kamp/searxng
1
0
Fork 0
mirror of https://github.com/searxng/searxng.git synced 2024-11-22 12:10:11 +01:00
Code

Checker: change image check

Browse Source 
In the master branch, the checker starts to stream the response
and cut the connection. However it creates a lot of read error,
which are false negative. I don't know how to fix the issue.

This commit change the checker to download the whole image.
The error reporting is also changed to report only one line,
instead of the whole stacktrace.

Also, if a timeout occurs, the checker waits for one second
before retry.

Do note that I have not test the checker running in background.
This feature seems forgotten and lack of interrest despite the
initial move few years ago.
This commit is contained in:
Alexandre FLAMENT 2024-03-09 09:52:02 +00:00
parent 42515d98f7
commit f49d1a9b90
1 changed files with 11 additions and 10 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

21
searx/search/checker/impl.py
View File

@ -1,11 +1,11 @@
# SPDX-License-Identifier: AGPL-3.0-or-later # SPDX-License-Identifier: AGPL-3.0-or-later
import gc import gc
import time
import typing import typing
import types import types
import functools import functools
import itertools import itertools
from time import time
from timeit import default_timer from timeit import default_timer
from urllib.parse import urlparse from urllib.parse import urlparse
@ -65,15 +65,15 @@ def _download_and_check_if_image(image_url: str) -> bool:
This function should not be called directly: use _is_url_image This function should not be called directly: use _is_url_image
otherwise the cache of functools.lru_cache contains data: URL which might be huge. otherwise the cache of functools.lru_cache contains data: URL which might be huge.
""" """
retry = 2 retry = 3
while retry > 0: while retry > 0:
a = time() a = default_timer()
try: try:
# use "image_proxy" (avoid HTTP/2) # use "image_proxy" (avoid HTTP/2)
network.set_timeout_for_thread(2)
network.set_context_network_name('image_proxy') network.set_context_network_name('image_proxy')
r, stream = network.stream( r = network.get(
'GET',
image_url, image_url,
timeout=10.0, timeout=10.0,
allow_redirects=True, allow_redirects=True,
@ -89,19 +89,20 @@ def _download_and_check_if_image(image_url: str) -> bool:
'Cache-Control': 'max-age=0', 'Cache-Control': 'max-age=0',
}, },
) )
r.close()
if r.status_code == 200: if r.status_code == 200:
is_image = r.headers.get('content-type', '').startswith('image/') is_image = r.headers.get('content-type', '').startswith('image/')
else: else:
is_image = False is_image = False
del r del r
del stream
return is_image return is_image
except httpx.TimeoutException: except httpx.TimeoutException:
logger.error('Timeout for %s: %i', image_url, int(time() - a)) logger.error('Timeout for %s: %i', image_url, int(default_timer() - a))
time.sleep(1)
retry -= 1 retry -= 1
except httpx.HTTPError: except httpx.HTTPStatusError as e:
logger.exception('Exception for %s', image_url) logger.error('Exception for %s: HTTP status=%i', image_url, e.response.status_code)
except httpx.HTTPError as e:
logger.error('Exception for %s: %s, %s', image_url, e.__class__.__name__, ",".join(e.args))
return False return False
return False return False