From 4e8407758618bece136addffe301ba8357366de3 Mon Sep 17 00:00:00 2001 From: Bryan Newbold Date: Fri, 7 Jan 2022 18:02:09 -0800 Subject: enqueue PLATFORM PDFs for crawl --- notes/tasks/2022-01-07_grobid_platform_pdfs.md | 23 +++++++++++++++++++++++ 1 file changed, 23 insertions(+) create mode 100644 notes/tasks/2022-01-07_grobid_platform_pdfs.md (limited to 'notes/tasks') diff --git a/notes/tasks/2022-01-07_grobid_platform_pdfs.md b/notes/tasks/2022-01-07_grobid_platform_pdfs.md new file mode 100644 index 0000000..b5422c2 --- /dev/null +++ b/notes/tasks/2022-01-07_grobid_platform_pdfs.md @@ -0,0 +1,23 @@ + +Martin crawled more than 10 million new PDFs from various platform domains. We +should get these processed and included in sandcrawler-db. + +## Select CDX Rows + + COPY ( + SELECT DISTINCT ON (cdx.sha1hex) row_to_json(cdx) + FROM cdx + LEFT JOIN grobid ON grobid.sha1hex = cdx.sha1hex + WHERE + grobid.sha1hex IS NULL + AND cdx.sha1hex IS NOT NULL + AND cdx.warc_path LIKE 'PLATFORM-CRAWL-2020%' + -- LIMIT 5; + ) + TO '/srv/sandcrawler/tasks/ungrobided_platform_crawl.2022-01-07.cdx.json' + WITH NULL ''; + => COPY 8801527 + + cat /srv/sandcrawler/tasks/ungrobided_platform_crawl.2022-01-07.cdx.json | rg -v "\\\\" | jq . -c | pv -l | kafkacat -P -b wbgrp-svc263.us.archive.org -t sandcrawler-prod.ungrobided-pg -p -1 + + # for pdfextract, would be: sandcrawler-prod.unextracted -- cgit v1.2.3