mirrors/ArchiveBox
1
0
Fork 0
mirror of https://github.com/ArchiveBox/ArchiveBox.git synced 2025-05-15 07:34:27 -04:00
Code Issues Projects Releases Packages Wiki Activity Actions

wip

Browse source
This commit is contained in:
Nick Sweeting 2024-10-25 01:06:12 -07:00
parent 4b6f08b0fe
commit 5d9a32c364
No known key found for this signature in database
178 changed files with 2982 additions and 1322 deletions
Download patch file Download diff file Expand all files Collapse all files

20
packages/abx-plugin-readability-extractor/extractors.py Normal file
View file

@ -0,0 +1,20 @@
__package__ = 'plugins_extractor.readability'
from pathlib import Path
from pydantic_pkgr import BinName
from abx.archivebox.base_extractor import BaseExtractor
from .binaries import READABILITY_BINARY
class ReadabilityExtractor(BaseExtractor):
name: str = 'readability'
binary: BinName = READABILITY_BINARY.name
def get_output_path(self, snapshot) -> Path:
return Path(snapshot.link_dir) / 'readability' / 'content.html'
READABILITY_EXTRACTOR = ReadabilityExtractor()