def walk(self, top, topdown=True, ignore_file_handler=None):
"""Directory tree generator.
See `os.walk` for the docs. Differences:
- no support for symlinks
- it could raise exceptions, there is no onerror argument
"""
def onerror(e):
raise e
for root, dirs, files in dvc_walk(
top,
topdown=topdown,
onerror=onerror,
ignore_file_handler=ignore_file_handler,
):
yield os.path.normpath(root), dirs, files
def get_mtime_and_size(path):
size = os.path.getsize(path)
mtime = os.path.getmtime(path)
if os.path.isdir(path):
for root, dirs, files in dvc_walk(str(path)):
for name in dirs + files:
entry = os.path.join(root, name)
try:
stat = os.stat(entry)
except OSError as exc:
# NOTE: broken symlink case.
if exc.errno != errno.ENOENT:
raise
continue
size += stat.st_size
entry_mtime = stat.st_mtime
if entry_mtime > mtime:
mtime = entry_mtime
# State of files handled by dvc is stored in db as TEXT.
# We cast results to string for later comparisons with stored values.
return str(int(nanotime.timestamp(mtime))), str(size)
def walk(self, path_info):
return dvc_walk(path_info, self.repo.dvcignore)
def __init__(self, root_dir):
self.ignores = {DvcIgnoreDirs([".git", ".hg", ".dvc"])}
self._update(root_dir)
for root, dirs, _ in dvc_walk(root_dir, self):
for d in dirs:
self._update(os.path.join(root, d))
