aboutsummaryrefslogtreecommitdiffstats
path: root/python
diff options
context:
space:
mode:
authorBryan Newbold <bnewbold@archive.org>2019-12-27 18:11:10 -0800
committerBryan Newbold <bnewbold@archive.org>2020-01-02 18:12:58 -0800
commit6093c9a0c9b65cdf790f200395e2d44d4fe6278b (patch)
tree967145462d5804e05195088fb56eb6374e3a2c34 /python
parent2a7a51e9ee154170ad6d48b36281f6ee702da582 (diff)
downloadsandcrawler-6093c9a0c9b65cdf790f200395e2d44d4fe6278b.tar.gz
sandcrawler-6093c9a0c9b65cdf790f200395e2d44d4fe6278b.zip
remove unused filter in grobid worker
Diffstat (limited to 'python')
-rw-r--r--python/sandcrawler/persist.py1
1 files changed, 0 insertions, 1 deletions
diff --git a/python/sandcrawler/persist.py b/python/sandcrawler/persist.py
index 45df8c7..c3f6b08 100644
--- a/python/sandcrawler/persist.py
+++ b/python/sandcrawler/persist.py
@@ -256,7 +256,6 @@ class PersistGrobidWorker(SandcrawlerWorker):
r['metadata'] = metadata
if not self.s3_only:
- grobid_batch = [r['grobid'] for r in batch if r.get('grobid')]
resp = self.db.insert_grobid(self.cur, batch, on_conflict="update")
self.counts['insert-grobid'] += resp[0]
self.counts['update-grobid'] += resp[1]