On 02/09/2010, at 1:45 AM, David Glick wrote:

> On 9/1/10 12:00 AM, Dylan Jay wrote:
>> We recently had to create a PFG action adapter that could store large
>> numbers of records in the ZODB and retrieve efficiently using
>> different indexes. We might rewrite it and release it so it could  
>> be a
>> drop in replacement for any PFG form.
> That sounds useful. Did you use collective.soup or something custom  
> for
> storage/indexing?

We just used custom BTree structures for our needs but had we more  
time we would have built it more generic using catalog. I haven't seen  
collective.soup but it seems to be use catalog too which is nice.

>> We might rewrite it and release it so it could be a drop in
>> replacement for any PFG form. If so we'd love it this "SQL PFG action
>> adapter" was SQL agnostic in its api so we and others could use these
>> two adapters interchangeably.
> Is there a more specific interface that needs to be implemented to
> ensure this, beyond the basic PFG action adapter contract (provide an
> onSuccess method that does something with form fields) ?

What I had in mind is a Generic CSVContentType that could be useful  
when you wanted to store a lot tabular data programatically but also  
want to have a simple UI for the users to get at the data.
It would have a simple UI that would let upload and download as csv  
like the PFG save adaptor does. The api we needed was to add, remove  
and insert rows and do queries (like soup). For our use case we really  
only needed looking up by primary keys but the kind of catalog queries  
soup uses gives would be nice. It would implement the onSuccess method  
and be able add columns on the fly so it can be a drop in replacement  
for a pfg. And as a bonus a nicer UI to browse the data and maybe edit  
it would be nice (e.g. atspreadsheet).

>

_______________________________________________
Product-Developers mailing list
[email protected]
http://lists.plone.org/mailman/listinfo/product-developers

Reply via email to