rss-feedler/src/app.py

47 lines
1.4 KiB
Python
Raw Normal View History

2024-10-16 00:04:01 +03:00
import argparse
import json
import requests.exceptions
from requests import Session
from requests.adapters import HTTPAdapter
from urllib3.util import Retry
from src.logger import logger as l
logger = l.getChild(__name__)
logger.info("Starting")
parser = argparse.ArgumentParser(__name__)
parser.add_argument('-s', '--settings', required=True, help='Settings JSON')
parser.add_argument('-d', '--directory', required=True, help='Directory to store XMLs')
args = parser.parse_args()
logger.info("Loading settings from settings JSON")
settings = []
with open(args.settings) as json_file:
settings = json.load(json_file)
assert type(settings) == list
for sets in settings:
logger.info(f"Working set: {sets}")
assert type(sets) == dict
assert 'src' in sets
assert 'rss' in sets
s = Session()
retries = Retry(
total=3,
backoff_factor=0.1,
status_forcelist=[502, 503, 504],
allowed_methods={'GET'},
)
s.mount(sets['src'], HTTPAdapter(max_retries=retries))
try:
r = s.get(sets['src'], timeout=3, stream=True)
if r.status_code == 200:
logger.info(f"Saving to file: {sets['rss']}")
with open("{0}/{1}".format(args.directory, sets['rss']), 'wb') as rss:
for chunk in r:
rss.write(chunk)
except requests.exceptions.ConnectionError as e:
logger.warning(f"Unable to fetch {sets['src']}", e)