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

fix: Remove title from extractors for oneshot

Browse source
This commit is contained in:
Cristian 2020-07-31 10:24:58 -05:00
parent 8bcb171e74
commit e6c571beb2
4 changed files with 37 additions and 14 deletions
Download patch file Download diff file Expand all files Collapse all files

32
archivebox/extractors/__init__.py
View file

@ -32,22 +32,32 @@ from .git import should_save_git, save_git
from .media import should_save_media, save_media
from .archive_org import should_save_archive_dot_org, save_archive_dot_org
def get_default_archive_methods():
return [
('title', should_save_title, save_title),
('favicon', should_save_favicon, save_favicon),
('wget', should_save_wget, save_wget),
('pdf', should_save_pdf, save_pdf),
('screenshot', should_save_screenshot, save_screenshot),
('dom', should_save_dom, save_dom),
('git', should_save_git, save_git),
('media', should_save_media, save_media),
('archive_org', should_save_archive_dot_org, save_archive_dot_org),
]
@enforce_types
def ignore_methods(to_ignore: List[str]):
ARCHIVE_METHODS = get_default_archive_methods()
methods = filter(lambda x: x[0] not in to_ignore, ARCHIVE_METHODS)
methods = map(lambda x: x[1], methods)
return list(methods)
@enforce_types
def archive_link(link: Link, overwrite: bool=False, methods: Optional[Iterable[str]]=None, out_dir: Optional[str]=None, skip_index: bool=False) -> Link:
"""download the DOM, PDF, and a screenshot into a folder named after the link's timestamp"""
ARCHIVE_METHODS = [
('title', should_save_title, save_title),
('favicon', should_save_favicon, save_favicon),
('wget', should_save_wget, save_wget),
('pdf', should_save_pdf, save_pdf),
('screenshot', should_save_screenshot, save_screenshot),
('dom', should_save_dom, save_dom),
('git', should_save_git, save_git),
('media', should_save_media, save_media),
('archive_org', should_save_archive_dot_org, save_archive_dot_org),
]
ARCHIVE_METHODS = get_default_archive_methods()
if methods is not None:
ARCHIVE_METHODS = [
method for method in ARCHIVE_METHODS