correct bug in rename files with extensions (dropseq tests passing) (30588308) · Commits · github_fork / Universc

launch_universc.sh

+67 −24

Original line number	Diff line number	Diff line
		@@ -1019,13 +1019,13 @@ for key in ${keys[@]}; do
		fi

		if [[ -h $read ]]; then
		path=`readlink -f $read`
		fullpath=`readlink -f $read`
		if [[ $verbose ]]; then
		echo "*Warning: file $read not in current directory. Path to the file captured instead.*"
		echo " (file) $read"
		echo " (path) $path"
		echo " (path) $fullpath"
		fi
		read=${path}
		read=${fullpath}
		fi
		case $read in
		#check if contains lane before read
		@@ -1073,15 +1073,58 @@ for key in ${keys[@]}; do
		echo " $read compatible with suffix"
		fi
		;;
		*)
		.)
		#rename file
		if [[ $verbose ]]; then
		echo "*Warning: file $read does not have suffix in its name. Suffix 001 is given.*"
		echo " renaming $read ..."
		fi
		rename -f "s/_${readkey}.\./_${readkey}_001\./" ${read}
		if [[ -f $(find $(dirname ${read}) -name $(basename ${read})'*.gz') ]]; then
		rename -f "s/_${readkey}.(.).gz/_${readkey}_001\.\$1.gz/g" ${read}gz
		# read=`echo $read \| sed -e "s/_${readkey}/_${readkey}_001/g"`
		fi
		if [[ ${read} == *.gz ]]; then
		rename -f "s/_${readkey}.(.*).gz/_${readkey}_001\.\$1.gz/g" ${read}
		# read=`echo $read \| sed -e "s/_${readkey}.(.*).gz/_${readkey}_001\.\$1.gz/g"`
		fi
		if [[ -f $(find $(dirname ${read}) -name $(basename ${read})'*.fastq') ]]; then
		rename -f "s/_${readkey}.(.)/_${readkey}_001\.\$1/" ${read}.fastq
		# read=`echo $read \| sed -e "s/_${readkey}(.*).fastq/_${readkey}_001.fastq/g"`
		fi
		if [[ -f $(find $(dirname ${read}) -name $(basename ${read})'*.fq') ]]; then
		rename -f "s/_${readkey}.(.)/_${readkey}_001\.\$1/" ${read}.fq
		# read=`echo $read \| sed -e "s/_${readkey}(.*).fq/_${readkey}_001.fq/g"`
		fi
		if [[ ${read} == *.fastq ]]; then
		rename -f "s/_${readkey}(.*).fastq/_${readkey}_001\.fastq/" ${read} ${read}.gz
		# read=`echo $read \| sed -e "s/_${readkey}*.fastq/_${readkey}_001.fastq/g"`
		fi
		if [[ ${read} == *.fq ]]; then
		rename -f "s/_${readkey}(.*).fq/_${readkey}_001\.fq/" ${read}
		# read=`echo $read \| sed -e "s/_${readkey}*.fq/_${readkey}_001\.fq/g"`
		fi
		#update file variable
		if [[ ${read} == .gz ]] \|\| [[ ${read} == .fastq ]] \|\| [[ ${read} == *.fq ]] \|\| [[ -f ${read} ]]; then
		#assumes read name already contains . in file extension
		read=`echo $read \| sed -e "s/_${readkey}.*\./_${readkey}_001\./g"`
		else
		#replace everything after read key (R1, R2, I1, I2) with 001 suffix (detects file later)
		rename -f "s/_${readkey}.*/_${readkey}_001/g" ${read}
		read=`echo $read \| sed -e "s/_${readkey}.*/_${readkey}_001/g"`
		fi
		#remove characters after read key (R1, R2, I1, I2) required as above
		if [[ ${read} != _${readkey}_001. ]] && [[ ${read} != . ]]; then
		rename -f "s/_${readkey}_*\./_${readkey}_001\./" ${read}
		read=`echo $read \| sed -e "s/_${readkey}.*\./_${readkey}_001\./g"`
		elif [[ ${read} != _${readkey}_001 ]] \|\| [[ ${read} != _${readkey}*00#1 ]]; then
		rename -f "s/_${readkey}*_001/_${readkey}_001/" ${read#}
		read=`echo $read \| sed -e "s/_${readkey}*_001/_${readkey}_001/g"`
		fi
		if [[ ${read} == _${readkey}__001.* ]]; then
		rename -f "s/_${readkey}.*_001\./_${readkey}_001\./" ${read}
		read=`echo $read \| sed -e "s/_${readkey}*_001\./_${readkey}_001\./g"`
		fi

		list[$j]=$read
		;;
		esac

test/run_tests_dropseq.sh

+17 −8

Original line number	Diff line number	Diff line
		@@ -27,29 +27,38 @@ if [[ -f test/shared/dropseq-test/*fastq.gz ]]; then
		fi
		# test manual setup
		bash launch_universc.sh -t "nadia" --setup

		# remove processed files
		if [[ -f SRR1873277_S1_L001_R[12]_001.fastq ]]; then
		rm SRR1873277_S1_L001_R[12]_001.fastq
		fi
		if [[ -d input4cellranger_test-dropseq ]]; then
		rm -rf input4cellranger_test-dropseq
		fi
		# call on dropseq with files
		bash launch_universc.sh --id "test-dropseq" --technology "nadia" \
		--reference "test/cellranger_reference/cellranger-tiny-ref/3.0.0" \
		--read1 "test/shared/dropseq-test/SRR1873277_Sample1_R1" \
		--read2 "test/shared/dropseq-test/SRR1873277_Sample1_R2"
		--read2 "test/shared/dropseq-test/SRR1873277_Sample1_R2" \
		--verbose

		#reset test data (files names)
		if [[ -f test/shared/dropseq-test/SRR1873277_S1_L001_R[12]_001.fastq ]]; then
		gzip test/shared/dropseq-test/SRR1873277_S1_L001_R[12]_001.fastq
		#reset test data (compress)
		if [[ -f test/shared/dropseq-test/SRR1873277_Sample1_S1_L001_R1_001.fastq ]]; then
		gzip -f test/shared/dropseq-test/SRR1873277_Sample1_S1_L001_R1_001.fastq
		fi
		if [[ -f test/shared/dropseq-test/SRR1873277_Sample1_S1_L001_R2_001.fastq ]]; then
		gzip -f test/shared/dropseq-test/SRR1873277_Sample1_S1_L001_R2_001.fastq
		fi
		#reset test data (files names)
		if [[ ! -f test/shared/dropseq-test/SRR1873277_Sample1_R1.fastq.gz ]]; then
		mv test/shared/dropseq-test/SRR1873277_S1_L001_R1_001.fastq.gz test/shared/dropseq-test/SRR1873277_Sample1_R1.fastq.gz
		mv test/shared/dropseq-test/SRR1873277_Sample1_S1_L001_R1_001.fastq.gz test/shared/dropseq-test/SRR1873277_Sample1_R1.fastq.gz
		else
		rm test/shared/dropseq-test/SRR1873277_S1_L001_R1_001.fastq.gz
		rm test/shared/dropseq-test/SRR1873277_Sample1_S1_L001_R1_001.fastq.gz
		fi
		if [[ ! -f test/shared/dropseq-test/SRR1873277_Sample1_R2.fastq.gz ]]; then
		mv test/shared/dropseq-test/SRR1873277_S1_L001_R2_001.fastq.gz test/shared/dropseq-test/SRR1873277_Sample1_R1.fastq.gz
		mv test/shared/dropseq-test/SRR1873277_Sample1_S1_L001_R2_001.fastq.gz test/shared/dropseq-test/SRR1873277_Sample1_R2.fastq.gz
		else
		rm test/shared/dropseq-test/SRR1873277_S1_L001_R2_001.fastq.gz
		rm test/shared/dropseq-test/SRR1873277_Sample1_S1_L001_R2_001.fastq.gz
		fi

		# compress all input files

Admin message