mirrors/ArchiveBox
1
0
Fork 0
mirror of https://github.com/ArchiveBox/ArchiveBox.git synced 2025-05-13 06:34:25 -04:00
Code Issues Projects Releases Packages Wiki Activity Actions 1

only start parsing json after the first open brace

Browse source
This commit is contained in:
Nick Sweeting 2023-09-03 21:40:12 -07:00 • committed by GitHub
parent eb97fd427b
commit aaca74f6a8
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
1 changed files with 4 additions and 2 deletions
Download patch file Download diff file Expand all files Collapse all files

6
archivebox/parsers/generic_json.py
View file

@ -17,8 +17,10 @@ def parse_generic_json_export(json_file: IO[str], **_kwargs) -> Iterable[Link]:
"""Parse JSON-format bookmarks export files (produced by pinboard.in/export/, or wallabag)""" """Parse JSON-format bookmarks export files (produced by pinboard.in/export/, or wallabag)"""
json_file.seek(0) json_file.seek(0)
next(json_file)
links = json.load(json_file) # sometimes the first line is a comment or filepath, so we get everything after the first {
json_file_json_str = '{' + json_file.read().split('{', 1)[-1]
links = json.loads(json_file_json_str)
json_date = lambda s: datetime.strptime(s, '%Y-%m-%dT%H:%M:%S%z') json_date = lambda s: datetime.strptime(s, '%Y-%m-%dT%H:%M:%S%z')
for link in links: for link in links: