yequari
/
discussfeed
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

add logging and cli arguments

This commit is contained in:
yequari 2023-09-14 22:17:57 -07:00
parent 1befc62b6c
commit 04280d1b9b
1 changed files with 43 additions and 22 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

65
rss.py
View File

@ -1,11 +1,19 @@
import os import os
import argparse
import logging
import xml.etree.ElementTree as ET import xml.etree.ElementTree as ET
import requests import requests
from datetime import datetime, timezone from datetime import datetime, timezone
FEED = 'https://discuss.32bit.cafe/all/new.atom' LOGGER = logging.getLogger(__name__)
FILENAME = os.path.basename(FEED) LOGGER.setLevel(logging.INFO)
WEBHOOK = 'https://discord.com/api/webhooks/1152044037017186417/igGNMxiVhLSW9y4kL8cs7y6m0AutxqXHxMTWSXWN0jbCUnsxLT_er8pqWQKoAjGpPywg'
ch = logging.StreamHandler()
ch.setLevel(logging.INFO)
formatter = logging.Formatter('%(asctime)s - %(name)s - %(levelname)s - %(message)s')
ch.setFormatter(formatter)
LOGGER.addHandler(ch)
def tag(tagname: str) -> str: def tag(tagname: str) -> str:
@ -21,46 +29,45 @@ class Entry:
self.date = datetime.fromisoformat(entry.find(tag('published')).text) self.date = datetime.fromisoformat(entry.find(tag('published')).text)
def __str__(self): def __str__(self):
return f'{self.title} by {self.author} at {self.date} \n {self.url}' return f'New post by {self.author}: {self.title}\n{self.url}'
def get_last_update_time(): def get_last_update_time(file: str) -> datetime:
try: try:
stat = os.stat(FILENAME) stat = os.stat(file)
return datetime.fromtimestamp(stat.st_mtime, return datetime.fromtimestamp(stat.st_mtime,
datetime.now().astimezone().tzinfo) datetime.now().astimezone().tzinfo)
except FileNotFoundError: except FileNotFoundError:
return datetime.now(timezone.utc) return datetime.now(timezone.utc)
def update_rss_file(): def update_rss_file(file, feed):
r = requests.get(FEED, stream=True) r = requests.get(feed, stream=True)
with open(FILENAME, 'w') as fd: with open(file, 'w') as fd:
for chunk in r.iter_content(chunk_size=1024, decode_unicode=True): for chunk in r.iter_content(chunk_size=1024, decode_unicode=True):
fd.write(chunk) fd.write(chunk)
def parse_rss_file(): def parse_rss_file(file) -> ET.Element:
tree = ET.parse(FILENAME) tree = ET.parse(file)
return tree.getroot() return tree.getroot()
def get_new_posts(time, root): def get_new_posts(time, root) -> list[Entry]:
posts = [] posts = []
for entry in root.findall(tag('entry')): for entry in root.findall(tag('entry')):
e = Entry(entry) e = Entry(entry)
print(e.date)
if e.date > time: if e.date > time:
posts.append(e) posts.append(e)
return posts return posts
def send_message(entry): def send_message(entry, url) -> int:
obj = { obj = {
'content': str(entry) 'content': str(entry)
} }
r = requests.post(WEBHOOK, obj) r = requests.post(url, obj)
print(r.status_code) return r.status_code
# Check last mod date of current rss file # Check last mod date of current rss file
# Download new version of file # Download new version of file
@ -69,11 +76,25 @@ def send_message(entry):
if __name__ == "__main__": if __name__ == "__main__":
last_updated = get_last_update_time() parser = argparse.ArgumentParser(
print(last_updated) prog='discussfeed',
update_rss_file() description='Checks for new entries from an atom feed and sends them to a discord webhook')
root = parse_rss_file() parser.add_argument('-w', '--webhook')
parser.add_argument('feed_url')
args = parser.parse_args()
filename = os.path.basename(args.feed_url)
last_updated = get_last_update_time(filename)
LOGGER.info(f'last updated: {last_updated}')
update_rss_file(filename, args.feed_url)
root = parse_rss_file(filename)
posts = get_new_posts(last_updated, root) posts = get_new_posts(last_updated, root)
print(posts) LOGGER.info(f'found {len(posts)} new posts')
for post in posts: for post in posts:
send_message(post) status = send_message(post, args.webhook)
if status >= 300:
LOGGER.error(f'Response code {status}')
else:
LOGGER.info(f'response code {status}')