2a551773f1cca84e8acb73cf1856f78111d53002 angie Fri Jan 22 10:57:17 2021 -0800 Use curl not wget to avoid annoying .1, .2 files on retries. diff --git src/hg/utils/otto/sarscov2phylo/getCogUk.sh src/hg/utils/otto/sarscov2phylo/getCogUk.sh index ab711c7..b2affda 100755 --- src/hg/utils/otto/sarscov2phylo/getCogUk.sh +++ src/hg/utils/otto/sarscov2phylo/getCogUk.sh @@ -1,27 +1,25 @@ #!/bin/bash set -beEux -o pipefail # Download latest COG-UK fasta and metadata; update $ottoDir/cogUk.latest link. scriptDir=$(dirname "${BASH_SOURCE[0]}") source $scriptDir/util.sh today=$(date +%F) ottoDir=/hive/data/outside/otto/sarscov2phylo cogUrlBase=https://cog-uk.s3.climb.ac.uk/phylogenetics/latest mkdir -p $ottoDir/cogUk.$today cd $ottoDir/cogUk.$today -wget -q $cogUrlBase/cog_all.fasta -wget -q $cogUrlBase/cog_metadata.csv - -xz -f -T 50 cog_all.fasta +curl -S -s $cogUrlBase/cog_all.fasta | xz -T 50 > cog_all.fasta.xz +curl -S -s $cogUrlBase/cog_metadata.csv > cog_metadata.csv tail -n +2 cog_metadata.csv \ | awk -F, '{print $1 "\t" $4;}' | sort > cogUkToDate rm -f $ottoDir/cogUk.latest ln -s cogUk.$today $ottoDir/cogUk.latest