Hi John - thank you for your quick reply.  Awesome! - your suggested 
solution is similar to what I did to get things to work for us: 
https://gist.github.com/3753803
But you are right, the 'create_table' code is still broken.  Since we 
wanted more control over the table creation process in Rails, I created a 
migration (https://gist.github.com/3753821) instead of using your 
'create_table' method, so that will still need to be addressed.

Thank you for your help!
Geoff


On Wednesday, September 19, 2012 4:41:31 PM UTC-4, John Mettraux wrote:
>
>
> On Wed, Sep 19, 2012 at 10:32:12AM -0700, gmh wrote: 
> > 
> > First off - thank you for such a cool piece of software.  The ruote 
> > workflow has such a rich toolset - it will save us tons of time no 
> doubt. 
> > 
> > (...) 
> > 
> > If there is a possibility for the data in the doc column to be larger 
> than 
> > the 4000 limit, then the CLOB type will need to be supported.  We 
> > experimented by calling the '.read' method of the passed back 
> <OCI8::CLOB> 
> > object in storage.rb (extracting the text from the column).  This seems 
> to 
> > work, but we are wondering if anyone else has already done this?  And is 
> > this really necessary if a simple VARCHAR2 type will be large enough? 
>
> Hello Geoff, 
>
> welcome to the ruote mailing list! 
>
> Yes, VARCHAR2 might be limited. Ruote-sequel naively encodes as JSON the 
> documents and places them in the "doc" column, those 4000 bytes feel not 
> that 
> far. 
>
> Maybe I could adapt ruote-sequel so that it calls #read before doing the 
> JSON 
> decode on the doc column (if the resulting object responds to #read). 
> That'd 
> make ruote-sequel adapted to your situation out of the box (though I 
> wonder 
> how one tells Sequel to use CLOB instead of VARCHAR2 when :text => 
> true...) 
>
> What do you think ? 
>
> -- 
> John Mettraux - http://lambda.io/jmettraux 
>
>

-- 
you received this message because you are subscribed to the "ruote users" group.
to post : send email to [email protected]
to unsubscribe : send email to [email protected]
more options : http://groups.google.com/group/openwferu-users?hl=en

Reply via email to