mirror/archivebox
1
0
Fork 0
mirror of https://github.com/pirate/ArchiveBox.git synced 2025-08-21 05:41:54 +02:00
Code Issues Releases Wiki Activity

fix: Invalid file path being generated on add

Browse Source
This commit is contained in:
Cristian
2021-01-11 08:28:41 -05:00
parent 1ce6130202
commit d5cabe05af
2 changed files with 1 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

1
archivebox/core/models.py
View File

@@ -213,7 +213,6 @@ class Snapshot(models.Model):
return self.history['title'][-1].output.strip() return self.history['title'][-1].output.strip()
return None return None
@cached_property @cached_property
def domain(self) -> str: def domain(self) -> str:
from ..util import domain from ..util import domain

2
archivebox/main.py
View File

@@ -576,7 +576,7 @@ def add(urls: Union[str, List[str]],
log_crawl_started(new_snapshots) log_crawl_started(new_snapshots)
for new_snapshot in new_snapshots: for new_snapshot in new_snapshots:
# TODO: Check if we need to add domain to the Snapshot model # TODO: Check if we need to add domain to the Snapshot model
downloaded_file = save_file_as_source(new_snapshot.url, filename=f'{new_snapshot.timestamp}-crawl-{new_snapshot.url}.txt', out_dir=out_dir) downloaded_file = save_file_as_source(new_snapshot.url, filename=f'{new_snapshot.timestamp}-crawl-{new_snapshot.domain}.txt', out_dir=out_dir)
new_snapshots_depth += parse_snapshots_from_source(downloaded_file, root_url=new_snapshot.url) new_snapshots_depth += parse_snapshots_from_source(downloaded_file, root_url=new_snapshot.url)
imported_snapshots = [Snapshot(url=snapshot.url) for snapshot in new_snapshots + new_snapshots_depth] imported_snapshots = [Snapshot(url=snapshot.url) for snapshot in new_snapshots + new_snapshots_depth]