FreeTDS Development Group

[Bob Kline <bkline AT rksystems.com>]

On Sun, 4 Jun 2000, Carsten Rhod Gregersen wrote:

> Hi Bob,
> 
> Just tell me if you want any help.
> 
> Some weeks ago I've posted a patch that makes it posible to
> handle TEXT columns. My guess is the NTEXT and TEXT columns
> are handled the same way.
> 
> We haven't tested the patch alot, it seems to work and we havn't had
> any problems.
> 
> I send the sources to Justin Schwartz (justin AT ivorytower.co.za) who
> had the same problem as I had, and therefore was requesting a
> solution. Maybe he can say something about the patch.
> 
> About the unicode problem. Would it not be enough just to do
> a "mystring.getBytes()" ???
> 

Carsten:

Thanks very much for the tip.

In a separate branch of this thread I posted a set of my own patches
which implement the ability to select NTEXT columns and get back the
data either as wide-character strings or as narrow-character strings
with the wide characters stripped back to ASCII.  The second flavor is
too crude, of course.  It should really offer UTF-8 or conversion to the
local encoding (reporting errors for characters which can't be
converted), but since (a) I'm not sure how much interest there will be
in these patches and, more to the point (b) my own purpose in dabbling
with the C code was to figure out how to get it working in the JDBC
driver, where most of these problems won't exist (Java supports Unicode
out of the box).

Thanks again!

-- 
Bob Kline
mailto:bkline AT rksystems.com
http://www.rksystems.com