mirrors/ArchiveBox
1
0
Fork 0
mirror of https://github.com/ArchiveBox/ArchiveBox.git synced 2025-05-29 22:15:21 -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

4
archivebox/index/__init__.py
View file

@ -265,14 +265,14 @@ def load_main_index_meta(out_dir: Path=OUTPUT_DIR) -> Optional[dict]:
@enforce_types
def parse_links_from_source(source_path: str, root_url: Optional[str]=None) -> Tuple[List[Link], List[Link]]:
def parse_links_from_source(source_path: str, root_url: Optional[str]=None, parser: str="auto") -> Tuple[List[Link], List[Link]]:
from ..parsers import parse_links
new_links: List[Link] = []
# parse and validate the import file
raw_links, parser_name = parse_links(source_path, root_url=root_url)
raw_links, parser_name = parse_links(source_path, root_url=root_url, parser=parser)
new_links = validate_links(raw_links)
if parser_name: