alex/ArchiveBox
1
0
Fork 0
mirror of https://github.com/ArchiveBox/ArchiveBox.git synced 2026-04-02 22:07:58 +10:00
Code Issues Packages Projects Releases Wiki Activity

fix: Guess timestamps and add placeholders to support older indices

Browse Source
This commit is contained in:
Cristian
2020-07-24 09:24:52 -05:00
parent 523c384e62
commit 100fa5d1f5
3 changed files with 49 additions and 16 deletions
Download Patch File Download Diff File Expand all files Collapse all files

16
archivebox/index/__init__.py
View File

@@ -529,8 +529,16 @@ def get_unrecognized_folders(links, out_dir: str=OUTPUT_DIR) -> Dict[str, Option
link = None
try:
link = parse_json_link_details(entry.path)
except Exception:
pass
except KeyError:
# Try to fix index
if index_exists:
try:
# Last attempt to repair the detail index
link_guessed = parse_json_link_details(entry.path, guess=True)
write_json_link_details(link_guessed, out_dir=entry.path)
link = parse_json_link_details(entry.path)
except Exception as e:
pass
if index_exists and link is None:
# index exists but it's corrupted or unparseable
@@ -555,9 +563,9 @@ def is_valid(link: Link) -> bool:
return False
if dir_exists and index_exists:
try:
parsed_link = parse_json_link_details(link.link_dir)
parsed_link = parse_json_link_details(link.link_dir, guess=True)
return link.url == parsed_link.url
except Exception:
except Exception as e:
pass
return False