diff options
author | Bryan Newbold <bnewbold@archive.org> | 2020-10-21 12:20:52 -0700 |
---|---|---|
committer | Bryan Newbold <bnewbold@archive.org> | 2020-10-21 12:20:54 -0700 |
commit | 200bf734bd459dd3c7a147b3dfe127dbf0ed7f70 (patch) | |
tree | 4f010e66a059271ac3b9c496d15a3bc90bd763c4 /python/sandcrawler/ingest.py | |
parent | 33249f2679851afb64142c428be45d16f35f5539 (diff) | |
download | sandcrawler-200bf734bd459dd3c7a147b3dfe127dbf0ed7f70.tar.gz sandcrawler-200bf734bd459dd3c7a147b3dfe127dbf0ed7f70.zip |
differential wayback-error from wayback-content-error
The motivation here is to distinguish errors due to current content in
wayback (eg, in WARCs) from operational errors (eg, wayback machine is
down, or network failures/disruption).
Diffstat (limited to 'python/sandcrawler/ingest.py')
-rw-r--r-- | python/sandcrawler/ingest.py | 6 |
1 files changed, 5 insertions, 1 deletions
diff --git a/python/sandcrawler/ingest.py b/python/sandcrawler/ingest.py index 322859a..57988e8 100644 --- a/python/sandcrawler/ingest.py +++ b/python/sandcrawler/ingest.py @@ -8,7 +8,7 @@ import requests from http.server import BaseHTTPRequestHandler, HTTPServer from collections import namedtuple -from sandcrawler.ia import SavePageNowClient, CdxApiClient, WaybackClient, WaybackError, SavePageNowError, CdxApiError, PetaboxError, cdx_to_dict, ResourceResult +from sandcrawler.ia import SavePageNowClient, CdxApiClient, WaybackClient, WaybackError, WaybackError, SavePageNowError, CdxApiError, PetaboxError, cdx_to_dict, ResourceResult from sandcrawler.grobid import GrobidClient from sandcrawler.pdfextract import process_pdf, PdfExtractResult from sandcrawler.misc import gen_file_metadata, clean_url @@ -388,6 +388,10 @@ class IngestFileWorker(SandcrawlerWorker): result['status'] = 'wayback-error' result['error_message'] = str(e)[:1600] return result + except WaybackContentError as e: + result['status'] = 'wayback-content-error' + result['error_message'] = str(e)[:1600] + return result except NotImplementedError as e: result['status'] = 'not-implemented' result['error_message'] = str(e)[:1600] |