mirrors/ArchiveBox
1
0
Fork 0
mirror of https://github.com/ArchiveBox/ArchiveBox.git synced 2025-05-13 22:54:27 -04:00
Code Issues Projects Releases Packages Wiki Activity Actions

add functions to parse link details jsons and list+apply migrations

Browse source
This commit is contained in:
Nick Sweeting 2019-04-24 04:07:46 -04:00
parent 0b27f33d2e
commit 0f2497a2a6
2 changed files with 41 additions and 1 deletions
Download patch file Download diff file Expand all files Collapse all files

10
archivebox/legacy/storage/json.py
View file

@ -15,6 +15,7 @@ from ..config import (
GIT_SHA,
DEPENDENCIES,
JSON_INDEX_FILENAME,
ARCHIVE_DIR_NAME,
)
from ..util import (
enforce_types,
@ -98,3 +99,12 @@ def parse_json_link_details(out_dir: str) -> Optional[Link]:
link_json = json.load(f)
return Link.from_json(link_json)
return None
@enforce_types
def parse_json_links_details(out_dir: str) -> Iterator[Link]:
"""read through all the archive data folders and return the parsed links"""
for entry in os.scandir(os.path.join(out_dir, ARCHIVE_DIR_NAME)):
if entry.is_dir(follow_symlinks=True):
if os.path.exists(os.path.join(entry.path, 'index.json')):
yield parse_json_link_details(entry.path)