# HG changeset patch
# User Henry S. Thompson <ht@markup.co.uk>
# Date 1544452998 0
# Node ID 2a0dab424418854a9c866205b9c5ca7e1230bb5f
# Parent  7a4e49689935e90a4f41e75a31593837bdad5388
cci path hack changed for 2018.04

diff -r 7a4e49689935 -r 2a0dab424418 workers/bin/ptimedWhich.sh
--- a/workers/bin/ptimedWhich.sh	Mon Dec 03 21:10:02 2018 +0000
+++ b/workers/bin/ptimedWhich.sh	Mon Dec 10 14:43:18 2018 +0000
@@ -66,7 +66,8 @@
 while read s
 do
  url="https://commoncrawl.s3.amazonaws.com/$s"
- cci=$(echo $s | tr '/-' ' ' | awk '{print $3,$4,$8,$13}' |tr ' ' \-)
+ # below for 2018-04, for 2017-04 needs $13 instead of $14
+ cci=$(echo $s | tr '/-' ' ' | awk '{print $3,$4,$8,$14}' |tr ' ' \-)
  echo $url /var/data/$cci
 done < ifile.txt 2>> $res/errs | \
  parallel --pipe -N$((N / wp)) -j $wp "bash -c \"tryread 2>>$res/errs{#}\"" 2>>$res/errs || pRes=$?