From eac269a36d648db8712eb4fda4c0a896d5f4dcbc Mon Sep 17 00:00:00 2001 From: Bryan Newbold Date: Thu, 6 Feb 2020 11:58:13 -0800 Subject: ingest worker: handle missing ingest_request_source Seeing a bunch of these due to re-ingests not including this field because of an earlier persist bug. --- python/fatcat_tools/importers/ingest.py | 3 +++ 1 file changed, 3 insertions(+) (limited to 'python/fatcat_tools') diff --git a/python/fatcat_tools/importers/ingest.py b/python/fatcat_tools/importers/ingest.py index bdfd2835..17dafe91 100644 --- a/python/fatcat_tools/importers/ingest.py +++ b/python/fatcat_tools/importers/ingest.py @@ -54,6 +54,9 @@ class IngestFileResultImporter(EntityImporter): self.counts['skip-hit'] += 1 return False source = row['request'].get('ingest_request_source') + if not source: + self.counts['skip-ingest_request_source'] += 1 + return False if self.ingest_request_source_whitelist and source not in self.ingest_request_source_whitelist: self.counts['skip-ingest_request_source'] += 1 return False -- cgit v1.2.3