From ed27bf6fb7afda158812c0822498bc7408572b8e Mon Sep 17 00:00:00 2001 From: Bryan Newbold Date: Wed, 15 Apr 2020 12:42:42 -0700 Subject: COVID-19 chinese paper ingest --- notes/ingest/2020-04-13_covid19.md | 73 ++++++++++++++++++++++++++++++++++++++ 1 file changed, 73 insertions(+) create mode 100644 notes/ingest/2020-04-13_covid19.md (limited to 'notes') diff --git a/notes/ingest/2020-04-13_covid19.md b/notes/ingest/2020-04-13_covid19.md new file mode 100644 index 0000000..b442d69 --- /dev/null +++ b/notes/ingest/2020-04-13_covid19.md @@ -0,0 +1,73 @@ + +Want to ensure seedlists from Wanfang and CNKI are captured in wayback. + +Wanfang URLs seem normal. Let's just submit them in a single queue via SPNv2. +They are heterogenous after redirect. + +CNKI are trickier. The PDF URLs definitely can't be crawled directly... but the +info ones probably can, then crawl on to PDF? At least some seem to capture Ok. + +Need scope and identifiers for ingest requests. Let's do: + + cnki_covid19 / + wanfang_covid19 / + +Source: scrape-covid19 + +## Commands + + # in sandcrawler pipenv + cat ~/code/covid19.fatcat.wiki/extra/scrape/cnki_metadata.2020-04-14.json | ./scripts/covid2ingestrequest.py - > ~/code/covid19.fatcat.wiki/extra/scrape/cnki_ingest_request.2020-04-14.json + cat ~/code/covid19.fatcat.wiki/extra/scrape/wanfang*.2020-04-14.json | ./scripts/covid2ingestrequest.py - > ~/code/covid19.fatcat.wiki/extra/scrape/wanfang_ingest_request.2020-04-14.json + + + cat /tmp/wanfang_ingest_request.2020-04-14.json | shuf | jq . -c | kafkacat -P -b wbgrp-svc263.us.archive.org -t sandcrawler-prod.ingest-file-requests -p 4 + cat /tmp/cnki_ingest_request.2020-04-14.json | shuf | jq . -c | kafkacat -P -b wbgrp-svc263.us.archive.org -t sandcrawler-prod.ingest-file-requests -p 8 + +## Status + + SELECT ingest_request.ingest_type, + ingest_file_result.status, + COUNT(*) + FROM ingest_file_result + LEFT JOIN ingest_request + ON ingest_file_result.ingest_type = ingest_request.ingest_type + AND ingest_file_result.base_url = ingest_request.base_url + WHERE + ingest_request.ingest_type = 'pdf' + AND ingest_request.ingest_request_source = 'scrape-covid19' + GROUP BY ingest_request.ingest_type, ingest_file_result.status + ORDER BY COUNT(*) DESC; + +2020-04-15: + + ingest_type | status | count + -------------+-------------------------------------+------- + pdf | spn2-cdx-lookup-failure | 1588 + pdf | success | 671 + pdf | gateway-timeout | 507 + pdf | no-pdf-link | 181 + pdf | wayback-error | 30 + pdf | spn2-error:job-failed | 20 + pdf | spn2-error | 7 + pdf | spn2-error:soft-time-limit-exceeded | 3 + pdf | spn2-error:pending | 2 + (9 rows) + +## Re-Try + + 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' + AND ingest_request.ingest_request_source = 'scrape-covid19' + AND ingest_file_result.ingest_type = 'pdf' + AND ingest_file_result.hit = false + AND ingest_file_result.status != 'no-pdf-link' + AND ingest_file_result.status != 'link-loop' + ) TO '/grande/snapshots/reingest_covid19.rows.json'; + + ./scripts/ingestrequest_row2json.py /grande/snapshots/reingest_covid19.rows.json | shuf > reingest_covid19.json + + cat reingest_covid19.json | shuf | jq . -c | kafkacat -P -b wbgrp-svc263.us.archive.org -t sandcrawler-prod.ingest-file-requests -p 9 + -- cgit v1.2.3