import sys
import os, os.path

EXTENSIONS = [".cs"]
MIN_MATCH = 75

class HashedFile(object):
    def __init__(self, path):
        self.path = path
        f = open(path)
        self.lineHashes = [hash(l) for l in f.xreadlines()]
        f.close()

class Block(object):
    def __init__(self, path1, start1, path2, start2, data=""):
        self.path1 = path1
        self.start1 = start1
        self.path2 = path2
        self.start2 = start2
        self.data = data
        
def get_filenames(path):
    for root, dirs, files in os.walk(path):
        for filename in files:
            name, ext = os.path.splitext(filename)
            if ext in EXTENSIONS:
                yield os.path.join(root, filename)
    
def hash(text):
    import md5
    m = md5.new()
    m.update(text)
    return m.digest()
    
def compare(hashedFile1, hashedFile2):
    import difflib
    sm = difflib.SequenceMatcher(None, hashedFile1.lineHashes, hashedFile2.lineHashes)
    for (start1, start2, lineCount) in sm.get_matching_blocks():
        if lineCount >= MIN_MATCH:
            yield Block(hashedFile1.path, start1+1, hashedFile2.path, start2+1)

if __name__ == "__main__":
    # generate hashes for all lines in all files
    hashedFiles = []
    for filename in get_filenames("."):
        sys.stderr.write(".")
        hashedFile = HashedFile(filename)
        hashedFiles.append(hashedFile)

    print >>sys.stderr
    
    # compare all files to each other using hashed lines
    for i in xrange(len(hashedFiles)):
        print >>sys.stderr, hashedFiles[i].path, "..."
        
        for j in xrange(i+1, len(hashedFiles)):
            for block in compare(hashedFiles[i], hashedFiles[j]):
                print "file 1: %s:%d" % (block.path1, block.start1)
                print "file 2: %s:%d" % (block.path2, block.start2)