02d1894196c846ac4eedfee782ff1bf182f2adbf angie Wed Mar 17 11:56:18 2021 -0700 Tolerate another condition where grep finding 0 results crashed the script. diff --git src/hg/utils/otto/sarscov2phylo/updatePublicTree.sh src/hg/utils/otto/sarscov2phylo/updatePublicTree.sh index bc1f847..8a750ff 100755 --- src/hg/utils/otto/sarscov2phylo/updatePublicTree.sh +++ src/hg/utils/otto/sarscov2phylo/updatePublicTree.sh @@ -104,35 +104,38 @@ #conda install -c bioconda mafft alignedFa=new.aligned.fa rm -f $alignedFa export TMPDIR=/dev/shm time bash ~angie/github/sarscov2phylo/scripts/global_profile_alignment.sh \ -i new.fa \ -o $alignedFa \ -t 50 faSize $alignedFa # Now make a renaming that keeps all the prevNames and adds full names (with shortened dates) # for the new seqs. renaming=oldAndNewNames tawk '{print $1, $1;}' prevNames > $renaming if (( $(fastaNames newCogUk.filtered.fa | wc -l) > 0 )) ; then + # This grep failed 2021-03-14, crashing the script. I complained to COG-UK; tolerate. + set +o pipefail fastaNames newCogUk.filtered.fa \ | grep -Fwf - $cogUkDir/cog_metadata.csv \ | awk -F, '{print $1 "\t" $1 "|" $5;}' \ | sed -re 's/20([0-9][0-9])(-[0-9-]+)?$/\1\2/;' \ >> $renaming + set -o pipefail fi if (( $(fastaNames newGenBank.filtered.fa | wc -l) > 0 )) ; then fastaNames newGenBank.filtered.fa \ | sed -re 's/[ |].*//' \ | grep -Fwf - $ncbiDir/ncbi_dataset.plusBioSample.tsv \ | tawk '{ if ($3 == "") { $3 = "?"; } if ($6 != "") { print $1 "\t" $6 "|" $1 "|" $3; } else { print $1 "\t" $1 "|" $3; } }' \ | cleanGenbank \ | sed -re 's/ /_/g' \ | sed -re 's/20([0-9][0-9])(-[0-9-]+)?$/\1\2/;' \ >> $renaming fi wc -l $renaming