diff options
Diffstat (limited to 'mapreduce')
| -rw-r--r-- | mapreduce/README.md | 7 | ||||
| -rwxr-xr-x | mapreduce/extraction_cdx_grobid.py | 8 |
2 files changed, 12 insertions, 3 deletions
diff --git a/mapreduce/README.md b/mapreduce/README.md index b63e84b..aebc160 100644 --- a/mapreduce/README.md +++ b/mapreduce/README.md @@ -33,6 +33,7 @@ running on a devbox and GROBID running on a dedicated machine: Running from the cluster: # Create tarball of virtualenv + export PIPENV_VENV_IN_PROJECT=1 pipenv shell export VENVSHORT=`basename $VIRTUAL_ENV` tar -czf $VENVSHORT.tar.gz -C /home/bnewbold/.local/share/virtualenvs/$VENVSHORT . @@ -60,9 +61,9 @@ Actual invocation to run on Hadoop cluster (running on an IA devbox, where hadoop environment is configured): # Create tarball of virtualenv - pipenv shell - export VENVSHORT=`basename $VIRTUAL_ENV` - tar -czf $VENVSHORT.tar.gz -C /home/bnewbold/.local/share/virtualenvs/$VENVSHORT . + export PIPENV_VENV_IN_PROJECT=1 + pipenv install --deploy + tar -czf venv-current.tar.gz -C .venv . ./backfill_hbase_from_cdx.py \ --hbase-host wbgrp-svc263.us.archive.org \ diff --git a/mapreduce/extraction_cdx_grobid.py b/mapreduce/extraction_cdx_grobid.py index 7771e45..ed82a5e 100755 --- a/mapreduce/extraction_cdx_grobid.py +++ b/mapreduce/extraction_cdx_grobid.py @@ -36,6 +36,10 @@ from grobid2json import teixml2json # Yep, a global. Gets DSN from `SENTRY_DSN` environment variable sentry_client = raven.Client() +# Specific poison-pill rows we should skip +KEY_BLACKLIST = ( + 'sha1:DLCCSMMVTCCIR6LRXHEQLZ4PWO6NG2YT', # "failed to guess ARC header format" +) class MRExtractCdxGrobid(MRJob): @@ -203,6 +207,10 @@ class MRExtractCdxGrobid(MRJob): yield _, status return key = info['key'] + if key in KEY_BLACKLIST: + self.increment_counter('lines', 'blacklist') + yield _, dict(status='blacklist', key=key) + return # Note: this may not get "cleared" correctly sentry_client.extra_context(dict(row_key=key)) |