diff msp_split.py @ 8:8de790489301 draft

planemo upload for repository https://github.com/computational-metabolomics/dma-tools-galaxy commit af689d3f20c86f69aa824545e668280bcd5e0cca

--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/msp_split.py	Mon May 14 08:15:07 2018 -0400
@@ -0,0 +1,73 @@
+from __future__ import print_function
+import argparse
+import textwrap
+import os
+import re
+import csv
+import math
+
+def msp_split(i, o, n):
+    spec_total = lcount('NAME', i)
+    spec_lim = math.ceil(spec_total/float(n))
+    spec_c = 0
+    filelist = []
+    header = ''
+    print('spec_lim', spec_lim)
+    with open(i, 'r') as msp_in:
+        for i in range(1, n+1):
+            with open(os.path.join(o, 'file{}.msp'.format(str(i).zfill(len(str(n))))), 'w+') as msp_out:
+                while spec_c <= spec_lim:
+                    if header:
+                        msp_out.write(header)
+                        header = ''
+                    line = msp_in.readline()
+
+                    if not line:
+                        break  # end of file
+
+                    if re.match('^NAME:.*$', line, re.IGNORECASE):
+                        header = line
+                        spec_c += 1
+                    else:
+                        msp_out.write(line)
+                spec_c = 1 
+
+    return filelist
+
+def lcount(keyword, fname):
+    with open(fname, 'r') as fin:
+        return sum([1 for line in fin if keyword in line])
+
+def main():
+
+    p = argparse.ArgumentParser(prog='PROG',
+                                formatter_class=argparse.RawDescriptionHelpFormatter,
+                                description='''Create filelist for DMA DIMS nearline workflow''',
+                                epilog=textwrap.dedent('''
+                            -------------------------------------------------------------------------
+
+                            Example Usage
+
+                            python dma-filelist-generation.py -i [dir with sample files], [dir with blank files] -o .
+
+                            '''))
+
+    p.add_argument('-i', dest='i', help='dir with sample files', required=True)
+    p.add_argument('-o', dest='o', help='out dir', required=True)
+    p.add_argument('-n', dest='n',)
+
+
+    args = p.parse_args() 
+
+    if not os.path.exists(args.o):
+        os.makedirs(args.o)
+    print('in file', args.i)
+    print('out dir', args.o)
+    print('nm files', args.n)
+
+    msp_split(args.i, args.o, int(args.n))
+
+
+if __name__ == '__main__':
+    main()
+