diff options
| author | Bryan Newbold <bnewbold@archive.org> | 2020-05-13 14:09:43 -0700 |
|---|---|---|
| committer | Bryan Newbold <bnewbold@archive.org> | 2020-05-13 14:09:43 -0700 |
| commit | d2c95fb7a98e051639ef875c1d9aa4ba4b794509 (patch) | |
| tree | 8f8dec94ca329415a0b7d2e923865ee0d6cc210d | |
| parent | 959544196679de53c379d8d71feff20c59cd40f2 (diff) | |
| download | fatcat-covid19-d2c95fb7a98e051639ef875c1d9aa4ba4b794509.tar.gz fatcat-covid19-d2c95fb7a98e051639ef875c1d9aa4ba4b794509.zip | |
improve pipeline commands
| -rwxr-xr-x | bin/run_pipeline.sh | 25 | ||||
| -rw-r--r-- | notes/pipeline_commands.md | 6 |
2 files changed, 23 insertions, 8 deletions
diff --git a/bin/run_pipeline.sh b/bin/run_pipeline.sh index 7ace3b3..5c1f4de 100755 --- a/bin/run_pipeline.sh +++ b/bin/run_pipeline.sh @@ -19,31 +19,39 @@ set -x mkdir -p metadata fulltext_web if [ ! -f metadata/cord19.$CORDDATE.csv ]; then - wget https://archive.org/download/s2-cord19-dataset/cord19.$CORDDATE.csv -O metadata/cord19.$CORDDATE.csv --no-clobber || true + rm metadata/cord19.$CORDDATE.csv.wip || true + wget https://archive.org/download/s2-cord19-dataset/cord19.$CORDDATE.csv -O metadata/cord19.$CORDDATE.csv.wip --no-clobber + mv metadata/cord19.$CORDDATE.csv.wip metadata/cord19.$CORDDATE.csv fi if [ ! -f metadata/cord19.$CORDDATE.json ]; then - ./covid19_tool.py parse-cord19 metadata/cord19.$CORDDATE.csv > metadata/cord19.$CORDDATE.json + ./covid19_tool.py parse-cord19 metadata/cord19.$CORDDATE.csv > metadata/cord19.$CORDDATE.json.wip + mv metadata/cord19.$CORDDATE.json.wip metadata/cord19.$CORDDATE.json fi if [ ! -f metadata/cord19.$CORDDATE.enrich.json ]; then - cat metadata/cord19.$CORDDATE.json | parallel -j10 --linebuffer --round-robin --pipe ./covid19_tool.py enrich-fatcat - | pv -l > metadata/cord19.$CORDDATE.enrich.json + cat metadata/cord19.$CORDDATE.json | parallel -j10 --linebuffer --round-robin --pipe ./covid19_tool.py enrich-fatcat - | pv -l > metadata/cord19.$CORDDATE.enrich.json.wip + mv metadata/cord19.$CORDDATE.enrich.json.wip metadata/cord19.$CORDDATE.enrich.json fi if [ ! -f metadata/cord19.$CORDDATE.missing.json ]; then - cat metadata/cord19.$CORDDATE.enrich.json | jq 'select(.release_id == null) | .cord19_paper' -c > metadata/cord19.$CORDDATE.missing.json + cat metadata/cord19.$CORDDATE.enrich.json | jq 'select(.release_id == null) | .cord19_paper' -c > metadata/cord19.$CORDDATE.missing.json.wip + mv metadata/cord19.$CORDDATE.missing.json.wip metadata/cord19.$CORDDATE.missing.json fi if [ ! -f metadata/fatcat_hits.$TODAY.enrich.json ]; then - ./covid19_tool.py query-fatcat | pv -l > metadata/fatcat_hits.$TODAY.enrich.json + ./covid19_tool.py query-fatcat | pv -l > metadata/fatcat_hits.$TODAY.enrich.json.wip + mv metadata/fatcat_hits.$TODAY.enrich.json.wip metadata/fatcat_hits.$TODAY.enrich.json fi if [ ! -f metadata/combined.$TODAY.enrich.json ]; then - cat metadata/fatcat_hits.$TODAY.enrich.json metadata/cord19.$CORDDATE.enrich.json | ./covid19_tool.py dedupe | pv -l > metadata/combined.$TODAY.enrich.json + cat metadata/fatcat_hits.$TODAY.enrich.json metadata/cord19.$CORDDATE.enrich.json | ./covid19_tool.py dedupe | pv -l > metadata/combined.$TODAY.enrich.json.wip + mv metadata/combined.$TODAY.enrich.json.wip metadata/combined.$TODAY.enrich.json fi if [ ! -f fatcat_web_$TODAY.log ]; then - cat metadata/combined.$TODAY.enrich.json | jq .fatcat_release -c | parallel -j20 --linebuffer --round-robin --pipe ./bin/deliver_file2disk.py --disk-dir fulltext_web - | pv -l > fatcat_web_$TODAY.log + cat metadata/combined.$TODAY.enrich.json | jq .fatcat_release -c | parallel -j20 --linebuffer --round-robin --pipe ./bin/deliver_file2disk.py --disk-dir fulltext_web - | pv -l > fatcat_web_$TODAY.log.wip + mv fatcat_web_$TODAY.log.wip fatcat_web_$TODAY.log fi if [ ! -f metadata/derivatives.$TODAY.stamp ]; then @@ -52,7 +60,8 @@ if [ ! -f metadata/derivatives.$TODAY.stamp ]; then fi if [ ! -f metadata/combined.$TODAY.fulltext.json ]; then - ./covid19_tool.py enrich-derivatives metadata/combined.$TODAY.enrich.json --base-dir fulltext_web/ | pv -l > metadata/combined.$TODAY.fulltext.json + ./covid19_tool.py enrich-derivatives metadata/combined.$TODAY.enrich.json --base-dir fulltext_web/ | pv -l > metadata/combined.$TODAY.fulltext.json.wip + mv metadata/combined.$TODAY.fulltext.json.wip metadata/combined.$TODAY.fulltext.json fi echo "## Fulltext Inclusion Counts" diff --git a/notes/pipeline_commands.md b/notes/pipeline_commands.md index ca3106c..94d7480 100644 --- a/notes/pipeline_commands.md +++ b/notes/pipeline_commands.md @@ -67,8 +67,14 @@ Create and index existing `fatcat_release` schema: # in fatcat python directory, pipenv shell export LC_ALL=C.UTF-8 + export TODAY="`TZ=UTC date --iso-8601=date`" cat /srv/fatcat_covid19/src/metadata/combined.$TODAY.enrich.json | jq .fatcat_release -c | rg -v '^null$' | pv -l | ./fatcat_transform.py elasticsearch-releases - - | esbulk -verbose -size 1000 -id ident -w 8 -index covid19_fatcat_release -type release +Attempt crawl/ingest of missing papers: + + ./fatcat_ingest.py --env prod --enqueue-kafka --elasticsearch-index covid19_fatcat_release --fatcat-api-url https://api.fatcat.wiki/v0 --kafka-hosts wbgrp-svc263.us.archive.org --allow-non-oa query "pmcid:*" + ./fatcat_ingest.py --env prod --enqueue-kafka --elasticsearch-index covid19_fatcat_release --fatcat-api-url https://api.fatcat.wiki/v0 --kafka-hosts wbgrp-svc263.us.archive.org --allow-non-oa query "-_exists_:pmcid doi:*" + ## GROBID Processing zip -r fulltext_web.zip fulltext_web |