mirrors/ArchiveBox
1
0
Fork 0
mirror of https://github.com/ArchiveBox/ArchiveBox.git synced 2025-05-23 11:17:02 -04:00
Code Issues Projects Releases Packages Wiki Activity Actions

refactor: archive_org uses snapshot instead of link

Browse source
This commit is contained in:
Cristian 2020-12-29 14:11:31 -05:00
parent 5ea1b9e39f
commit de3c82730c
1 changed files with 8 additions and 6 deletions
Download patch file Download diff file Expand all files Collapse all files

14
archivebox/extractors/archive_org.py
View file

@ -5,6 +5,8 @@ from pathlib import Path
from typing import Optional, List, Dict, Tuple from typing import Optional, List, Dict, Tuple
from collections import defaultdict from collections import defaultdict
from django.db.models import Model
from ..index.schema import Link, ArchiveResult, ArchiveOutput, ArchiveError from ..index.schema import Link, ArchiveResult, ArchiveOutput, ArchiveError
from ..system import run, chmod_file from ..system import run, chmod_file
from ..util import ( from ..util import (
@ -25,9 +27,9 @@ from ..logging_util import TimedProgress
@enforce_types @enforce_types
def should_save_archive_dot_org(link: Link, out_dir: Optional[Path]=None) -> bool: def should_save_archive_dot_org(snapshot: Model, out_dir: Optional[Path]=None) -> bool:
out_dir = out_dir or Path(link.link_dir) out_dir = out_dir or Path(snapshot.snapshot_dir)
if is_static_file(link.url): if is_static_file(snapshot.url):
return False return False
if (out_dir / "archive.org.txt").exists(): if (out_dir / "archive.org.txt").exists():
@ -37,13 +39,13 @@ def should_save_archive_dot_org(link: Link, out_dir: Optional[Path]=None) -> boo
return SAVE_ARCHIVE_DOT_ORG return SAVE_ARCHIVE_DOT_ORG
@enforce_types @enforce_types
def save_archive_dot_org(link: Link, out_dir: Optional[Path]=None, timeout: int=TIMEOUT) -> ArchiveResult: def save_archive_dot_org(snapshot: Model, out_dir: Optional[Path]=None, timeout: int=TIMEOUT) -> ArchiveResult:
"""submit site to archive.org for archiving via their service, save returned archive url""" """submit site to archive.org for archiving via their service, save returned archive url"""
out_dir = out_dir or Path(link.link_dir) out_dir = out_dir or Path(snapshot.snapshot_dir)
output: ArchiveOutput = 'archive.org.txt' output: ArchiveOutput = 'archive.org.txt'
archive_org_url = None archive_org_url = None
submit_url = 'https://web.archive.org/save/{}'.format(link.url) submit_url = 'https://web.archive.org/save/{}'.format(snapshot.url)
cmd = [ cmd = [
CURL_BINARY, CURL_BINARY,
*CURL_ARGS, *CURL_ARGS,