Hi,
I am trying to insert clobs into Oracle 9i with Oracle 10g JDBC drivers.
The problems I'm facing are:
1) Tried using iBatis ClobTypeHandlerCallback, but encountered error
of unable to setString with > 32K
2) Override ClobTypeHandlerCallback's setParameter to remove the last line
//setter.setString((String)parameter);
so that I can bypass the error. It works if I am inserting 1 clob
only. If I'm inserting 2 clobs,
e.g. (key, clob1, clob2), it will work fine if the clobs are small. If
i'm inserting 2 very large clobs, the order of the clobs in the
database is reversed, i.e. it becomes (key,clob2,clob1). My fellow
developers and I have totally no idea why this actually happened.
So what I did was to do an insert for the first clob, then follow by
an update on the 2nd clob.
3) Once in a while, I am getting "io exception: software caused
connection abort: socket write error" or "No more data to read" when I
do an insertion of clob. This really kills my App. Does anyone has the
solution to this?
4) I thought the above problem may be caused by the removal of
setter.setString((String)parameter), hence I tried the alternative of
using Spring's ClobStringTypeHandler instead. However, I got a
ClassCastException when it is parsing the SqlMap's Parameter.
Does anyone has solution for inserting clobs which are > 32K and will
not encounter the socket write error?
Is anyone using the oracle.sql.CLOB solution? Ever hit the socket
write error or no more data to read error?
I will appreciate some help here. Thanks!
Regards,
Ronn
>Can you use the Oracle 10g driver?
>Cheers,
>Clinton
On Wed, 9 Feb 2005 17:42:45 -0000, Peter Nunn <[EMAIL PROTECTED]> wrote:
> I've been trying (for a while now) to implement a TypeHandlerCallback that
> can WRITE clobs under Oracle 9i. It appears that this isn't possible. If
> anyone has a solution to this then I'd be grateful for your assistance. Up to
> now I haven't seen a workable solution.
>
> So far I have tried the following:
>
> 1) Call setter.setString(s).
>
> This fails because setString(x) has an upper limit of 4K.
>
> 2) Call setter.setCharacterStream(reader, s.length()). i.e. use same approach
> as 2.0.9 ClobTypeHandlerCallback implementation.
>
> This doesn't work either.
>
> 3) Attempt to work with Oracle extension, namely oracle.sql.CLOB. The
> preferred solution.
>
> This isn't possible because you cannot create an instance of oracle.sql.CLOB
> directly. You need to either:
>
> a) create a temporary CLOB via a call to CLOB.createTemporary(...). This
> requires a Connection object. Since ParameterSetter doesn't expose the
> PreparedStatement we cannot get hold of the Connection object. So this
> approach is a non-starter. Even if we did have the connection things aren't
> straightforward because we need an OracleConnection... which you won't have
> if you are using a DataSource to obtain your connections.
>
> b) execute a SELECT statement to obtain a ResultSet then obtain the CLOB
> instance by calling ResultSet.getClob("field"). This clearly isn't an option
> from within a TypeHandlerCallback.
>
> I'd really appreciate some help here... I've hit a brick wall. As of now I've
> come to the conclusion that iBatis cannot handle CLOBS in Oracle prior to
> 10g. This is clearly an issue.
>
> So the only remaining option is to write a JDBC Dao implementation and bypass
> SQLMAPS...any improvements upon this?
>
> Peter Nunn
> Senior Java Developer
> EMAP UK IT
> Telephone: 020 7017 3601
> Mobile: 07866 670 530
>
> ** For Emap magazine subscriptions & gift offers visit
> http://www.greatmagazines.co.uk/emap **
>
> --------------------------------------------------------------------------------------------------------------
> The information in this email is intended only for the addressee(s) named
> above. Access to this email by anyone else is unauthorised.
>
> If you are not the intended recipient of this message any disclosure,
> copying, distribution or any action taken in reliance on it is prohibited and
> may be unlawful.
>
> Emap plc and or its subsidiaries do not warrant that any attachments are free
> from viruses or other defects and accept no liability for any losses
> resulting from infected email transmissions.
>
> Please note that any views expressed in this email may be those of the
> originator and do not necessarily reflect those of this organisation.
> --------------------------------------------------------------------------------------------------------------
>
>
