translating many-to-many relationship from CSV to database?

[James Hartley]

I have a CSV file with lots of redundant data which models many-to-many relationships.  I'm needing to scrub the data as it is inserted into the database littered with unique constraints.  Is there a way to insert the data once without querying for each object before inserting?

I'm sure this is a common CSV translation nit, but it is new to me.

Thanks for any insight provided!

[Michael Bayer]

typically I load all the records I’m going to be working with at once, and assemble them into dictionaries or other data structures so that as I process the csv data, I can look things up in the local dictionary for the modify.

if the data you get from your CSV can all relate to the primary key of the objects, then you can use session.merge(), applying the existing primary key to a new object, which will be merged into the state of the existing row.   This also emits a SELECT for those rows, but again if you load them up ahead of time, the check is against the local identity map.

--
You received this message because you are subscribed to the Google Groups "sqlalchemy" group.
To unsubscribe from this group and stop receiving emails from it, send an email to sqlalchemy+...@googlegroups.com.
To post to this group, send email to sqlal...@googlegroups.com.
Visit this group at http://groups.google.com/group/sqlalchemy.
For more options, visit https://groups.google.com/d/optout.