Sorry if this is a duplicate...I never saw it come through...
--- Begin Message ---
Freaky. If you can send me both the source + stylesheets and the result
pre-serializer then I can put some logging statements in and try and
root out the problem. This sounds like a Cocoon problem, it sounds like
something isn't being sent that HSSF is trying to validate.
This basically confirms to me that its not at least a direct serializer
problem as I can't imagine how a saved file would work yet a direct
sent. I'm posting this to the list as well since someone may know of this.
To recap:
Seth has an XMl source of some sort that he applies multiple layers of
transformation/stylesheets to.
He's getting a null pointer thrown from the serializer
If he limits the number of these transformations he does not get the
null pointer
If he saves the result with all transformations minus the serializer
then reruns the serialized results through the serializer, no null
pointer is thrown and the sheet runs well.
So it appears that through the pipeline an expected hunk of data is NULL
and/or a validation in the serializer is failing (though its peculiar
that its in the element processor parent and not a child, which means
its probably when its idenitying the event aka tag) -- the same data is
NOT null when saved and passed to the serializer directly.
Seth correct me if I've left anything out?
Seth if you can send me both sets of data/transformations I can probably
insert the relevant log statements and trace this down further (at the
very least). Any sensitive data will be treated with all due respect.
-Andy
Seth Ladd wrote:
>On Fri, 2002-04-26 at 19:39, Andrew C. Oliver wrote:
>
>>Okay I can't say for sure but from the looks of this, I'm betting its a
>>validation error. U/F this isn't really enough to go on. I can't
>>install this and run with it. I'm finding it doubtful I could reproduce
>>the error based only on this, I think the data you're passing in is
>>probably the cause of the error. At some point when you have a bit more
>>time to invest in this I'd like to look a bit more in depth into this.
>>
>
>I'm pretty sure you're right. The only thing that makes me think my
>data is good is this: I take the result XML (right before the
>HSSFSerializer) and save it to file. Then I take that (as a file, not
>as sax events) and give it to the HSSF serializer. It works great that
>way. I can even send you the XML file for you to pump through
>HSSFSerializer. The only thing I can think of is that the data is
>different as it travels through as SAX events.
>
>Thanks,
>Seth
>
>
>
--- End Message ---
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, email: [EMAIL PROTECTED]
