comparison bin/plinks.sh @ 6:0f494c76a887

refactor to address tarred-up pdfs
author Henry S. Thompson <ht@inf.ed.ac.uk>
date Sun, 23 Feb 2020 16:48:34 +0000
parents 65a56c0d1c1f
children 25ca3505b4d7
comparison
equal deleted inserted replaced
5:a28d731977da 6:0f494c76a887
1 #!/usr/bin/bash 1 #!/usr/bin/bash
2 mkdir -p $TMPDIR 2 mkdir -p $TMPDIR
3 echo $(date) $(hostname) 3 echo $(date) $(hostname)
4 cd /beegfs/common_crawl/CC-MAIN-2019-35/pdfs
5 h=$(hostname) 4 h=$(hostname)
6 hn=${h##*n} 5 hn=${h##*n}
7 parallel --will-cite -j 71 --pipepart -a lm/$(hostname)_pdfFilesWithLM doPlinks.sh ${hn}_'{#}' 6 if [ $hn -eq 0 ]; then echo {001..002}; else echo {003..004}; fi |\
7 tr ' ' '\012' |parallel --will-cite -j 71 -N 1 doPlinks.sh ${hn} '{#}' '{}'
8 echo $(date) $(hostname) 8 echo $(date) $(hostname)