That worked (of course) would never of gotten that (still wokring on what
** dash squiggle ~!! * means in C).

The neat thing is there is very little code to ad as I just reuse the
fetch_lob code that is already part of DBD oracle.

Turns out it will be a 4 or 5 line patch.

cheers
> Thanks Paul was suspecting that myself but as usual in OCI is is one
> little reff to it buried in the docs.
>
> Iwill give it a try
>> On Wed, 12 Mar 2008 10:22:11 -0400, <[EMAIL PROTECTED]> wrote:
>>
>>> Hi Mark been playing with the code you sent me for the past day or two
>>> and
>>> unfortunetly I have have made little progress with it.
>>>
>>> Seems I get the value out of the Database no problem of the type
>>> OCI_TYPECODE_CLOB,  But I cannot tell if it is a lob_locator or not and
>>> there seems no way to bind a decriptor to it.
>>>
>>
>> It is a pointer to a lob locator.   You need to do something like this:
>>
>> OCILobLocator *lobloc = *(OCILobLocator**)valuep;
>>
>> and then use OCILobGetLength and OCILobRead on the locator to get the
>> data
>> out.
>>
>> -P
>>
>>
>>
>>
>>
>>> So I am not sure I will be able to build in support for embedded clobs
>>> in
>>> the next version of DBD::Oracle
>>>
>>> I will work on it a little more today as there is still one thing I
>>> have
>>> not tried but I do not see much of a future in it.
>>>
>>> cheers
>>> John Scoles
>>>>
>>>>
>>>> dear sir
>>>>
>>>> i attach a tarball containing a sqlplus script, mt.sql which creates a
>>>> table and type, both of which contain an embedded clob field.
>>>> the file mt.lst is the sqlplus spooled output of mt.sql, and you can
>>>> see
>>>> that the data is in the clob, and is accessible in the type.
>>>>
>>>> the other two files are mt.pl and output mt.txt.
>>>> the clob field doesn't appear in the embedded version of the query.
>>>>
>>>> i think this is a problem with DBD::Oracle.
>>>>
>>>> following your instructions, i am sending this to yourselves first.
>>>>
>>>> if i need to work on a fix, can u pls advise whereabouts i should
>>>> look.
>>>>
>>>> thanks for your time and for any feedback you can provide.
>>>>
>>>> mr
>>>>
>>>
>>>
>>
>>
>>
>
>
>


Reply via email to