mirrors/ArchiveBox
1
0
Fork 0
mirror of https://github.com/ArchiveBox/ArchiveBox.git synced 2025-05-17 16:44:26 -04:00
Code Issues Projects Releases Packages Wiki Activity Actions 1

add new SEARCH_BACKEND_TIMEOUT config option defaulted to 90sec

Browse source
This commit is contained in:
Nick Sweeting 2021-04-09 00:27:08 -04:00
parent 34371d027a
commit 32764347ce
2 changed files with 3 additions and 2 deletions
Download patch file Download diff file Expand all files Collapse all files

1
archivebox/config.py
View file

@ -157,6 +157,7 @@ CONFIG_SCHEMA: Dict[str, ConfigDefaultDict] = {
# SONIC # SONIC
'SONIC_COLLECTION': {'type': str, 'default': 'archivebox'}, 'SONIC_COLLECTION': {'type': str, 'default': 'archivebox'},
'SONIC_BUCKET': {'type': str, 'default': 'snapshots'}, 'SONIC_BUCKET': {'type': str, 'default': 'snapshots'},
'SEARCH_BACKEND_TIMEOUT': {'type': int, 'default': 90},
}, },
'DEPENDENCY_CONFIG': { 'DEPENDENCY_CONFIG': {

4
archivebox/search/backends/ripgrep.py
View file

@ -2,7 +2,7 @@ import re
from subprocess import run, PIPE from subprocess import run, PIPE
from typing import List, Generator from typing import List, Generator
from archivebox.config import ARCHIVE_DIR, RIPGREP_VERSION from archivebox.config import ARCHIVE_DIR, RIPGREP_VERSION, SEARCH_BACKEND_TIMEOUT
from archivebox.util import enforce_types from archivebox.util import enforce_types
RG_IGNORE_EXTENSIONS = ('css','js','orig','svg') RG_IGNORE_EXTENSIONS = ('css','js','orig','svg')
@ -32,7 +32,7 @@ def search(text: str) -> List[str]:
from core.models import Snapshot from core.models import Snapshot
rg_cmd = ['rg', RG_ADD_TYPE, RG_IGNORE_ARGUMENTS, RG_DEFAULT_ARGUMENTS, RG_REGEX_ARGUMENT, text, str(ARCHIVE_DIR)] rg_cmd = ['rg', RG_ADD_TYPE, RG_IGNORE_ARGUMENTS, RG_DEFAULT_ARGUMENTS, RG_REGEX_ARGUMENT, text, str(ARCHIVE_DIR)]
rg = run(rg_cmd, stdout=PIPE, stderr=PIPE, timeout=60) rg = run(rg_cmd, stdout=PIPE, stderr=PIPE, timeout=SEARCH_BACKEND_TIMEOUT)
file_paths = [p.decode() for p in rg.stdout.splitlines()] file_paths = [p.decode() for p in rg.stdout.splitlines()]
timestamps = set() timestamps = set()
for path in file_paths: for path in file_paths: