mirrors/ArchiveBox
1
0
Fork 0
mirror of https://github.com/ArchiveBox/ArchiveBox.git synced 2025-05-18 17:14:39 -04:00
Code Issues Projects Releases Packages Wiki Activity Actions

Added yt-dlp everywhere

Browse source
This commit is contained in:
Joseph Turian 2022-09-12 20:34:02 +00:00
parent e41f313fa3
commit f5f7aff3b4
10 changed files with 18 additions and 11 deletions
Download patch file Download diff file Expand all files Collapse all files

3
archivebox/config.py
View file

@ -203,7 +203,8 @@ CONFIG_SCHEMA: Dict[str, ConfigDefaultDict] = {
'SINGLEFILE_BINARY': {'type': str, 'default': lambda c: bin_path('single-file')},
'READABILITY_BINARY': {'type': str, 'default': lambda c: bin_path('readability-extractor')},
'MERCURY_BINARY': {'type': str, 'default': lambda c: bin_path('mercury-parser')},
'YOUTUBEDL_BINARY': {'type': str, 'default': 'youtube-dl'},
#'YOUTUBEDL_BINARY': {'type': str, 'default': 'youtube-dl'},
'YOUTUBEDL_BINARY': {'type': str, 'default': 'yt-dlp'},
'NODE_BINARY': {'type': str, 'default': 'node'},
'RIPGREP_BINARY': {'type': str, 'default': 'rg'},
'CHROME_BINARY': {'type': str, 'default': None},

1
archivebox/extractors/__init__.py
View file

@ -1,6 +1,7 @@
__package__ = 'archivebox.extractors'
import os
import sys
from pathlib import Path
from typing import Optional, List, Iterable, Union

1
archivebox/extractors/media.py
View file

@ -72,6 +72,7 @@ def save_media(link: Link, out_dir: Optional[Path]=None, timeout: int=MEDIA_TIME
timer.end()
# add video description and subtitles to full-text index
# Let's try a few different
index_texts = [
text_file.read_text(encoding='utf-8').strip()
for text_file in (