Reputation: 183
Connected to an Oracle database, I need to run the following query:
SELECT
ad_campaign_id as campaign_id,
TO_DATE("DAY", 'YYYY-MM-DD') as DAY,
NVL(spent, 0) as spent,
impressions,
clicks
FROM sometable
How can I do that with SQLAlchemy? How I can send paramaters to the func.to_date
function?
Upvotes: 1
Views: 2992
Reputation: 15718
SQLAlchemy's func will accept any Oracle function as an attribute.
from sqlalchemy import func
campaign_id = sometable.c.ad_campaign_id
day = func.to_date(sometable.c.day, 'YYYY-MM-DD') # Is this what you need?
spent = func.nvl(sometable.c.spent, 0)
clicks = sometable.c.clicks
cols = [campaign_id, day, spent, clicks]
# Feel free to skip, this is just so you get the labels you want
labels = "campaign_id day spent clicks".split()
col_label = zip(cols, labels)
labelled = [c.label(l) for (c,l) in col_label]
q = session.query(*labelled) # or use *cols if you don't care about labels
q.all()
Upvotes: 4
Reputation: 16212
Here's how you execute arbitrary queries:
engine.execute(query_string)
setting up the engine is another story. It's pretty well documented though.
If you are asking how to do this using some model that's a different story again...
So that's the first part of your question answered (possibly). But what do you mean by the func.to_date function? Please clarify
Upvotes: 0