mastotuit/mastotuit.py

387 líneas
9,4 KiB
Python
Original Vista normal Històric

2021-09-10 20:32:11 +02:00
import os
import feedparser
2021-09-29 19:06:31 +02:00
from bs4 import BeautifulSoup
2021-09-10 20:32:11 +02:00
from mastodon import Mastodon
import psycopg2
import sys
import time
2021-09-29 19:06:31 +02:00
import requests
import shutil
2021-09-10 20:32:11 +02:00
import tweepy
from tweepy import TweepyException
2021-09-10 20:32:11 +02:00
import logging
import filetype
import ffmpeg
2021-09-10 20:32:11 +02:00
import pdb
logger = logging.getLogger()
2021-10-08 14:30:00 +02:00
def get_toot(title):
soup = BeautifulSoup(title, 'html.parser')
toot_text = soup.get_text()
sub_str = 'http'
find_link = toot_text.find(sub_str)
if find_link != -1:
tuit_text = toot_text[:toot_text.index(sub_str)]
else:
tuit_text = toot_text
links_lst = ''
for links in soup.find_all('a'):
find_tag = links.get('href').find('/tags/')
if find_tag == -1:
links_lst += links.get('href')
if len(links_lst) > 0:
last_text = toot_text[len(tuit_text) + len(links_lst):]
else:
last_text = ''
tuit_text = f'{tuit_text} {links_lst} {last_text}'
return tuit_text
2021-10-15 13:15:52 +02:00
def compose_tweet(tuit_text, with_images, is_reply):
2021-10-19 15:51:21 +02:00
images_id_lst = []
if with_images:
2021-10-19 15:51:21 +02:00
for image in images_list:
2021-10-19 15:51:21 +02:00
kind = filetype.guess('images/' + image)
is_video = True if kind.mime == 'video/mp4' else False
if is_video:
2021-10-19 15:51:21 +02:00
try:
2021-10-19 15:51:21 +02:00
probe = ffmpeg.probe(f'images/{image}')
video_duration = float(
2021-10-19 15:51:21 +02:00
probe['streams'][0]['duration']
)
2021-10-19 15:51:21 +02:00
except Exception as e:
2021-10-19 15:51:21 +02:00
print(f'Error while trying to probe {image}\n{e}')
sys.exit(e)
if video_duration > 139:
2021-10-19 15:51:21 +02:00
print(f'video duration is too large: {video_duration}')
continue
try:
2021-10-19 15:51:21 +02:00
media_upload = api.media_upload(
f'images/{image}',
2021-10-19 15:51:21 +02:00
media_category='tweet_video' if is_video else 'tweet_image'
)
2021-10-19 15:51:21 +02:00
if is_video:
if media_upload.processing_info['state'] == 'succeeded':
images_id_lst.append(media_upload.media_id)
else:
images_id_lst.append(media_upload.media_id)
except TweepyException as err:
2021-10-19 15:51:21 +02:00
print('Error while uploading media!\n')
sys.exit(err)
# Compose tuit
2021-10-16 15:58:41 +02:00
tuit_text2 = ''
if len(tuit_text) > 280:
2021-10-19 15:51:21 +02:00
2021-10-16 15:58:41 +02:00
tuit_max_length = 250 if with_images else 275
2021-10-19 15:51:21 +02:00
tuit_text1 = '{0} (1/2)'.format(
tuit_text[:tuit_max_length].rsplit(' ', 1)[0]
)
tuit_text2 = '{0} (2/2)'.format(
2021-10-19 15:51:21 +02:00
tuit_text[len(tuit_text1) - 6:]
)
try:
2021-10-19 15:51:21 +02:00
first_tweet = api.update_status(
2021-10-19 15:51:21 +02:00
status=tuit_text1,
in_reply_to_status_id=tweet_id if is_reply else ''
)
tweet = api.update_status(
status=tuit_text2,
in_reply_to_status_id=first_tweet.id,
2021-10-19 15:51:21 +02:00
media_ids=images_id_lst
)
except TweepyException as err:
2021-10-19 15:51:21 +02:00
print('Error while trying to publish split tweet.\n')
sys.exit(err)
2021-10-15 13:15:52 +02:00
else:
2021-10-19 15:51:21 +02:00
try:
2021-10-19 15:51:21 +02:00
tweet = api.update_status(
status=tuit_text,
in_reply_to_status_id=tweet_id if is_reply else '',
2021-10-19 15:51:21 +02:00
media_ids=images_id_lst
)
2021-10-19 15:51:21 +02:00
except TweepyException as err:
2021-10-19 15:51:21 +02:00
print('Error while trying to publish tweet.\n')
sys.exit(err)
return tweet
2021-10-15 13:15:52 +02:00
2021-10-07 20:23:17 +02:00
def get_tweet_id(toot_id):
tweet_id = 0
try:
conn = None
conn = psycopg2.connect(database = feeds_db, user = feeds_db_user, password = "", host = "/var/run/postgresql", port = "5432")
cur = conn.cursor()
cur.execute('select tweet_id from id where toot_id=(%s)', (toot_id,))
row = cur.fetchone()
if row != None:
tweet_id = row[0]
cur.close()
return(tweet_id)
except (Exception, psycopg2.DatabaseError) as error:
print(error)
finally:
if conn is not None:
conn.close()
2021-09-29 19:06:31 +02:00
def write_image(image_url):
2021-09-10 20:32:11 +02:00
2021-09-29 19:06:31 +02:00
if not os.path.exists('images'):
os.makedirs('images')
filename = image_url.split("/") [-1]
r = requests.get(image_url, stream = True)
r.raw.decode_content = True
with open('images/' + filename, 'wb') as f:
shutil.copyfileobj(r.raw, f)
return filename
2021-09-10 20:32:11 +02:00
def create_api():
auth = tweepy.OAuthHandler(api_key, api_key_secret)
auth.set_access_token(access_token, access_token_secret)
api = tweepy.API(auth)
2021-09-10 20:32:11 +02:00
try:
api.verify_credentials()
2021-09-11 18:27:54 +02:00
logged_in = True
2021-09-10 20:32:11 +02:00
except Exception as e:
logger.error("Error creating API", exc_info=True)
raise e
logger.info("API created")
2021-09-11 18:27:54 +02:00
return (api, logged_in)
2021-09-10 20:32:11 +02:00
def mastodon():
# Load secrets from secrets file
secrets_filepath = "secrets/secrets.txt"
uc_client_id = get_parameter("uc_client_id", secrets_filepath)
uc_client_secret = get_parameter("uc_client_secret", secrets_filepath)
uc_access_token = get_parameter("uc_access_token", secrets_filepath)
# Load configuration from config file
config_filepath = "config/config.txt"
mastodon_hostname = get_parameter("mastodon_hostname", config_filepath)
# Initialise Mastodon API
mastodon = Mastodon(
client_id=uc_client_id,
client_secret=uc_client_secret,
access_token=uc_access_token,
api_base_url='https://' + mastodon_hostname,
)
# Initialise access headers
headers = {'Authorization': 'Bearer %s'%uc_access_token}
return (mastodon, mastodon_hostname)
def db_config():
# Load db configuration from config file
db_config_filepath = "config/db_config.txt"
feeds_db = get_parameter("feeds_db", db_config_filepath)
feeds_db_user = get_parameter("feeds_db_user", db_config_filepath)
feeds_url = get_parameter("feeds_url", db_config_filepath)
return (feeds_db, feeds_db_user, feeds_url)
def twitter_config():
twitter_config_filepath = "config/keys_config.txt"
api_key = get_parameter("api_key", twitter_config_filepath)
api_key_secret = get_parameter("api_key_secret", twitter_config_filepath)
access_token = get_parameter("access_token", twitter_config_filepath)
access_token_secret = get_parameter("access_token_secret", twitter_config_filepath)
return(api_key, api_key_secret, access_token, access_token_secret)
# Returns the parameter from the specified file
def get_parameter( parameter, file_path ):
# Check if secrets file exists
if not os.path.isfile(file_path):
print("File %s not found, exiting."%file_path)
sys.exit(0)
# Find parameter in file
with open( file_path ) as f:
for line in f:
if line.startswith( parameter ):
return line.replace(parameter + ":", "").strip()
# Cannot find parameter, exit
print(file_path + " Missing parameter %s "%parameter)
sys.exit(0)
###############################################################################
# main
if __name__ == '__main__':
2021-10-07 20:23:17 +02:00
mastodon, mastodon_hostname = mastodon()
2021-09-10 20:32:11 +02:00
feeds_db, feeds_db_user, feeds_url = db_config()
api_key, api_key_secret, access_token, access_token_secret = twitter_config()
2021-10-07 20:23:17 +02:00
2021-09-11 18:27:54 +02:00
logged_in = False
2021-09-10 20:32:11 +02:00
try:
newsfeeds = feedparser.parse(feeds_url)
except:
print(newsfeeds.status)
sys.exit(0)
for entry in newsfeeds.entries:
2021-09-29 19:06:31 +02:00
publish = False
with_images = False
2021-10-07 20:23:17 +02:00
is_reply = False
2021-09-29 19:06:31 +02:00
2021-09-10 20:32:11 +02:00
title = entry['summary']
id = entry['id']
link = entry['link']
2021-10-07 20:23:17 +02:00
toot_id = link.rsplit('/')[4]
tweet_id = get_tweet_id(toot_id)
2021-10-07 20:23:17 +02:00
if tweet_id == 0:
2021-09-29 19:06:31 +02:00
publish = True
2021-09-29 19:06:31 +02:00
reply_id = mastodon.status(toot_id).in_reply_to_id
if reply_id != None:
2021-10-09 13:30:18 +02:00
is_reply = True
tweet_id = get_tweet_id(reply_id)
2021-09-10 20:32:11 +02:00
if len(entry.links) >= 2:
2021-09-10 20:32:11 +02:00
with_images = True
images_list = []
images = len(entry.links) - 1
2021-09-10 20:32:11 +02:00
i = 0
while i < images:
2021-09-10 20:32:11 +02:00
image_url = entry.links[i+1].href
image_filename = write_image(image_url)
images_list.append(image_filename)
i += 1
2021-09-10 20:32:11 +02:00
###########################################################
2021-09-29 19:06:31 +02:00
if publish:
2021-09-10 20:32:11 +02:00
2021-10-08 14:30:00 +02:00
tuit_text = get_toot(title)
2021-09-10 20:32:11 +02:00
print("Tooting...")
2021-09-30 18:05:35 +02:00
print(tuit_text)
2021-09-11 18:27:54 +02:00
if not logged_in:
2021-09-29 19:06:31 +02:00
2021-09-11 18:27:54 +02:00
api, logged_in = create_api()
2021-10-07 20:23:17 +02:00
2021-10-15 13:15:52 +02:00
tweet = compose_tweet(tuit_text, with_images, is_reply)
2021-09-10 20:32:11 +02:00
#########################################################
2021-10-09 13:30:18 +02:00
sql_insert_ids = 'INSERT INTO id(toot_id, tweet_id) VALUES (%s,%s)'
2021-10-07 20:23:17 +02:00
conn = None
try:
conn = psycopg2.connect(database = feeds_db, user = feeds_db_user, password = "", host = "/var/run/postgresql", port = "5432")
cur = conn.cursor()
2021-10-09 13:30:18 +02:00
cur.execute(sql_insert_ids, (toot_id, tweet.id))
2021-10-07 20:23:17 +02:00
conn.commit()
cur.close()
except (Exception, psycopg2.DatabaseError) as error:
print(error)
finally:
if conn is not None:
conn.close()
2021-10-08 14:30:00 +02:00
#########################################################
time.sleep(2)
2021-09-10 20:32:11 +02:00
else:
print("Any new feeds")
sys.exit(0)