import sys
import os
import hashlib
import time
import shutil

def chunk_reader(fobj, chunk_size=1024):
    while True:
        chunk = fobj.read(chunk_size)
        if not chunk:
            return
        yield chunk
        
dub_list=[]
mv_list_src=[]
mv_list_dst=[]

def check_for_duplicates(paths, hash=hashlib.sha1):
    hashes = {}
    dub_list=[]
    mv_list=[]
    for path in paths:
        if not os.path.exists(path+'\duplicate'):
            os.mkdir(path+'\duplicate')
        for dirpath, dirnames, filenames in os.walk(path):
            for filename in filenames:
                full_path = os.path.join(dirpath, filename)
                mtime=time.strftime("Ostatnia modyfikacja: %b/%Y", time.localtime(os.path.getmtime(full_path)))
                mtime_path=time.strftime("\\%Y\\%m", time.gmtime(os.path.getmtime(full_path)))
                print("File:",filename,mtime)
                dst_dir=dirpath+mtime_path
                hashobj = hash()
                for chunk in chunk_reader(open(full_path, 'rb')):
                    hashobj.update(chunk)
                file_id = (hashobj.digest(), os.path.getsize(full_path))
                duplicate = hashes.get(file_id, None)
                if duplicate:
                    print("Znalezione duplikaty: %s and %s" % (full_path, duplicate))
                    if os.path.getmtime(full_path) < os.path.getmtime(duplicate):
                        dub_list.append(full_path)
                    else:
                        if not os.path.exists(dst_dir):
                            os.makedirs(dst_dir)
                        mv_list_src.append(full_path)
                        mv_list_dst.append(dst_dir)
                        
                else:
                    hashes[file_id] = full_path
                    if not os.path.exists(dst_dir):
                        os.makedirs(dst_dir)
                    mv_list_src.append(full_path)
                    mv_list_dst.append(dst_dir)
    for x in dub_list:
        shutil.move(x,path+'\duplicate\\')
    for src,dst in zip(mv_list_src, mv_list_dst):
        shutil.move(src,dst)

if sys.argv[1:]:
    check_for_duplicates(sys.argv[1:])
else:
    print("Podaj sciezke jako argument")