diff data_manager/install_primer_scheme_bedfiles.py @ 2:b3710e492ee4 draft

"planemo upload for repository https://github.com/galaxyproject/tools-iuc/tree/master/data_managers/data_manager_primer_scheme_bedfiles commit 61b727022bd1b24a8843830a9c5f34ef66835694"

--- a/data_manager/install_primer_scheme_bedfiles.py	Fri Apr 24 16:10:30 2020 -0400
+++ b/data_manager/install_primer_scheme_bedfiles.py	Mon Jun 22 18:26:10 2020 -0400
@@ -1,20 +1,39 @@
 #!/usr/bin/env python
 
-from __future__ import print_function, division
+from __future__ import division, print_function
 
 import argparse
+try:
+    from io import StringIO
+except ImportError:
+    from StringIO import StringIO
 import json
 import os
 import os.path
 import re
 import sys
-import shutil
 
 import requests
 
 DATA_TABLE_NAME = "primer_scheme_bedfiles"
 
 
+def write_good_bed(input_file, bed_output_filename):
+    with open(bed_output_filename, "w") as bed_output_file:
+        for line in input_file:
+            fields = line.split("\t")
+            if len(fields) < 6:
+                # too short to encode the strand format
+                exit("invalid format in BED file: {}".format(line.rstrip()))
+            try:
+                # try and parse field 5 as a number
+                float(fields[4])
+            except ValueError:
+                # ARTIC with broken BED, set field 5 to 60
+                fields[4] = "60"
+            bed_output_file.write("\t".join(fields))
+
+
 def fetch_artic_primers(output_directory, primers):
     primer_sets = {
         "SARS-CoV-2-ARTICv1": "https://raw.githubusercontent.com/artic-network/artic-ncov2019/master/primer_schemes/nCoV-2019/V1/nCoV-2019.bed",
@@ -37,7 +56,7 @@
             )
             exit(response.status_code)
         bed_output_filename = os.path.join(output_directory, name + ".bed")
-        open(bed_output_filename, "w").write(response.text)
+        write_good_bed(StringIO(response.text), bed_output_filename)
         description = name[:-2] + " " + name[-2:] + " primer set"
         data.append(dict(value=name, path=bed_output_filename, description=description))
     return data
@@ -48,7 +67,8 @@
 ):
     name = re.sub(r"\W", "", str(primer_name).replace(" ", "_"))
     output_filename = os.path.join(output_directory, name + ".bed")
-    shutil.copyfile(input_filename, output_filename)
+    with open(input_filename) as input_file:
+        write_good_bed(input_file, output_filename)
     data = [dict(value=name, description=primer_description, path=output_filename)]
     return data