mirror of
https://github.com/paperless-ngx/paperless-ngx.git
synced 2025-04-29 11:09:27 -05:00
Add timeout for Tika client as suggested in
https://github.com/paperless-ngx/paperless-ngx/discussions/8509
This commit is contained in:
parent
3f7a0802a4
commit
e76c8689e9
1
.github/workflows/ci.yml
vendored
1
.github/workflows/ci.yml
vendored
@ -12,6 +12,7 @@ on:
|
||||
pull_request:
|
||||
branches-ignore:
|
||||
- 'translations**'
|
||||
workflow_dispatch:
|
||||
|
||||
env:
|
||||
# This is the version of pipenv all the steps will use
|
||||
|
@ -33,7 +33,10 @@ class TikaDocumentParser(DocumentParser):
|
||||
|
||||
def extract_metadata(self, document_path, mime_type):
|
||||
try:
|
||||
with TikaClient(tika_url=settings.TIKA_ENDPOINT) as client:
|
||||
with TikaClient(
|
||||
tika_url=settings.TIKA_ENDPOINT,
|
||||
timeout=settings.CELERY_TASK_TIME_LIMIT,
|
||||
) as client:
|
||||
parsed = client.metadata.from_file(document_path, mime_type)
|
||||
return [
|
||||
{
|
||||
@ -54,7 +57,10 @@ class TikaDocumentParser(DocumentParser):
|
||||
self.log.info(f"Sending {document_path} to Tika server")
|
||||
|
||||
try:
|
||||
with TikaClient(tika_url=settings.TIKA_ENDPOINT) as client:
|
||||
with TikaClient(
|
||||
tika_url=settings.TIKA_ENDPOINT,
|
||||
timeout=settings.CELERY_TASK_TIME_LIMIT,
|
||||
) as client:
|
||||
try:
|
||||
parsed = client.tika.as_text.from_file(document_path, mime_type)
|
||||
except httpx.HTTPStatusError as err:
|
||||
|
Loading…
x
Reference in New Issue
Block a user