in pyodbc. BTW are you going to
contact pyodbc's author to submit this patch?
Yes, once i have done a little more testing with it, i will send it
to the pyodbc authors.
I've applied the patches, and I c
Hi Patrick!
thanks for your fixes, especially for doing extra work
finding the problem in pyodbc. BTW are you going to
contact pyodbc's author to submit this patch?
I've applied the patches, and I can run my non-unicode
queries with virtodbc32u_r.so fine now.
However, running the pyodbctests.py
Hi Anton,
Thanks for the ODBC trace which I have passed on to development of analysis.
My observation shows the following call sequences being the source of the
failure when using the Unicode driver which works with the ANSI driver:
python B7DB26C0 ENTER SQLExecDirect
SQL
Hi Hugh,
please find trace files attached. First one is for
pyodbc used with virtodbc32_r.so, second is for
pyodbc with virtodbc32u_r.so.
The mxODBC driver seems to work okay with
virtodbc32u_r.so, but it has problems working under
python 2.5, so I'd like to see if pyodbc can be used
with unicode
Hi Anton,
Can you please provide an iODBC trace of your attempts to insert unicode
data with the Virtuoso unicode driver (virtodbc32u_r.so) with the mxODBC and
pyodbc drivers. The iODBC tracing can be enabled by adding the following
entry to your odbc.ini file:
[ODBC]
Trace=1
TraceFile=
This wil
Hi Hugh,
I tried to do some tests, and found that pyodbc when
used with virtodbc32u_r.so fails to execute any
queries with error:
"[HY001] [iODBC][Driver Manager]Memory allocation
error (-1211581185)"
though it works for non-unicode queries with
virtodbc32_r.so.
The mxODBC seem to work with v
Hi Anton,
Can you please provide details on how you are attempting to save these
unicode strings as RDF literals ?
>From on OpenLink point of view both iODBC and the Virtuoso ODBC driver
support unicode, although note for the Virtuoso ODBC Driver you need to
ensure you are using the virtodbc32u_r
Hi,
yesterday I tried to do few simple tests saving python
unicode string as RDF literals in Virtuoso, but with
no success. So I wonder if there is any existing guide
on
setting system up for Unicode support?
What db driver should I use mxODBC or pyodbc?
What ODBC manager unixODBC or iODBC?
What
