diff column_remove_by_header.py @ 0:372967836e98 draft

planemo upload for repository https://github.com/galaxyproject/tools-iuc/tree/master/tools/column_remove_by_header commit 2150a3264364471090b650bdffde9f9c0b47ac39
author iuc
date Wed, 12 Apr 2017 17:17:29 -0400
parents
children 2040e4c2750a
line wrap: on
line diff
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/column_remove_by_header.py	Wed Apr 12 17:17:29 2017 -0400
@@ -0,0 +1,35 @@
+#!/usr/bin/env python
+
+import subprocess
+import sys
+
+AWK_CMD = """BEGIN{FS="%s"; OFS="%s";} {print %s;}"""
+
+input_filename = sys.argv[1]
+output_filename = sys.argv[2]
+delimiter = sys.argv[3]
+keep_columns = sys.argv[4]
+strip_characters = sys.argv[5]
+
+if keep_columns == "--keep":
+    keep_columns = True
+else:
+    keep_columns = False
+
+names = []
+for name in sys.argv[6:]:
+    names.append( name )
+
+header = None
+with open( input_filename, 'r' ) as fh:
+    header = fh.readline().strip( '\r\n' )
+header = header.split( delimiter )
+columns = []
+for i, key in enumerate( header, 1 ):
+    if i == 1 and strip_characters:
+        key = key.lstrip( strip_characters )
+    if ( keep_columns and key in names ) or ( not keep_columns and key not in names ):
+        columns.append( i )
+print( "Kept", len( columns ), "of", len( header ), "columns." )
+awk_cmd = AWK_CMD % ( delimiter, delimiter, ",".join( map( lambda x: "$%s" % x, columns ) ) )
+sys.exit( subprocess.call( [ 'gawk', awk_cmd, input_filename ], stdout=open( output_filename, 'wb+' ), shell=False ) )