Tracking records that could not be processed

NotCompleted is a special result type provided by the scinexus package. Apps return a NotCompleted result when a condition is not met (FALSE type) or an unexpected error occurs (ERROR type). These objects evaluate to False, carry information about the failure, and propagate through composed pipelines so that subsequent steps are skipped.

A cogent3 example

Below, an app that selects specific sequences returns a NotCompleted because the requested sequence "Mouse" does not exist in the primate data set.

from cogent3 import get_app

reader = get_app("load_aligned", format_name="fasta")
select_seqs = get_app("take_named_seqs", "Mouse", "Human")
aln = reader("data/primate_brca1.fasta")
result = select_seqs(aln)
assert result == False
result

NotCompleted(type=FAIL, origin=take_named_seqs, source="primate_brca1", message="named seq(s) {'Mouse'} not in ('Galago', 'HowlerMon', 'Rhesus', 'Orangutan', 'Gorilla', 'Human', 'Chimpanzee')")

The NotCompleted instance has attributes identifying what data failed,

result.source

'primate_brca1'

where the failure occurred

result.origin

'take_named_seqs'

and the reason for the failure

result.message

"named seq(s) {'Mouse'} not in ('Galago', 'HowlerMon', 'Rhesus', 'Orangutan', 'Gorilla', 'Human', 'Chimpanzee')"