comparison bin/_nl1.sh @ 4:f27061e8a9da

convert to no longer need uniq -c
author Henry S. Thompson <ht@inf.ed.ac.uk>
date Wed, 20 Jul 2022 19:38:30 +0100
parents b4801f5696b2
children
comparison
equal deleted inserted replaced
3:668579197bec 4:f27061e8a9da
28 28
29 doit () { 29 doit () {
30 echo $(date) start $1 $task $PARALLEL_SEQ 1>&2 30 echo $(date) start $1 $task $PARALLEL_SEQ 1>&2
31 fgrep ' w ' $srcdir/$1.tsv | \ 31 fgrep ' w ' $srcdir/$1.tsv | \
32 awk 'BEGIN {while (getline < "'$langfile'") {l[$0]=1}} 32 awk 'BEGIN {while (getline < "'$langfile'") {l[$0]=1}}
33 {if (l[$4]) {print $1,$4}}' | uniq -c | \ 33 {if (l[$5]) {print $1,$2,$5}}' | \
34 $W/shared/bin/uniq_merge.py > $resdir/${langs}_$1.tsv 34 $W/shared/bin/uniq_merge.py > $resdir/${langs}_$1.tsv
35 echo $(date) end $1 $task $PARALLEL_SEQ 1>&2 35 echo $(date) end $1 $task $PARALLEL_SEQ 1>&2
36 } 36 }
37 37
38 export -f doit 38 export -f doit
39 export srcdir resdir task langs langfile 39 export srcdir resdir task langs langfile
40 40
41 seq $s1 $sn | while read i 41 $W/hst/bin/share_by_task.sh -f "%03g\n" -s $s1 $sn $n $task | parallel -j $pjobs doit '{}'
42 do if [ $((i % $n)) -eq $task ]
43 then printf '%03g\n' $i
44 fi
45 done | \
46 parallel -j $pjobs doit '{}'
47 42
48 echo $(date) task $n.$task on $nodename:$N.$node end 1>&2 43 echo $(date) task $n.$task on $nodename:$N.$node end 1>&2
49 44