[Zope3-dev] Queries (was Re: personal conceptual questions)

[Shane Hathaway]

Jeffrey P Shell wrote:
> Whee!  Happy to pop back in for a moment.  I'll try to find time to 
> follow up on all the relations discussions because I still find myself 
> caring about that a lot.  Queries are the other big thing.
> 
> I really like the concept of APE, but found that I couldn't use it with 
> my existing database application (instead, I'm ending up with something 
> that's halfway between APE and Modeling, using a lot of Table Gateway 
> pattern based objects).  The main reason that I couldn't go with APE was 
> that I couldn't find an easy way to pull objects out of a relational 
> storage (since APE is so transparent) according to a set of criteria, 
> such as "Get me all Shows between May 5 2003 and June 6 2003".  If I had 
> written the application for Zope, I would have had one or more catalogs 
> that I could query with the catalog query language.  But since it's in 
> an RDBMS, I already have this all set up with SQL.

Here's one approach you might take with Ape.  Write a SQL query that 
returns primary keys, then fetch objects by key through an Ape 
connection.  It might look something like this:

cursor.execute("select key from shows where "
                "date >= '2003-05-05' and date <= '2003-06-06'")
rows = cursor.fetchall()
objs = [jar.loadStub((row[0],)) for row in rows]

There are good things about this (caching, for instance) and bad things 
about this (caching, for instance.) :-)

> What would be nice is to be able to have some sort of query language 
> that could go down closer to the storage level and not the application 
> level (where catalog is).  This would make working with and querying 
> other storage types easier.

I've been discussing relationships for exactly this purpose.  I want 
relationship (and relational) capabilities in ZODB so that I can mimic 
them in Ape.  I'd like to hide all SQL from the application.  But I just 
read an article by Fowler that convinced me that hiding SQL is not 
always the best thing to do.  So now I'm considering a direct-query 
interface again.  (I wonder if I got the URL from you? ;-) )

http://www.martinfowler.com/articles/dblogic.html

> I'd be happy with something like this that could query ZODB, and could 
> query other storages (via APE or something else) in the same manner.  
> Not sure if it can happen, but it's another little dream of mine.  
> Another dream I have no time to do anything about.

+1 ;-)

Thanks for the information about Modeling, though.  I still wonder if 
mixing Modeling and Ape might work.

Shane