#!/usr/bin/python
import string, os, sys
from openeye.oechem import *
from var_functions import start_and_end_with
##### Looks for identical molecules - and removes them (time consuming, but safe -> Paranoia to loose sth

#**************************************************************************
def check_4_doubles(smi_dir, infile, log, ofs):
    ifs.open(infile)
    for mol in ifs.GetOEMols():
        ism = OECreateIsoSmiString(mol)
        if smi_dir.has_key(ism):
            log.write(mol.GetTitle()+' is equal to '+ism_mol_dir[ism]+'\n')
        else:
            ism_mol_dir[ism] = mol.GetTitle()
            OEWriteMolecule(ofs_1, mol)
    ifs.close()

    return ism_mol_dir
#**************************************************************************

try:
    start = sys.argv[1]
    end = sys.argv[2]
except IndexError:
    print '\nUsage: b9_remove_doubles.py start-file-pattern end-file-pattern\n'
    raise SystemExit()

ifs = oemolistream()
log = open('b9_doubles.log', 'a')

ofs_1 = oemolostream()
ofs_1.open('b9_'+end[:-4]+'.ism')

ism_mol_dir = {}

for file in os.listdir('.'):
    if start_and_end_with(file, start, end):
        ism_mol_dir = check_4_doubles(ism_mol_dir, file, log, ofs_1)
    else:
        continue

log.close()        
ofs_1.close()