Mercurial > repos > r-lannes > cut_include_exclude

comparison cut_fr.py @ 0:93f267c26057 draft

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
Uploaded
author r-lannes
date Wed, 16 Dec 2015 08:45:06 -0500
parents
children
comparison
equal deleted inserted replaced
-1:000000000000 0:93f267c26057
1 #!usr/bin/pyhton3
2 '''<cut_fr.py is script allwing more convenient use of cut in galaxy environment>
3 Copyright (C) <2015> <Lannes Romain>
4
5 This program is free software: you can redistribute it and/or modify
6 it under the terms of the GNU General Public License as published by
7 the Free Software Foundation, either version 3 of the License, or
8 (at your option) any later version.
9
10 This program is distributed in the hope that it will be useful,
11 but WITHOUT ANY WARRANTY; without even the implied warranty of
12 MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
13 GNU General Public License for more details.
14
15 You should have received a copy of the GNU General Public License
16 along with this program. If not, see <http://www.gnu.org/licenses/>.'''
17
18 #-------- import
19 import os
20 import argparse
21 import sys
22
23 #------- argument
24 parser = argparse.ArgumentParser(description = "renvoie le fichier inputs moins les collones sellectionner")
25 parser.add_argument('--c', help = "collumn to be removed" )
26 parser.add_argument('--i', help = "input datasets" )
27 parser.add_argument('--type_cut', help = "type of cut" )
28 parser.add_argument('--extend_print', help = "add sdtrout print" )
29 parser.add_argument('--o', help = "output datasets" )
30
31 args = parser.parse_args()
32
33
34 if(args.type_cut == "exclude"):
35 #------- First step count number of column
36 dataset_initial = open(args.i,'r')
37 firstline_dataset = dataset_initial.readline()
38
39 column_header = firstline_dataset.split('\t')
40 number_column = len(column_header)
41
42 # Collumn are in foramt c1,c2,c3
43 # need to get them
44 column_list = args.c
45 column_list = column_list.split(',')
46
47 # remove the letter c
48 # there is certainly a more convenient way to do it
49 for i in range(len(column_list)):
50 column_list[i] = int(column_list[i][-1])
51
52 # Second Step prepare the command to be printed
53 str_argument = "-f"
54 for i in range( 1,number_column +1 ,1 ):
55 if i not in column_list:
56 str_argument = str_argument + str(i) + ','
57 str_argument = str_argument.strip(',')
58
59 # Create the command to be launch
60 str_cut = "cut " + str_argument + " " + args.i +" >"+ args.o
61
62 if(args.type_cut == "include"):
63 # Collumn are in foramt c1,c2,c3
64 # need to get them
65 column_list = args.c
66 column_list = column_list.split(',')
67 # remove the letter c
68 # there is certainly a more convenient way to do it
69 for i in range(len(column_list)):
70 column_list[i] = int(column_list[i][-1])
71 # Second Step prepare the command to be printed
72 str_argument = "-f"
73 for indice in column_list :
74 str_argument = str_argument + str(indice) + ','
75 str_argument = str_argument.strip(',')
76 # Create the command to be launch
77 str_cut = "cut " + str_argument + " " + args.i +" >"+ args.o
78
79 # launch the command
80 result = os.system( str_cut )
81 if (args.extend_print == "True"):
82 sys.stdout.write("input dataset : " + args.i + '\n')
83 sys.stdout.write("column selected : " + args.c + '\n')
84 sys.stdout.write("cut_type : " + args.type_cut+'\n')
85 sys.stdout.write("output dataset : " + args.o+'\n')
86 sys.stdout.write("command launch : " + str_cut+'\n' )
87
88
89
90
91
92
93