Mercurial > hg > cc > cirrus_home
comparison bin/doExtract.sh @ 35:ec99b2d1d2fc
sync up filenames and log names,
attempt to save more smaller tar files
author | Henry S. Thompson <ht@inf.ed.ac.uk> |
---|---|
date | Thu, 26 Mar 2020 12:24:30 +0000 |
parents | 4c23b1766692 |
children | e912ed51146a |
comparison
equal
deleted
inserted
replaced
34:e34cda3e7483 | 35:ec99b2d1d2fc |
---|---|
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}/${segid}/CC-MAIN-*.warc.gz | \ |
10 parallel --joblog job_${jobid}.log -j 18 -N 1 'id=$(echo {} | cut -f 6 -d / | cut -f 3- -d - | cut -f 1 -d .) ; \ | 10 parallel --joblog job_${jobid}.log -j 18 -N 1 'id=$(echo {} | cut -f 6 -d / | cut -f 3- -d - | cut -f 1 -d .) ; \ |
11 echo starting $id $(date) >> logs/${jobid}_{#}_log ; \ | 11 fid=$(printf "%03.0f" $(({#} - 1))) ; \ |
12 unpigz -dp 1 -c {} | $HOME/lib/valhalla/bin/warc.sh ${id} application/pdf 2>> logs/${jobid}_{#}_log ; \ | 12 lf=logs/${jobid}_${fid}_log ; \ |
13 echo finished ${id} $(date) >> logs/${jobid}_{#}_log' | 13 echo starting $id $(date) >> $lf ; \ |
14 unpigz -dp 1 -c {} | $HOME/lib/valhalla/bin/warc.sh ${id} application/pdf 2>> $lf ; \ | |
15 echo finished ${id} $(date) >> $lf' | |
14 res=$? | 16 res=$? |
15 echo $(date) $(hostname) $jobid $segid $res | 17 echo $(date) $(hostname) $jobid $segid $res |
16 if [ $res = 0 ] | 18 if [ $res = 0 ] |
17 then | 19 then |
18 cat ../by11s.txt | while read | 20 pfx=$(ls /beegfs/common_crawl/CC-MAIN-${ccid}/${segid}/CC-MAIN-*-00000.warc.gz |\ |
19 tar -cf /beegfs/common_crawl/CC-MAIN-${ccid}/${segid}/extracts.tar * && | 21 cut -f 6 -d / | cut -f 3,4 -d -) |
20 echo $(date) $(hostname) $jobid /beegfs/common_crawl/CC-MAIN-${ccid}/${segid}/extracts.tar | 22 |
23 cat ../by11s.txt | while read i j | |
24 do ((n=i/11)) | |
25 tar -cf /beegfs/common_crawl/CC-MAIN-${ccid}/${segid}/extract_${n}.tar.gz \ | |
26 $(seq $i $j | xargs -I ^ bash -c '{ k=^; printf "${pfx}-%05.0f_* logs/?_%03.0f_log\n" $k $k ; }') | |
27 done && | |
28 echo $(date) $(hostname) $jobid /beegfs/common_crawl/CC-MAIN-${ccid}/${segid}/extracts_\{0..${n}\}.tar | |
21 fi | 29 fi |