diff options
| author | Bryan Newbold <bnewbold@archive.org> | 2019-10-04 14:28:40 -0700 |
|---|---|---|
| committer | Bryan Newbold <bnewbold@archive.org> | 2019-10-04 14:28:40 -0700 |
| commit | ab1f3fe806e40122a93c975d2253f2c14035952e (patch) | |
| tree | 30d565fa83d1f324bf8cb4d30b908a3d02c9c2bc | |
| parent | 1a9df830d128cbd0bf80ff585785e226a6cb9019 (diff) | |
| download | sandcrawler-ab1f3fe806e40122a93c975d2253f2c14035952e.tar.gz sandcrawler-ab1f3fe806e40122a93c975d2253f2c14035952e.zip | |
grobid-output-pg, not grobid-output-json
| -rwxr-xr-x | python/grobid_tool.py | 6 |
1 files changed, 2 insertions, 4 deletions
diff --git a/python/grobid_tool.py b/python/grobid_tool.py index 9af0ab2..e787cdf 100755 --- a/python/grobid_tool.py +++ b/python/grobid_tool.py @@ -6,9 +6,7 @@ might go to stdout, or might go to Kafka topic. Example of large parallel run, locally: - cat /srv/sandcrawler/tasks/ungrobided.2019-09-23.json \ - | parallel -j6 --pipe \ - ./grobid_tool.py --kafka-env qa --kafka-hosts wbgrp-svc263.us.archive.org:9092,wbgrp-svc284.us.archive.org:9092,wbgrp-svc285.us.archive.org:9092 --kafka-mode --grobid-host http://localhost:8070 -j10 extract-json - + cat /srv/sandcrawler/tasks/ungrobided.2019-09-23.json | pv -l | parallel -j30 --pipe ./grobid_tool.py --kafka-env prod --kafka-hosts wbgrp-svc263.us.archive.org:9092,wbgrp-svc284.us.archive.org:9092,wbgrp-svc285.us.archive.org:9092 --kafka-mode --grobid-host http://localhost:8070 -j0 extract-json - """ import sys @@ -93,7 +91,7 @@ def main(): args.sink = None if args.kafka_mode: - produce_topic = "sandcrawler-{}.grobid-output-json".format(args.kafka_env) + produce_topic = "sandcrawler-{}.grobid-output-pg".format(args.kafka_env) print("Running in kafka output mode, publishing to {}\n".format(produce_topic)) args.sink = KafkaGrobidSink(kafka_hosts=args.kafka_hosts, produce_topic=produce_topic) |