Skip to content

Commit f43a316

Browse files
committed
feat(datasets): load dataset only when necessary to reduce blocking time
1 parent f6b4a44 commit f43a316

1 file changed

Lines changed: 8 additions & 3 deletions

File tree

codesectools/datasets/core/dataset.py

Lines changed: 8 additions & 3 deletions
Original file line numberDiff line numberDiff line change
@@ -57,12 +57,17 @@ def __init__(self, lang: str | None = None) -> None:
5757
"""
5858
self.directory = USER_CACHE_DIR / self.name
5959
self.lang = lang
60+
self._files = []
6061
if self.lang:
6162
self.full_name = f"{self.name}_{self.lang}"
6263
assert self.full_name in self.list_dataset_full_names()
63-
self.files: list[File] = self.load_dataset()
64-
else:
65-
self.files = []
64+
65+
@property
66+
def files(self) -> list:
67+
"""Get the list of dataset files, loading them if necessary."""
68+
if self.lang:
69+
self._files = self.load_dataset()
70+
return self._files
6671

6772
@classmethod
6873
def is_cached(cls) -> bool:

0 commit comments

Comments
 (0)