Mercurial > hg > cc > cirrus_home
comparison bin/doExtract.sh @ 29:a95a1b31f5e9
try to fix multi-line lossage
author | Henry S. Thompson <ht@inf.ed.ac.uk> |
---|---|
date | Wed, 18 Mar 2020 21:54:47 +0000 |
parents | 7a2bc060230d |
children | 1e479e3ea9c4 |
comparison
equal
deleted
inserted
replaced
28:70c04a452599 | 29:a95a1b31f5e9 |
---|---|
1 #!/bin/bash | 1 #!/bin/bash |
2 # Usage extract.sh ccid jobid segid | 2 # Usage doExtract.sh ccid jobid segid |
3 ccid=$1 | 3 ccid=$1 |
4 jobid=$2 | 4 export jobid=$2 |
5 segid=$3 | 5 segid=$3 |
6 echo $(date) $(hostname) $jobid $segid | 6 echo $(date) $(hostname) $jobid $segid |
7 mkdir -p $segid/logs | 7 mkdir -p $segid/logs |
8 cd $segid | 8 cd $segid |
9 ls /beegfs/common-crawl/CC-MAIN-${ccid}/${segid}/CC-MAIN-*.warc.gz | \ | 9 ls /beegfs/common_crawl/CC-MAIN-${ccid}/${segid}/CC-MAIN-*.warc.gz | \ |
10 parallel -j 12 -N 1 '{ | 10 parallel -j 12 -N 1 id=$(echo {} | cut -f 6 -d / | cut -f 3- -d - | cut -f 1 -d .) \; \ |
11 id=$(echo {} | cut -f 6 -d / | cut -f 3- -d - | cut -f 1 -d .) | 11 echo starting \$id \$(date) >> logs/${jobid}_{#}_log \; \ |
12 echo starting $id $(date) >> logs/${jobid}_{#}_log | 12 unpigz -dp 1 -c {} | $HOME/lib/valhalla/bin/warc.sh ${id} application/pdf 2>> logs/${jobid}_{#}_log \; \ |
13 unpigz -dp 1 -c {} | $HOME/lib/valhalla/bin/warc.sh ${id} application/pdf 2>> logs/${jobid}_{#}_log | |
14 echo finished ${id} $(date) >> logs/${jobid}_{#}_log | 13 echo finished ${id} $(date) >> logs/${jobid}_{#}_log |
15 }' | 14 res=$? |
16 echo $(date) $(hostname) $jobid $segid $? | 15 echo $(date) $(hostname) $jobid $segid $res |
17 tar -cf /beegfs/common-crawl/CC-MAIN-${ccid}/${segid}/extracts.tar * | 16 if [ $res = 0 ] |
18 echo $(date) $(hostname) $jobid /beegfs/common-crawl/CC-MAIN-${ccid}/${segid}/extracts.tar | 17 then |
18 tar -cf /beegfs/common_crawl/CC-MAIN-${ccid}/${segid}/extracts.tar * | |
19 echo $(date) $(hostname) $jobid /beegfs/common_crawl/CC-MAIN-${ccid}/${segid}/extracts.tar | |
20 fi |