diff options
author | Bryan Newbold <bnewbold@archive.org> | 2022-01-11 14:49:37 -0800 |
---|---|---|
committer | Bryan Newbold <bnewbold@archive.org> | 2022-01-11 14:49:37 -0800 |
commit | ff6894043576d3d51c9ab16623053f91780edc89 (patch) | |
tree | 983ed2f5a61c05c088df96416017ef13cf439dda | |
parent | 4e8407758618bece136addffe301ba8357366de3 (diff) | |
download | sandcrawler-ff6894043576d3d51c9ab16623053f91780edc89.tar.gz sandcrawler-ff6894043576d3d51c9ab16623053f91780edc89.zip |
spn: handle blocked-url (etc) better
-rw-r--r-- | python/sandcrawler/ia.py | 10 |
1 files changed, 10 insertions, 0 deletions
diff --git a/python/sandcrawler/ia.py b/python/sandcrawler/ia.py index dc9aae5..d123c9f 100644 --- a/python/sandcrawler/ia.py +++ b/python/sandcrawler/ia.py @@ -1040,6 +1040,16 @@ class SavePageNowClient: and "You have already reached the limit of active sessions" in resp_json["message"] ): raise SavePageNowBackoffError(resp_json["message"]) + elif resp_json.get("status") == "error": + return SavePageNowResult( + False, + resp_json.get("status_ext") or resp_json["status"], + None, + request_url, + None, + None, + None, + ) elif not resp_json or "job_id" not in resp_json or not resp_json["job_id"]: raise SavePageNowError( "Didn't get expected 'job_id' field in SPN2 response: {}".format(resp_json) |