Re: :Oracle - Any advance on inserting CLOB to XMLTYPE?

Sorry, after starting this thread, I dropped out for a while folks.
John, did you get to a satisfactory solution in the end?
Paul

---------- Forwarded message ----------
From: "John Scoles"
To: dbi-users@perl.org
Date: Mon, 9 Apr 2007 14:33:41 -0400
Subject: Re: :Oracle - Any advance on inserting CLOB to XMLTYPE?
Well I have been hacking this one to death this weekend. Did manage to get
it to work using a SP that creats a lob pointer before we do the update but
this is a very bad hack on my part. I think this will require lob locators
as the function SYS.XMLType.CREATEXML requires all the CLOB there before it
runs, So we might have to

1) prepare the SQL
2) upload the CLOB
3) bind
4) execute the statment.

I do not like this myself.

I will look at the Lob locators and see if I can work it out with that later
this week.

cheers All

""John Scoles"" wrote in message
news:20070406013714.19853.qmail@lists.develooper.com...
> Sorry to be a Johnny come lately on this (pun intended)
>
> I have been playing with this myself and like you chaps have come up with
> a blank. End up with ORA-00942 error
>
> I did spend some time to see how the python guys did it but as I can see
> they just did it as a cursor which I think we can do as well in
> DBD::Oracle. It is too late at night for me to give it a try though.
>
> I have stepped though all the code and can see nothing awry.
>
> I think our only alternative is to build up the clob in hunks then try to
> run it using the LOB Locator Methods We could indicate this with a new
> ORA type called ORA_XMLTYPE or alike????
>
> Cheers
>
> Of by the way any of you going the
> http://vienna.yapceurope.org/ye2007/index.html YAPC in Viena in Aug.
>
>

Re: :Oracle - Any advance on inserting CLOB to XMLTYPE?

So still trying to track this one down. Tim B. Suggestes we look as some of
his code in his direct Oracle ::OCI.
and we use handles to get the lob out in chunks. On my to do list still
when I get a little free time.


""Paul Gallagher"" wrote in message
news:9b7819b20704202011h4f6abe95ic62b35bf117d1777@mail.gmail .com...
> Sorry, after starting this thread, I dropped out for a while folks.
> John, did you get to a satisfactory solution in the end?
> Paul
>
> ---------- Forwarded message ----------
> From: "John Scoles"
> To: dbi-users@perl.org
> Date: Mon, 9 Apr 2007 14:33:41 -0400
> Subject: Re: :Oracle - Any advance on inserting CLOB to XMLTYPE?
> Well I have been hacking this one to death this weekend. Did manage to
> get
> it to work using a SP that creats a lob pointer before we do the update
> but
> this is a very bad hack on my part. I think this will require lob
> locators
> as the function SYS.XMLType.CREATEXML requires all the CLOB there before
> it
> runs, So we might have to
>
> 1) prepare the SQL
> 2) upload the CLOB
> 3) bind
> 4) execute the statment.
>
> I do not like this myself.
>
> I will look at the Lob locators and see if I can work it out with that
> later
> this week.
>
> cheers All
>
> ""John Scoles"" wrote in message
> news:20070406013714.19853.qmail@lists.develooper.com...
>> Sorry to be a Johnny come lately on this (pun intended)
>>
>> I have been playing with this myself and like you chaps have come up with
>> a blank. End up with ORA-00942 error
>>
>> I did spend some time to see how the python guys did it but as I can see
>> they just did it as a cursor which I think we can do as well in
>> DBD::Oracle. It is too late at night for me to give it a try though.
>>
>> I have stepped though all the code and can see nothing awry.
>>
>> I think our only alternative is to build up the clob in hunks then try to
>> run it using the LOB Locator Methods We could indicate this with a new
>> ORA type called ORA_XMLTYPE or alike????
>>
>> Cheers
>>
>> Of by the way any of you going the
>> http://vienna.yapceurope.org/ye2007/index.html YAPC in Viena in Aug.
>>
>>