def test_recurse_files(tmp_path):
"""
recurse_files() reads the files in a directory and its subdirectories.
"""
with tempfiles(tmp_path) as path:
chnl = file.recurse_files(path.parent)
assert isinstance(chnl, Channel)
files = set(chnl)
assert path in files
# Starting point is not included.
assert path.parent not in files
assert set(file.recurse_files(path)) == set()
all_files = set(file.recurse_files(tmp_path))
# Test that it really is just files, not directories
assert path.parent not in all_files
assert all(f.is_file() for f in all_files if f.name != 'deep')
assert all(file.longname(f).is_file() for f in all_files)
# Test that filenames starting . are included.
assert tmp_path.joinpath('.hidden') in all_files
# Test that long paths are traversed
assert any(f.name == 'deep' for f in all_files), all_files # pragma: no branch
# A containment check without set() could lead to abandoning the
# iterator. We'd need more files to confirm that really has caused an
# early exit in the worker thread, and anyway the Channel tests look into
# that, so let's just make sure the proper code works.
with file.recurse_files(tmp_path).cancel_context() as chnl:
assert path in chnl
def test_string_param(tmp_path):
"""
recurse_files() and recurse_filestats() accept a string argument.
"""
with tempfiles(tmp_path):
assert set(file.recurse_files(tmp_path)) == set(file.recurse_files(str(tmp_path)))
assert set(file.recurse_filestats(tmp_path)) == set(file.recurse_filestats(str(tmp_path)))
def test_channel_param(tmp_path):
"""
recurse_files() and recurse_filestats() accept a channel parameter.
"""
with tempfiles(tmp_path):
assert set(file.recurse_files(tmp_path)) == set(file.recurse_files(tmp_path, channel=QueueChannel()))
assert set(file.recurse_filestats(tmp_path)) == set(file.recurse_filestats(tmp_path, channel=QueueChannel()))
def test_hasher(count, tmp_hashfiles):
"""
A hasher is a crew of threads that computes the hashes of the contents
of the files we pass to it.
"""
requests, responses = file.hasher(count)
filenames = tuple(file.recurse_files(tmp_hashfiles))
assert len(filenames) == 20
requests.put_many(filenames).end()
results = tuple(responses)
assert len(results) == len(filenames)
assert set(result[0] for result in results) == set(filenames)
for filename, digest in results:
assert binascii.hexlify(digest).decode('ascii') == filename.name
def test_recurse_filestats(tmp_path):
"""
recurse_filestats() is like recurse_files(), but returns more data than
just the path of each file it finds.
"""
with tempfiles(tmp_path, file_size=27):
chnl = file.recurse_filestats(tmp_path)
assert isinstance(chnl, Channel)
all_files = set(chnl)
# Check the same inclusion/exclusion as recurse_files()
assert set(file.fullpath for file in all_files) == set(file.recurse_files(tmp_path))
nonempty = [file for file in all_files if file.fullpath == tmp_path / '.hidden']
assert len(nonempty) == 1
nonempty = nonempty[0]
# Check we have all the stats we want
assert nonempty.fullpath == tmp_path / '.hidden'
assert nonempty.basepath == tmp_path
assert nonempty.relpath == Path('.hidden')
assert nonempty.size == 27
assert all(f.size == 27 for f in all_files), [(f, f.size) for f in all_files if f.size != 27]