[fpc-devel] Oracle CLOB and UTF8

[LacaK]

Hi,

as far as I remember I have problems (when I have implemented support 
for CLOBs) you describe as well.

I do not have access to Oracle DB anymore so I can not fix it, worse I 
am not aware of easy way how to get from OCI API information about byte 
length (and connection character set)  ... I have never used Oracle DB 
except for one period when I have fixed some errors found in our db test 
suite.

There are some API functions like OCILobCharSetId(), which probably can 
be used to guess "character length" based on known character set ids.

If you are aware of working solution, please submit bug report and 
attach patch ...
Thanks
-Laco.


> Hi all,
>
> I've got a database with a CLOB column containing "abcdefgáéíóúøå"
>
> I'm using the TOracleConnection to select this data and it comes back 
> corrupted "abcdefgáéí�"
>
> This only happens when using a CLOB. If I put the same data in a 
> VARCHAR2 or NVARCHAR2 the results are correct.
>
> The OCILobGetLength in the procedure 
> TOracleConnection.LoadBlobIntoBuffer returns 14. This is the number of 
> characters in the string, but it is not taking the double-byte 
> characters into account. The length in the case should be 21.
>
> Is this a known issue?
>
> Many thanks,
> Dave
>
>
>
> _______________________________________________
> fpc-devel maillist  -  fpc-devel at lists.freepascal.org
> http://lists.freepascal.org/cgi-bin/mailman/listinfo/fpc-devel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.freepascal.org/pipermail/fpc-devel/attachments/20170906/496c6b4e/attachment.html>