diff options
| -rw-r--r-- | python/sandcrawler/ingest.py | 3 |
1 files changed, 3 insertions, 0 deletions
diff --git a/python/sandcrawler/ingest.py b/python/sandcrawler/ingest.py index 38ba55c..94eb9a5 100644 --- a/python/sandcrawler/ingest.py +++ b/python/sandcrawler/ingest.py @@ -273,6 +273,9 @@ class IngestFileWorker(SandcrawlerWorker): result['status'] = resource.status return result + if not resource.body: + result['status'] = 'null-body' + return result file_meta = gen_file_metadata(resource.body) if "html" in file_meta['mimetype'] or "xml" in file_meta['mimetype']: |