From f4c3da88c1233802fea46cc972a81dc3b5b51185 Mon Sep 17 00:00:00 2001 From: Peter Amstutz Date: Tue, 21 Apr 2020 15:37:58 -0400 Subject: Work around CWL content size limit by chunking Arvados-DCO-1.1-Signed-off-by: Peter Amstutz --- workflows/pangenome-generate/relabel-seqs.py | 22 ++++++++++++++++------ 1 file changed, 16 insertions(+), 6 deletions(-) (limited to 'workflows/pangenome-generate/relabel-seqs.py') diff --git a/workflows/pangenome-generate/relabel-seqs.py b/workflows/pangenome-generate/relabel-seqs.py index 1188ceb..970540f 100644 --- a/workflows/pangenome-generate/relabel-seqs.py +++ b/workflows/pangenome-generate/relabel-seqs.py @@ -1,5 +1,15 @@ -reads = $(inputs.readsFA) -subjects = $(inputs.subjects) +import os +import json + +reads = [] +b = 1 +while os.path.exists("block%i" % b): + with open("block%i" % b) as f: + reads.extend(json.load(f)) + b += 1 + +with open("subjects") as f: + subjects = json.load(f) relabeled_fasta = open("relabeledSeqs.fasta", "wt") original_labels = open("originalLabels.ttl", "wt") @@ -7,12 +17,12 @@ original_labels = open("originalLabels.ttl", "wt") for i, r in enumerate(reads): with open(r["path"], "rt") as fa: label = fa.readline() - original_labels.write("<%s> \\"%s\\" .\\n" % (subjects[i], label[1:].strip().replace('"', '\\\\"'))) - relabeled_fasta.write(">"+subjects[i]+"\\n") + original_labels.write("<%s> \"%s\" .\n" % (subjects[i], label[1:].strip().replace('"', '\\"'))) + relabeled_fasta.write(">"+subjects[i]+"\n") data = fa.read(8096) while data: relabeled_fasta.write(data) - endswithnewline = data.endswith("\\n") + endswithnewline = data.endswith("\n") data = fa.read(8096) if not endswithnewline: - relabeled_fasta.write("\\n") + relabeled_fasta.write("\n") -- cgit v1.2.3