Mirrors/ArchiveBox
2
0
Fork 0
mirror of https://github.com/ArchiveBox/ArchiveBox synced 2024-11-10 06:34:16 +00:00
Code Issues Projects Releases Packages Wiki Activity

Add exception handling for indexable content reader

Browse source
This commit is contained in:
jdcaballerov 2020-11-24 09:35:06 -05:00 committed by Nick Sweeting
parent b1d70185ed
commit 5a6b814c79
1 changed files with 10 additions and 2 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

12
archivebox/search/__init__.py
View file

@ -97,6 +97,14 @@ def index_links(links: Union[List[Link],None], out_dir: Path=OUTPUT_DIR):
for link in links:
if snap := Snapshot.objects.filter(url=link.url).first():
results = ArchiveResult.objects.indexable().filter(snapshot=snap)
texts = get_indexable_content(results)
log_index_started(link.url)
write_search_index(link, texts, out_dir=out_dir)
try:
texts = get_indexable_content(results)
except Exception as err:
stderr()
stderr(
f'[X] An Exception ocurred reading the indexable content={err}:',
color='red',
)
else:
write_search_index(link, texts, out_dir=out_dir)