import os, hashlib, sys

directories = os.listdir('.') ## Get the files in the current directory

files = []

def quickhash(filefullpath):
    try:
        with open(filefullpath, "rb" ) as f:
            md5 = hashlib.md5()
            block_size=1024*100
            data = f.read(block_size)
            md5.update(data)
        return md5.digest()
    except OSError as er:
        print(f"Error hashing {filefullpath}")
        return None

def direcCrawler():
    global directories, files
    for item in directories:
        if os.path.isdir(item):
            newdirItems = []
            newdir = os.listdir(item)
            for nwitem in newdir:
                directories.append(f'{item}/{nwitem}')

        elif os.path.isfile(item):
            files.append(item)
        directories.remove(item) # Its been crawled
    return True

def calchashes():
    global files, filehashes
    filehashes = {}
    for file in files:
        hash = quickhash(file)
        if hash == None: continue
        ## Inefficient to use a catch on every non-duplicate (most of the files)
        try:  filehashes[hash].append(file)
        except: filehashes[hash] = [file]
    return filehashes

def returnduplicates(filehashes:dict):
    duplicates = []
    for hlistkey in filehashes.keys():
        if len(filehashes[hlistkey]) > 1:
            duplicates.append(filehashes[hlistkey])
    return duplicates

            

def main():
    global files, directories
    # Crawl the directories (TD add a timeout here later)
    print("Collecting files & paths")
    while len(directories) > 0:
        direcCrawler()
    print(f'found {len(files)} files')
    # Calc hashes
    print("Calculating file hashes")
    filehashes = calchashes()

    duplicates = returnduplicates(filehashes)
    print(f"Found {len(duplicates)} duplicates")
    for dupe in duplicates:
        print(dupe)



main()


# quickhash('./ubuntu-22.04.2-desktop-amd64.iso')