Hello, all.

I'm trying to create a view that uses CASE statements in the column 
expression to categorize, so that all the logic is in setting up the view. 
Then nothing else needs to know all the rules, and if / when the rules 
change, I can simply drop and re-create the view without changing any other 
queries --- in SA or outside it.

EG

CREATE TABLE people ( name VARCHAR(10), income INTEGER );
INSERT INTO people (name, income) VALUES ( 'bob', 123456);
INSERT INTO people (name, income) VALUES ( 'barry', 35467);
INSERT INTO people (name, income) VALUES ( 'betty', 95672);
CREATE VIEW people_income_class AS SELECT name, income, CASE WHEN income < 
50054 THEN 'below median' ELSE 'above median' END AS income_class FROM 
people;

In this case I'd like to have the people_income_class view come from a SA 
select, with the median income, 50054, coming from a bound parameter.

The recipe I've found for view creation, 
http://stackoverflow.com/questions/9766940/how-to-create-an-sql-view-with-sqlalchemy,
 
is great at defining views with a bound parameter in the WHERE clause, but 
literal_binds=TRUE isn't picking up the parameter in the CASE statement. 
Before I do much more tracing of the paths through the compiler, I thought 
I'd ask.

Anyone have suggestions?


Richard


-- 
You received this message because you are subscribed to the Google Groups 
"sqlalchemy" group.
To view this discussion on the web visit 
https://groups.google.com/d/msg/sqlalchemy/-/WI-FkyMymsgJ.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to 
[email protected].
For more options, visit this group at 
http://groups.google.com/group/sqlalchemy?hl=en.

Reply via email to