def __call__(self, fobj):
for (chunk1, chunk2) in zip_longest(self.filehash,
hash_chunks(fobj.path),
fillvalue=''):
if chunk1 != chunk2:
return False
return True
def group_by_hash(fsobjects):
hashes = {}
for fobj in fsobjects:
chunks = hash_chunks(fobj.path)
chunk = next(chunks)
while chunk in hashes:
for dup in hashes[chunk]:
_, dup_chunks = dup
try:
hashes[next(dup_chunks)] = [dup]
hashes[chunk].remove(dup)
except StopIteration:
pass
try:
chunk = next(chunks)
except StopIteration:
hashes[chunk].append((fobj, chunks))
break
else:
hashes[chunk] = [(fobj, chunks)]
groups = []
for dups in hashes.values():
group = []
for (dup, _) in dups:
group.append(dup)
if group:
groups.append(group)
return groups
def __init__(self, filepath=None):
if filepath is None:
self.filepath = self.fm.thisfile.path
else:
self.filepath = filepath
if self.filepath is None:
self.fm.notify("Error: No file selected for hashing!", bad=True)
# TODO: Lazily generated list would be more efficient, a generator
# isn't enough because this object is reused for every fsobject
# in the current directory.
self.filehash = list(hash_chunks(abspath(self.filepath)))