#!/usr/local2/bin/python2.3
import string, sys
from openeye.oechem import *

##### Looks for identical molecules - and removes them (time consuming, but safe -> Paranoia to loose sth

try:
    molf = open(sys.argv[1], 'r')
    outf1 = open('a4'+sys.argv[1][string.find(sys.argv[1], '_'):string.find(sys.argv[1], '_out')]+'_out.ism', 'w')
    outf2 = open('a4'+sys.argv[1][string.find(sys.argv[1], '_'):string.find(sys.argv[1], '_out')]+'_doubles.ism', 'w')
    log = open('a4'+sys.argv[1][string.find(sys.argv[1], '_'):string.find(sys.argv[1], '_out')]+'.log', 'w')
    m1 = open(sys.argv[1][:string.find(sys.argv[1], '_out')]+'_metals_keep.txt', 'r')
    m2 = open(sys.argv[1][:string.find(sys.argv[1], '_out')]+'_metals.txt', 'r')
except IndexError:
    print '\nUsage: a4_rm_doubles.py molfile\n'
    raise SystemExit()

ism_mol_dir = {}

# Metal lists generated by a3.2_size_cutoff_filter.py

metal_list = []
for line in m2.readlines():
    metal_list.append(line[:-1])
metal_list_keep = []
for line in m1.readlines():
    metal_list_keep.append(line[:-1])

for line in molf.readlines():
    if string.split(line)[1] not in metal_list_keep and string.split(line)[1] in metal_list:
        continue
    else:
        if ism_mol_dir.has_key(string.split(line)[0]):
            outf2.write(line) 
            log.write(string.split(line)[1]+'\tidentical with '+ism_mol_dir[string.split(line)[0]]+'\n')
        else:
            ism_mol_dir[string.split(line)[0]] = string.split(line)[1]
            outf1.write(line)

log.close()        
outf1.close()
outf2.close()
molf.close()