diff options
Diffstat (limited to 'sql/dump_reingest_spn.sql')
-rw-r--r-- | sql/dump_reingest_spn.sql | 25 |
1 files changed, 18 insertions, 7 deletions
diff --git a/sql/dump_reingest_spn.sql b/sql/dump_reingest_spn.sql index 6ef08c1..a83125c 100644 --- a/sql/dump_reingest_spn.sql +++ b/sql/dump_reingest_spn.sql @@ -1,19 +1,28 @@ +BEGIN TRANSACTION ISOLATION LEVEL SERIALIZABLE READ ONLY DEFERRABLE; + COPY ( SELECT row_to_json(ingest_request.*) FROM ingest_request - LEFT JOIN ingest_file_result ON ingest_file_result.base_url = ingest_request.base_url - WHERE ingest_request.ingest_type = 'pdf' + LEFT JOIN ingest_file_result ON + ingest_file_result.base_url = ingest_request.base_url + AND ingest_file_result.ingest_type = ingest_request.ingest_type + WHERE + (ingest_request.ingest_type = 'pdf' + OR ingest_request.ingest_type = 'html' + OR ingest_request.ingest_type = 'xml' + OR ingest_request.ingest_type = 'component') AND ingest_file_result.hit = false AND ingest_request.created < NOW() - '6 hour'::INTERVAL AND ingest_request.created > NOW() - '180 day'::INTERVAL AND ingest_request.ingest_request_source = 'savepapernow-web' AND ( ingest_file_result.status like 'spn2-%' - -- OR ingest_file_result.status like 'cdx-error' - -- OR ingest_file_result.status like 'wayback-error' - -- OR ingest_file_result.status like 'wayback-content-error' - OR ingest_file_result.status like 'petabox-error' - -- OR ingest_file_result.status like 'gateway-timeout' + -- OR ingest_file_result.status = 'cdx-error' + -- OR ingest_file_result.status = 'wayback-error' + -- OR ingest_file_result.status = 'wayback-content-error' + OR ingest_file_result.status = 'petabox-error' + -- OR ingest_file_result.status = 'gateway-timeout' + OR ingest_file_result.status = 'no-capture' ) AND ingest_file_result.status != 'spn2-error:invalid-url-syntax' AND ingest_file_result.status != 'spn2-error:filesize-limit' @@ -23,3 +32,5 @@ COPY ( AND ingest_file_result.status != 'spn2-error:network-authentication-required' AND ingest_file_result.status != 'spn2-error:unknown' ) TO '/srv/sandcrawler/tasks/reingest_spn.rows.json'; + +ROLLBACK; |