Is filtering supposed to work for querying Annotations?

[fedorov]

I am confused about the behavior of the query below. Should it filter by patient ID, or not?

[notmatthancock]

Scans and Annotation objects are stored in different tables, and so to do this type of query, you need a join clause. See here.

[notmatthancock]

Also, to get counts from a query, you can just issue a count on the query, e.g.:

print pl.query(pl.Annotation).join(pl.Scan)\
        .filter(pl.Scan.patient_id == 'LIDC-IDRI-0011').count()
# 23

Alternatively, you can accomplish this same result by:

scan = pl.query(pl.Scan).filter(pl.Scan.patient_id == 'LIDC-IDRI-0011').first()
print len(scan.annotations)
# 23

[fedorov]

Thank you! Sorry for the confusion.

[notmatthancock]

No problem. FYI, pylidc is implemented with sqlalchemy, which is where the filter, count, join, etc ... functions come from. sqlalchemy has nice documentation if you want to dive deeper into more complex query capabilities.