From 35b19455b8014a513a9c31ca5795b2611bad06ce Mon Sep 17 00:00:00 2001 From: Bryan Newbold Date: Wed, 20 Jul 2022 18:12:56 -0700 Subject: bulk edit/import notes --- extra/bulk_edits/2022-07-12_jalc.md | 47 ++++++++++++++++++++++ extra/bulk_edits/2022-07-19_doaj.md | 78 +++++++++++++++++++++++++++++++++++++ extra/bulk_edits/CHANGELOG.md | 11 ++++++ 3 files changed, 136 insertions(+) create mode 100644 extra/bulk_edits/2022-07-12_jalc.md create mode 100644 extra/bulk_edits/2022-07-19_doaj.md (limited to 'extra') diff --git a/extra/bulk_edits/2022-07-12_jalc.md b/extra/bulk_edits/2022-07-12_jalc.md new file mode 100644 index 00000000..d9f09fee --- /dev/null +++ b/extra/bulk_edits/2022-07-12_jalc.md @@ -0,0 +1,47 @@ + +Import of a 2022-04 JALC DOI metadata snapshot. + +Note that we had downloaded a prior 2021-04 snapshot, but don't seem to have +ever imported it. + +## Download and Archive + +URL for bulk snapshot is available at the bottom of this page: + +More info: + + wget 'https://japanlinkcenter.org/lod/JALC-LOD-20220401.gz?jalcmetadatadl_1703' + wget 'http://japanlinkcenter.org/top/doc/JaLC_LOD_format.pdf' + wget 'http://japanlinkcenter.org/top/doc/JaLC_LOD_sample.pdf' + + mv 'JALC-LOD-20220401.gz?jalcmetadatadl_1703' JALC-LOD-20220401.gz + + ia upload jalc-bulk-metadata-2022-04 -m collection:ia_biblio_metadata jalc_logo.png JALC-LOD-20220401.gz JaLC_LOD_format.pdf JaLC_LOD_sample.pdf + +## Import + +As of 2022-07-19, 6,502,202 release hits for `doi_registrar:jalc`. + +Re-download the file: + + cd /srv/fatcat/datasets + wget 'https://archive.org/download/jalc-bulk-metadata-2022-04/JALC-LOD-20220401.gz' + gunzip JALC-LOD-20220401.gz + cd /srv/fatcat/src/python + + wc -l /srv/fatcat/datasets/JALC-LOD-20220401 + 9525225 + +Start with some samples: + + export FATCAT_AUTH_WORKER_JALC=[...] + shuf -n100 /srv/fatcat/datasets/JALC-LOD-20220401 | ./fatcat_import.py --batch-size 100 jalc - /srv/fatcat/datasets/ISSN-to-ISSN-L.txt + # Counter({'total': 100, 'exists': 89, 'insert': 11, 'skip': 0, 'update': 0}) + +Full import (single threaded): + + cat /srv/fatcat/datasets/JALC-LOD-20220401 | pv -l | ./fatcat_import.py --batch-size 100 jalc - /srv/fatcat/datasets/ISSN-to-ISSN-L.txt + # 9.53M 22:26:06 [ 117 /s] + # Counter({'total': 9510096, 'exists': 8589731, 'insert': 915032, 'skip': 5333, 'inserted.container': 119, 'update': 0}) + +Wow, almost a million new releases! 7,417,245 results for `doi_registrar:jalc`. diff --git a/extra/bulk_edits/2022-07-19_doaj.md b/extra/bulk_edits/2022-07-19_doaj.md new file mode 100644 index 00000000..d25f2dda --- /dev/null +++ b/extra/bulk_edits/2022-07-19_doaj.md @@ -0,0 +1,78 @@ + +Doing a batch import of DOAJ articles. Will need to do another one of these +soon after setting up daily (OAI-PMH feed) ingest. + +## Prep + + wget https://doaj.org/csv + wget https://doaj.org/public-data-dump/journal + wget https://doaj.org/public-data-dump/article + + mv csv journalcsv__doaj_20220719_2135_utf8.csv + mv journal doaj_journal_data_2022-07-19.tar.gz + mv article doaj_article_data_2022-07-19.tar.gz + + ia upload doaj_data_2022-07-19 -m collection:ia_biblio_metadata ../logo_cropped.jpg journalcsv__doaj_20220719_2135_utf8.csv doaj_journal_data_2022-07-19.tar.gz doaj_article_data_2022-07-19.tar.gz + + tar xvf doaj_journal_data_2022-07-19.tar.gz + cat doaj_journal_data_*/journal_batch_*.json | jq .[] -c | pv -l | gzip > doaj_journal_data_2022-07-19_all.json.gz + + tar xvf doaj_article_data_2022-07-19.tar.gz + cat doaj_article_data_*/article_batch*.json | jq .[] -c | pv -l | gzip > doaj_article_data_2022-07-19_all.json.gz + + ia upload doaj_data_2022-07-19 doaj_journal_data_2022-07-19_all.json.gz doaj_article_data_2022-07-19_all.json.gz + +On fatcat machine: + + cd /srv/fatcat/datasets + wget https://archive.org/download/doaj_data_2022-07-19/doaj_article_data_2022-07-19_all.json.gz + +## Prod Article Import + + git rev: 582495f66e5e08b6e257360097807711e53008d4 + (includes DOAJ container-id required patch) + + date: Tue Jul 19 22:46:42 UTC 2022 + + `doaj_id:*`: 1,335,195 hits + +Start with sample: + + zcat /srv/fatcat/datasets/doaj_article_data_2022-07-19_all.json.gz | shuf -n1000 > /srv/fatcat/datasets/doaj_article_data_2022-07-19_sample.json + + export FATCAT_AUTH_WORKER_DOAJ=[...] + cat /srv/fatcat/datasets/doaj_article_data_2022-07-19_sample.json | pv -l | ./fatcat_import.py doaj-article --issn-map-file /srv/fatcat/datasets/ISSN-to-ISSN-L.txt - + # Counter({'total': 1000, 'exists': 895, 'exists-fuzzy': 93, 'insert': 9, 'skip': 3, 'skip-no-container': 3, 'update': 0}) + +Pretty few imports. + +Full ingest: + + export FATCAT_AUTH_WORKER_DOAJ=[...] + zcat /srv/fatcat/datasets/doaj_article_data_2022-07-19_all.json.gz | pv -l | parallel -j6 --round-robin --pipe ./fatcat_import.py doaj-article --issn-map-file /srv/fatcat/datasets/ISSN-to-ISSN-L.txt - + # Counter({'total': 1282908, 'exists': 1145439, 'exists-fuzzy': 117120, 'insert': 16357, 'skip': 3831, 'skip-no-container': 2641, 'skip-title': 1190, 'skip-doaj-id-mismatch': 161, 'update': 0}) + +Times 6x, around 100k releases added. + +Got a bunch of: + + /1/srv/fatcat/src/python/fatcat_tools/importers/doaj_article.py:233: UserWarning: unexpected DOAJ ext_id match after lookup failed doaj=fcdb7a7a9729403d8d99a21f6970dd1d ident=wesvmjwihvblzayfmrvvgr4ulm + warnings.warn(warn_str) + /1/srv/fatcat/src/python/fatcat_tools/importers/doaj_article.py:233: UserWarning: unexpected DOAJ ext_id match after lookup failed doaj=1455dfe24583480883dbbb293a4bc0c6 ident=lfw57esesjbotms3grvvods5dq + warnings.warn(warn_str) + /1/srv/fatcat/src/python/fatcat_tools/importers/doaj_article.py:233: UserWarning: unexpected DOAJ ext_id match after lookup failed doaj=88fa65a33c8e484091fc76f4cda59c25 ident=22abqt5qe5e7ngjd5fkyvzyc4q + warnings.warn(warn_str) + /1/srv/fatcat/src/python/fatcat_tools/importers/doaj_article.py:233: UserWarning: unexpected DOAJ ext_id match after lookup failed doaj=eb7b03dc3dc340cea36891a68a50cce7 ident=ljedohlfyzdkxebgpcswjtd77q + warnings.warn(warn_str) + /1/srv/fatcat/src/python/fatcat_tools/importers/doaj_article.py:233: UserWarning: unexpected DOAJ ext_id match after lookup failed doaj=519617147ce248ea88d45ab098342153 ident=a63bqkttrbhyxavfr7li2w2xf4 + +Should investigate! + +Also, noticed that DOAJ importer is hitting `api.fatcat.wiki`, not the public +API endpoint. Guessing this is via fuzzycat. + +1,434,266 results for `doaj_id:*`. + +Then did a follow-up sandcrawler ingest, see notes in that repository. Note +that newer ingest can crawl doaj.org, bypassing the sandcrawler SQL load, but +the direct crawling is probably still faster. diff --git a/extra/bulk_edits/CHANGELOG.md b/extra/bulk_edits/CHANGELOG.md index f7b9e536..732cbb2f 100644 --- a/extra/bulk_edits/CHANGELOG.md +++ b/extra/bulk_edits/CHANGELOG.md @@ -16,6 +16,17 @@ Ran a journal-level metadata update, using chocula. Cleaned up just under 500 releases with missing `container_id` from an older DOAJ article import. +Imported roughly 100k releases from DOAJ, new since 2022-04. + +Imported roughly 2.7 million new ORCiD `creator` entities, using the 2021 dump +(first update since 2020 dump). + +Imported almost 1 million new DOI release entities from JALC, first update in +more than a year. + +Imported at least 400 new dblp containers, and an unknown number of new dblp +release entities. + ## 2022-04 -- cgit v1.2.3