diff options
author | Bryan Newbold <bnewbold@archive.org> | 2020-07-08 12:07:08 -0700 |
---|---|---|
committer | Bryan Newbold <bnewbold@archive.org> | 2020-07-08 12:07:08 -0700 |
commit | ab1bc8c0cd11458be062dca8c5c758b9c28e3259 (patch) | |
tree | a17082b9e774b14f2a6197491ddb12dc85269ea4 | |
parent | 073421d6317fc33d810c5195cbaee15d2a81b30e (diff) | |
download | sandcrawler-ab1bc8c0cd11458be062dca8c5c758b9c28e3259.tar.gz sandcrawler-ab1bc8c0cd11458be062dca8c5c758b9c28e3259.zip |
MAG 2020-07 ingest notes
-rw-r--r-- | notes/ingest/2020-07_mag.md | 159 |
1 files changed, 159 insertions, 0 deletions
diff --git a/notes/ingest/2020-07_mag.md b/notes/ingest/2020-07_mag.md new file mode 100644 index 0000000..c459e56 --- /dev/null +++ b/notes/ingest/2020-07_mag.md @@ -0,0 +1,159 @@ + +Using 2020-06-25 upstream MAG corpus snapshot. + +Ran munging from `scratch:ingest/mag` notes first. + +Expecting a couple million new ingest request URLs; this is the first "patch" +MAG ingest on top of existing already-run requests. + +Planning to skip the initial bulk ingest step, on the assumption that new URLs +have either been ingested already (eg, via continuous ingest pipeline) or need +crawling. + +## Generate Requests + + export LC_ALL=C + cat PaperUrls_mag_url_doi.all.txt | rg -a -v arxiv.org | rg -a "://" | ./mag_ingest_request.py - --created-date 2020-06-25 | pv -l > ingest_requests_mag-2020-06-25.json + => 28.7M 2:36:48 [3.06k/s] + + export LC_ALL=C + zcat PaperUrls_mag_url_pmid.txt.gz | rg -a -v arxiv.org | rg -a "://" | ./mag_ingest_request.py - --created-date 2020-06-25 --pmid | pv -l > ingest_requests_mag-2020-06-25.pmid.json + => 5.66M 0:29:28 [ 3.2k/s] + +## Persist Ingest Requests + + # small sample + head -n1000 /schnell/mag/20200625/ingest_requests_mag-2020-06-25.pmid.json | ./persist_tool.py ingest-request - + => Worker: Counter({'total': 1000, 'insert-requests': 319, 'update-requests': 0}) + + head -n1000 /schnell/mag/20200625/ingest_requests_mag-2020-06-25.json | ./persist_tool.py ingest-request - + Worker: Counter({'total': 1000, 'insert-requests': 304, 'update-requests': 0}) + + cat /schnell/mag/20200625/ingest_requests_mag-2020-06-25.pmid.json | ./persist_tool.py ingest-request - + => Worker: Counter({'total': 5662486, 'insert-requests': 1984605, 'update-requests': 0}) + + cat /schnell/mag/20200625/ingest_requests_mag-2020-06-25.json | ./persist_tool.py ingest-request - + => Worker: Counter({'total': 28743819, 'insert-requests': 7433465, 'update-requests': 0}) + +## Crawl/Dupe Status + +Overall status for old and new seeds, filtering out large (blocking) +publishers: + + SELECT ingest_file_result.status, COUNT(*) + FROM ingest_request + LEFT JOIN ingest_file_result + 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.link_source = 'mag' + AND ingest_request.base_url NOT LIKE '%journals.sagepub.com%' + AND ingest_request.base_url NOT LIKE '%pubs.acs.org%' + AND ingest_request.base_url NOT LIKE '%ahajournals.org%' + AND ingest_request.base_url NOT LIKE '%www.journal.csj.jp%' + AND ingest_request.base_url NOT LIKE '%aip.scitation.org%' + AND ingest_request.base_url NOT LIKE '%academic.oup.com%' + AND ingest_request.base_url NOT LIKE '%tandfonline.com%' + GROUP BY status + ORDER BY COUNT DESC + LIMIT 20; + + status | count + -------------------------------------+---------- + success | 19477651 + | 8238898 + redirect-loop | 2036494 + link-loop | 1330036 + no-pdf-link | 1304820 + terminal-bad-status | 648150 + no-capture | 545785 + gateway-timeout | 200143 + cdx-error | 149995 + spn2-cdx-lookup-failure | 80010 + wrong-mimetype | 57052 + wayback-error | 41032 + invalid-host-resolution | 37203 + petabox-error | 11167 + null-body | 6662 + spn2-error | 1698 + spn2-error:job-failed | 775 + spn2-error:invalid-url-syntax | 335 + spn2-error:soft-time-limit-exceeded | 191 + bad-redirect | 77 + (20 rows) + +Just the new seeds: + + SELECT ingest_file_result.status, COUNT(*) + FROM ingest_request + LEFT JOIN ingest_file_result + 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.link_source = 'mag' + AND ingest_request.base_url NOT LIKE '%journals.sagepub.com%' + AND ingest_request.base_url NOT LIKE '%pubs.acs.org%' + AND ingest_request.base_url NOT LIKE '%ahajournals.org%' + AND ingest_request.base_url NOT LIKE '%www.journal.csj.jp%' + AND ingest_request.base_url NOT LIKE '%aip.scitation.org%' + AND ingest_request.base_url NOT LIKE '%academic.oup.com%' + AND ingest_request.base_url NOT LIKE '%tandfonline.com%' + AND ingest_request.created > '2020-06-20' + GROUP BY status + ORDER BY COUNT DESC + LIMIT 20; + + status | count + -------------------------------------+--------- + | 8238851 + success | 787174 + no-capture | 42864 + redirect-loop | 31718 + terminal-bad-status | 31493 + no-pdf-link | 13025 + cdx-error | 11275 + wrong-mimetype | 6238 + link-loop | 3365 + wayback-error | 748 + gateway-timeout | 506 + null-body | 191 + spn2-cdx-lookup-failure | 99 + petabox-error | 89 + invalid-host-resolution | 70 + spn2-error | 7 + spn2-error:job-failed | 2 + spn2-error:soft-time-limit-exceeded | 1 + bad-gzip-encoding | 1 + (19 rows) + +## Heritrix Seedlist Generation + +Dump ingest requests (filtered for some domains that don't expect to crawl via +heritrix): + + COPY ( + SELECT row_to_json(ingest_request.*) FROM ingest_request + LEFT JOIN ingest_file_result + 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.link_source = 'mag' + AND (ingest_file_result.status = 'no-capture' + OR ingest_file_result.status IS NULL) + AND ingest_request.base_url NOT LIKE '%journals.sagepub.com%' + AND ingest_request.base_url NOT LIKE '%pubs.acs.org%' + AND ingest_request.base_url NOT LIKE '%ahajournals.org%' + AND ingest_request.base_url NOT LIKE '%www.journal.csj.jp%' + AND ingest_request.base_url NOT LIKE '%aip.scitation.org%' + AND ingest_request.base_url NOT LIKE '%academic.oup.com%' + AND ingest_request.base_url NOT LIKE '%tandfonline.com%' + ) TO '/grande/snapshots/mag_nocapture_20200708.rows.json'; + => 8784683 + + # in sandcrawler pipenv + ./scripts/ingestrequest_row2json.py /grande/snapshots/mag_nocapture_20200708.rows.json > /grande/snapshots/mag_nocapture_20200708.json + +Seedlist transform from here on covered in MAG crawl notes. |