From d675d96d81bfdec8e8e54719e1c7216860b1906e Mon Sep 17 00:00:00 2001
From: replikation <christian.jena@gmail.com>
Date: Wed, 19 Aug 2020 00:01:48 +0200
Subject: [PATCH] way better samtools error fail save

---
 modules/samtools.nf | 56 +++++++++++++++++++++++++++++++--------------
 1 file changed, 39 insertions(+), 17 deletions(-)

diff --git a/modules/samtools.nf b/modules/samtools.nf
index 2ee9a6a..12adfa2 100644
--- a/modules/samtools.nf
+++ b/modules/samtools.nf
@@ -1,29 +1,51 @@
 process samtools {
-      publishDir "${params.output}/${name}/phage_positive_contigs", mode: 'copy', pattern: "${name}_positive_contigs.fa"
-      label 'samtools'
+        publishDir "${params.output}/${name}/phage_positive_contigs", mode: 'copy', pattern: "${name}_positive_contigs.fa"
+        label 'samtools'
     input:
-      tuple val(name), file(file), file(list)
+        tuple val(name), file(file), file(list)
     output:
-      tuple val(name), file("${name}_positive_contigs.fa")
+        tuple val(name), file("${name}_positive_contigs.fa")
     script:
-      """
-      cat ${list} | sort | uniq > tmp_allctgs.txt
-      cat ${file} > all.fasta
-      xargs samtools faidx all.fasta < tmp_allctgs.txt > ${name}_positive_contigs.fa
-      """
+        """
+        cat ${list} | sort | uniq > tmp_allctgs.txt
+        cat ${file} > all.fasta
+
+        # get samtools but ignore "samtool fails"
+        while read fastaheader; do    
+            if grep -qw ">\$fastaheader" all.fasta; then
+                samtools faidx all.fasta \$fastaheader >> ${name}_positive_contigs.fa
+            else
+                echo "\$fastaheader not found" >> error_code_samtools.txt
+            fi
+        done < tmp_allctgs.txt
+
+        # old command
+        # xargs samtools faidx all.fasta < tmp_allctgs.txt > ${name}_positive_contigs.fa
+        """
 }
 
 process samtools_fastq {
-      publishDir "${params.output}/${name}/phage_positive_contigs", mode: 'copy', pattern: "${name}_positive_contigs.fa"
-      label 'samtools'
+        publishDir "${params.output}/${name}/phage_positive_contigs", mode: 'copy', pattern: "${name}_positive_contigs.fa"
+        label 'samtools'
     input:
-      tuple val(name), file(file), file(list)
+        tuple val(name), file(file), file(list)
     output:
-      tuple val(name), file("${name}_positive_contigs.fa")
+        tuple val(name), file("${name}_positive_contigs.fa")
     script:
-      """
-      cat ${list} | sort | uniq > tmp_allctgs.txt
-      cat ${file} > all.fasta
-      xargs samtools faidx all.fasta < tmp_allctgs.txt > ${name}_positive_contigs.fa
+        """
+        cat ${list} | sort | uniq > tmp_allctgs.txt
+        cat ${file} > all.fasta
+        
+        # get samtools but ignore "samtool fails"
+        while read fastaheader; do    
+            if grep -qw ">\$fastaheader" all.fasta; then
+                samtools faidx all.fasta \$fastaheader >> ${name}_positive_contigs.fa
+            else
+                echo "\$fastaheader not found" >> error_code_samtools.txt
+            fi
+        done < tmp_allctgs.txt
+
+        # old command
+        # xargs samtools faidx all.fasta < tmp_allctgs.txt > ${name}_positive_contigs.fa
       """
 }
\ No newline at end of file