aboutsummaryrefslogtreecommitdiffstats
path: root/python
diff options
context:
space:
mode:
authorBryan Newbold <bnewbold@archive.org>2022-05-03 17:12:48 -0700
committerBryan Newbold <bnewbold@archive.org>2022-05-03 17:12:48 -0700
commit00ae74378413e87f230c88113ff8163a6f969d63 (patch)
tree16cdcbde7a002704e80f494b7fd13fc5c19dd695 /python
parentef0421567dd67a248d0f92f32ad4e14ae0776920 (diff)
downloadsandcrawler-00ae74378413e87f230c88113ff8163a6f969d63.tar.gz
sandcrawler-00ae74378413e87f230c88113ff8163a6f969d63.zip
switch default kafka-broker host from wbgrp-svc263 to wbgrp-svc350
Diffstat (limited to 'python')
-rwxr-xr-xpython/grobid_tool.py2
-rwxr-xr-xpython/pdftrio_tool.py2
2 files changed, 2 insertions, 2 deletions
diff --git a/python/grobid_tool.py b/python/grobid_tool.py
index 029cbf1..3ffac98 100755
--- a/python/grobid_tool.py
+++ b/python/grobid_tool.py
@@ -5,7 +5,7 @@ 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 | 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 -
+ cat /srv/sandcrawler/tasks/ungrobided.2019-09-23.json | pv -l | parallel -j30 --pipe ./grobid_tool.py --kafka-env prod --kafka-hosts wbgrp-svc350.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 argparse
diff --git a/python/pdftrio_tool.py b/python/pdftrio_tool.py
index 9d3010e..24b749d 100755
--- a/python/pdftrio_tool.py
+++ b/python/pdftrio_tool.py
@@ -5,7 +5,7 @@ text extraction.
Example of large parallel run, locally:
-cat /srv/sandcrawler/tasks/something.cdx | pv -l | parallel -j30 --pipe ./pdftrio_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 --pdftrio-host http://localhost:3939 -j0 classify-pdf-json -
+cat /srv/sandcrawler/tasks/something.cdx | pv -l | parallel -j30 --pipe ./pdftrio_tool.py --kafka-env prod --kafka-hosts wbgrp-svc350.us.archive.org:9092,wbgrp-svc284.us.archive.org:9092,wbgrp-svc285.us.archive.org:9092 --kafka-mode --pdftrio-host http://localhost:3939 -j0 classify-pdf-json -
"""
import argparse