diff options
author | Bryan Newbold <bnewbold@archive.org> | 2020-11-06 18:32:35 -0800 |
---|---|---|
committer | Bryan Newbold <bnewbold@archive.org> | 2020-11-06 18:32:35 -0800 |
commit | 175019c96fced3e21d0f60ea1a4a37da6b8872ac (patch) | |
tree | f42fbbe9c8ac06ae9eb06373ab9eec96d2b3a177 /sql/dump_unmatched_glutton_pdf.sql | |
parent | b0b66c20c6ffb9d8acc626068964d7dfd5d3bcdc (diff) | |
parent | 47ca1a273912c8836630b0930b71a4e66fd2c85b (diff) | |
download | sandcrawler-175019c96fced3e21d0f60ea1a4a37da6b8872ac.tar.gz sandcrawler-175019c96fced3e21d0f60ea1a4a37da6b8872ac.zip |
Merge branch 'bnewbold-html-ingest'
Diffstat (limited to 'sql/dump_unmatched_glutton_pdf.sql')
-rw-r--r-- | sql/dump_unmatched_glutton_pdf.sql | 19 |
1 files changed, 19 insertions, 0 deletions
diff --git a/sql/dump_unmatched_glutton_pdf.sql b/sql/dump_unmatched_glutton_pdf.sql new file mode 100644 index 0000000..d089c7e --- /dev/null +++ b/sql/dump_unmatched_glutton_pdf.sql @@ -0,0 +1,19 @@ + +-- Run like: +-- psql sandcrawler < THING.sql > THING.2019-09-23.json + +BEGIN TRANSACTION ISOLATION LEVEL SERIALIZABLE READ ONLY DEFERRABLE; + +COPY ( + SELECT row_to_json(grobid) + FROM grobid + LEFT JOIN fatcat_file ON grobid.sha1hex = fatcat_file.sha1hex + WHERE fatcat_file.sha1hex IS NULL + AND grobid.fatcat_release IS NOT NULL + LIMIT 1000 +) +TO '/grande/snapshots/dump_unmatched_glutton_pdf.2020-06-30.json'; +--TO STDOUT +--WITH NULL ''; + +ROLLBACK; |