def _copy_if_git_file(self, to_path):
src_path = self.def_path
repo_dir = cached_clone(**self.def_repo)
if not self._is_git_file(repo_dir, src_path):
return False
src_full_path = os.path.join(repo_dir, src_path)
dst_full_path = os.path.abspath(to_path)
fs_copy(src_full_path, dst_full_path)
return True
def _copy_if_git_file(self, to_path):
src_path = self.def_path
repo_dir = cached_clone(**self.def_repo)
if not self._is_git_file(repo_dir, src_path):
return False
src_full_path = os.path.join(repo_dir, src_path)
dst_full_path = os.path.abspath(to_path)
fs_copy(src_full_path, dst_full_path)
self.def_repo[self.PARAM_REV_LOCK] = SCM(repo_dir).get_rev()
return True
def get(url, path, out=None, rev=None):
out = resolve_output(path, out)
if Stage.is_valid_filename(out):
raise GetDVCFileError()
# Creating a directory right beside the output to make sure that they
# are on the same filesystem, so we could take the advantage of
# reflink and/or hardlink. Not using tempfile.TemporaryDirectory
# because it will create a symlink to tmpfs, which defeats the purpose
# and won't work with reflink/hardlink.
dpath = os.path.dirname(os.path.abspath(out))
tmp_dir = os.path.join(dpath, "." + str(shortuuid.uuid()))
try:
try:
with external_repo(cache_dir=tmp_dir, url=url, rev=rev) as repo:
# Try any links possible to avoid data duplication.
#
# Not using symlink, because we need to remove cache after we
# are done, and to make that work we would have to copy data
# over anyway before removing the cache, so we might just copy
# it right away.
#
# Also, we can't use theoretical "move" link type here, because
# the same cache file might be used a few times in a directory.
repo.cache.local.cache_types = ["reflink", "hardlink", "copy"]
output = repo.find_out_by_relpath(path)
if output.use_cache:
_get_cached(repo, output, out)
return
# Non-cached output, fall through and try to copy from git.
except (NotDvcRepoError, NoOutputInExternalRepoError):
# Not a DVC repository or, possibly, path is not tracked by DVC.
# Fall through and try to copy from git.
pass
if os.path.isabs(path):
raise FileNotFoundError
repo_dir = cached_clone(url, rev=rev)
fs_copy(os.path.join(repo_dir, path), out)
except (OutputNotFoundError, FileNotFoundError):
raise PathMissingError(path, url)
finally:
remove(tmp_dir)
def _get_checksum(self, updated=False):
rev_lock = None
if not updated:
rev_lock = self.def_repo.get(self.PARAM_REV_LOCK)
try:
with self._make_repo(rev_lock=rev_lock) as repo:
return repo.find_out_by_relpath(self.def_path).info["md5"]
except (NotDvcRepoError, NoOutputInExternalRepoError):
# Fall through and clone
pass
repo_path = cached_clone(
self.def_repo[self.PARAM_URL],
rev=rev_lock or self.def_repo.get(self.PARAM_REV),
)
path = PathInfo(os.path.join(repo_path, self.def_path))
return self.repo.cache.local.get_checksum(path)