forks/archivebox
1
0
Fork 0
Code Releases Activity

fix: Handle case when update tries to re-add a link that is not in the sql index

Browse source
This commit is contained in:
Cristian 2020-11-04 15:02:54 -05:00
parent f292cface2
commit d064a3eeff
2 changed files with 7 additions and 2 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

6
archivebox/extractors/__init__.py
View file

@ -8,6 +8,7 @@ from datetime import datetime
from django.db.models import QuerySet from django.db.models import QuerySet
from ..index.schema import Link from ..index.schema import Link
from ..index.sql import write_link_to_sql_index
from ..index import ( from ..index import (
load_link_details, load_link_details,
write_link_details, write_link_details,
@ -68,7 +69,10 @@ def archive_link(link: Link, overwrite: bool=False, methods: Optional[Iterable[s
# TODO: Remove when the input is changed to be a snapshot. Suboptimal approach. # TODO: Remove when the input is changed to be a snapshot. Suboptimal approach.
if not skip_index: if not skip_index:
from core.models import Snapshot, ArchiveResult from core.models import Snapshot, ArchiveResult
snapshot = Snapshot.objects.get(url=link.url) try:
snapshot = Snapshot.objects.get(url=link.url) # TODO: This will be unnecessary once everything is a snapshot
except Snapshot.DoesNotExist:
write_link_to_sql_index(link)
ARCHIVE_METHODS = get_default_archive_methods() ARCHIVE_METHODS = get_default_archive_methods()

3
tests/test_update.py
View file

@ -6,7 +6,7 @@ def test_update_status_invalid(tmp_path, process, disable_extractors_dict):
subprocess.run(['archivebox', 'add', 'http://127.0.0.1:8080/static/example.com.html'], capture_output=True, env=disable_extractors_dict) subprocess.run(['archivebox', 'add', 'http://127.0.0.1:8080/static/example.com.html'], capture_output=True, env=disable_extractors_dict)
assert list((tmp_path / "archive").iterdir()) != [] assert list((tmp_path / "archive").iterdir()) != []
subprocess.run(['archivebox', 'remove', 'http://127.0.0.1:8080/static/example.com.html', '--yes'], capture_output=True) a_process = subprocess.run(['archivebox', 'remove', 'http://127.0.0.1:8080/static/example.com.html', '--yes'], capture_output=True)
conn = sqlite3.connect(str(tmp_path / "index.sqlite3")) conn = sqlite3.connect(str(tmp_path / "index.sqlite3"))
c = conn.cursor() c = conn.cursor()
@ -17,6 +17,7 @@ def test_update_status_invalid(tmp_path, process, disable_extractors_dict):
assert link is None assert link is None
update_process = subprocess.run(['archivebox', 'update', '--status=invalid'], capture_output=True, env=disable_extractors_dict) update_process = subprocess.run(['archivebox', 'update', '--status=invalid'], capture_output=True, env=disable_extractors_dict)
#breakpoint()
conn = sqlite3.connect(str(tmp_path / "index.sqlite3")) conn = sqlite3.connect(str(tmp_path / "index.sqlite3"))
c = conn.cursor() c = conn.cursor()