mirrors/ArchiveBox
1
0
Fork 0
mirror of https://github.com/ArchiveBox/ArchiveBox.git synced 2025-05-14 07:04:27 -04:00
Code Issues Projects Releases Packages Wiki Activity Actions

Merge pull request #669 from FliegendeWurst/fix-issue-235

Browse source 
add command: --parser option (fixes #235)
This commit is contained in:
Nick Sweeting 2021-03-31 00:53:47 -04:00 • committed by GitHub
commit 36f0646501
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
6 changed files with 73 additions and 21 deletions
Download patch file Download diff file Expand all files Collapse all files

3
archivebox/main.py
View file

@ -568,6 +568,7 @@ def add(urls: Union[str, List[str]],
overwrite: bool=False,
init: bool=False,
extractors: str="",
parser: str="auto",
out_dir: Path=OUTPUT_DIR) -> List[Link]:
"""Add a new URL or list of URLs to your archive"""
@ -594,7 +595,7 @@ def add(urls: Union[str, List[str]],
# save verbatim args to sources
write_ahead_log = save_text_as_source('\n'.join(urls), filename='{ts}-import.txt', out_dir=out_dir)
new_links += parse_links_from_source(write_ahead_log, root_url=None)
new_links += parse_links_from_source(write_ahead_log, root_url=None, parser=parser)
# If we're going one level deeper, download each link and look for more links
new_links_depth = []