1 files changed, 18 insertions, 0 deletions
diff --git a/sql/dump_ungrobid_pdf.sql b/sql/dump_ungrobid_pdf.sql
new file mode 100644
index 0000000..81caf18
--- /dev/null
+++ b/sql/dump_ungrobid_pdf.sql
@@ -0,0 +1,18 @@
+
+-- Run like:
+--   psql sandcrawler < dump_ungrobid_pdf.sql
+
+BEGIN TRANSACTION ISOLATION LEVEL SERIALIZABLE READ ONLY DEFERRABLE;
+
+COPY (
+  SELECT DISTINCT ON (cdx.sha1hex) row_to_json(cdx)
+  FROM cdx
+  WHERE cdx.mimetype = 'application/pdf'
+  AND NOT EXISTS (SELECT grobid.sha1hex FROM grobid WHERE cdx.sha1hex = grobid.sha1hex AND grobid.status IS NOT NULL)
+  -- uncomment/comment this to control whether only fatcat files are included
+  --AND EXISTS (SELECT fatcat_file.sha1hex FROM fatcat_file WHERE cdx.sha1hex = fatcat_file.sha1hex)
+)
+TO '/srv/sandcrawler/tasks/dump_ungrobided_pdf.fatcat.2020-08-04.json'
+WITH NULL '';
+
+ROLLBACK;