#!/usr/bin/bash
cat /beegfs/common_crawl/CC-MAIN-2019-35/bigtar.txt | \
  parallel --will-cite -j 8 -n 1 '
    segid={}
    echo $segid starting
    ccid=2019-35
    h=/beegfs/common_crawl/CC-MAIN-${ccid}
    mkdir -p /dev/shm/rex/${segid}
    cd /dev/shm/rex/${segid}
    tar -xf ${h}/${segid}/extracts.tar
    pfx=$(ls ${h}/${segid}/CC-MAIN-*-00000.warc.gz |\
	  cut -f 6 -d / | cut -f 3,4 -d -)
    cat ../by11s.txt | while read i j
       do ((n=i/11))
       tar -cf /beegfs/common_crawl/CC-MAIN-${ccid}/${segid}/extract_${n}.tar \
	 $(seq $i $j | xargs -I ^ bash -c "{ k=^; printf \"\${0}-%05.0f_* logs/*_%03.0f_log\n\" \$k \$k ; }" $pfx)
       done
    echo $(date) $segid done
'