def _process(dest_tree, src_tree, collected_files, download=False):
from dvc.cache.local import _log_exceptions
from_infos = []
to_infos = []
names = []
for from_info in collected_files:
from_infos.append(from_info)
fname = from_info.relative_to(src_tree.path_info)
names.append(str(fname))
to_infos.append(dest_tree.path_info / fname)
total = len(from_infos)
if download:
func = partial(
_log_exceptions(src_tree.download, "download"),
dir_mode=dest_tree.dir_mode,
file_mode=dest_tree.file_mode,
)
desc = "Downloading"
else:
func = partial(_log_exceptions(dest_tree.upload, "upload"))
desc = "Uploading"
with Tqdm(total=total, unit="file", desc=desc) as pbar:
func = pbar.wrap_fn(func)
# TODO: parallelize this, currently --jobs for repro applies to
# number of repro executors not download threads
with ThreadPoolExecutor(max_workers=1) as dl_executor:
fails = sum(dl_executor.map(func, from_infos, to_infos, names))
if fails:
if download:
raise DownloadError(fails)
raise UploadError(fails)
def pull(self, remote):
remote = self.repo.cloud.get_remote(remote)
return self._transfer(
_log_exceptions(remote.tree.download, "download"),
remote.tree,
self.repo.cache.local.tree,
)
