2020-10-04 09:51:12 +00:00
|
|
|
__filename__ = "newswire.py"
|
|
|
|
__author__ = "Bob Mottram"
|
|
|
|
__license__ = "AGPL3+"
|
2022-02-03 13:58:20 +00:00
|
|
|
__version__ = "1.3.0"
|
2020-10-04 09:51:12 +00:00
|
|
|
__maintainer__ = "Bob Mottram"
|
2021-09-10 16:14:50 +00:00
|
|
|
__email__ = "bob@libreserver.org"
|
2020-10-04 09:51:12 +00:00
|
|
|
__status__ = "Production"
|
2021-06-26 11:27:14 +00:00
|
|
|
__module_group__ = "Web Interface Columns"
|
2020-10-04 09:51:12 +00:00
|
|
|
|
|
|
|
import os
|
2021-02-12 11:28:00 +00:00
|
|
|
import json
|
2020-10-04 09:51:12 +00:00
|
|
|
import requests
|
2022-01-27 10:42:46 +00:00
|
|
|
import random
|
2022-04-27 20:02:56 +00:00
|
|
|
import time
|
2020-10-04 09:51:12 +00:00
|
|
|
from socket import error as SocketError
|
|
|
|
import errno
|
|
|
|
from datetime import datetime
|
2020-11-22 15:33:11 +00:00
|
|
|
from datetime import timedelta
|
2020-11-22 19:09:35 +00:00
|
|
|
from datetime import timezone
|
2020-10-04 09:51:12 +00:00
|
|
|
from collections import OrderedDict
|
2021-12-26 12:31:47 +00:00
|
|
|
from utils import valid_post_date
|
2021-12-29 21:55:09 +00:00
|
|
|
from categories import set_hashtag_category
|
2022-02-12 20:37:15 +00:00
|
|
|
from utils import get_domain_from_actor
|
2022-01-13 15:10:41 +00:00
|
|
|
from utils import valid_hash_tag
|
2021-12-27 21:44:48 +00:00
|
|
|
from utils import dangerous_svg
|
2021-12-26 16:01:32 +00:00
|
|
|
from utils import get_fav_filename_from_url
|
2021-12-26 11:29:40 +00:00
|
|
|
from utils import get_base_content_from_post
|
2021-12-26 10:57:03 +00:00
|
|
|
from utils import has_object_dict
|
2021-12-27 15:52:08 +00:00
|
|
|
from utils import first_paragraph_from_string
|
2021-12-28 14:41:10 +00:00
|
|
|
from utils import is_public_post
|
2021-12-26 20:36:08 +00:00
|
|
|
from utils import locate_post
|
2021-12-26 15:13:34 +00:00
|
|
|
from utils import load_json
|
2021-12-26 14:47:21 +00:00
|
|
|
from utils import save_json
|
2021-12-27 15:37:31 +00:00
|
|
|
from utils import is_suspended
|
2021-12-27 17:53:41 +00:00
|
|
|
from utils import contains_invalid_chars
|
2021-12-27 15:43:22 +00:00
|
|
|
from utils import remove_html
|
2021-12-26 18:46:43 +00:00
|
|
|
from utils import is_account_dir
|
2021-12-26 12:02:29 +00:00
|
|
|
from utils import acct_dir
|
2021-12-26 10:19:59 +00:00
|
|
|
from utils import local_actor_url
|
2021-12-28 21:55:38 +00:00
|
|
|
from blocking import is_blocked_domain
|
|
|
|
from blocking import is_blocked_hashtag
|
2021-12-29 21:55:09 +00:00
|
|
|
from filters import is_filtered
|
|
|
|
from session import download_image_any_mime_type
|
2020-10-04 09:51:12 +00:00
|
|
|
|
2020-10-16 12:11:05 +00:00
|
|
|
|
2021-12-29 21:55:09 +00:00
|
|
|
def _remove_cdata(text: str) -> str:
|
2020-11-22 12:18:43 +00:00
|
|
|
"""Removes any CDATA from the given text
|
|
|
|
"""
|
|
|
|
if 'CDATA[' in text:
|
|
|
|
text = text.split('CDATA[')[1]
|
|
|
|
if ']' in text:
|
|
|
|
text = text.split(']')[0]
|
|
|
|
return text
|
|
|
|
|
|
|
|
|
2021-12-29 21:55:09 +00:00
|
|
|
def rss2header(http_prefix: str,
|
2021-12-26 10:00:46 +00:00
|
|
|
nickname: str, domain_full: str,
|
2020-10-04 12:29:07 +00:00
|
|
|
title: str, translate: {}) -> str:
|
2020-10-06 09:22:23 +00:00
|
|
|
"""Header for an RSS 2.0 feed
|
|
|
|
"""
|
2022-01-03 12:37:09 +00:00
|
|
|
rss_str = \
|
2021-07-04 09:50:09 +00:00
|
|
|
"<?xml version=\"1.0\" encoding=\"UTF-8\" ?>" + \
|
|
|
|
"<rss version=\"2.0\">" + \
|
|
|
|
'<channel>'
|
2020-10-13 17:14:57 +00:00
|
|
|
|
2020-10-04 12:29:07 +00:00
|
|
|
if title.startswith('News'):
|
2022-01-03 12:37:09 +00:00
|
|
|
rss_str += \
|
2021-07-04 09:50:09 +00:00
|
|
|
' <title>Newswire</title>' + \
|
2021-12-26 10:00:46 +00:00
|
|
|
' <link>' + http_prefix + '://' + domain_full + \
|
2020-10-04 12:29:07 +00:00
|
|
|
'/newswire.xml' + '</link>'
|
2020-10-13 17:14:57 +00:00
|
|
|
elif title.startswith('Site'):
|
2022-01-03 12:37:09 +00:00
|
|
|
rss_str += \
|
2021-12-26 10:00:46 +00:00
|
|
|
' <title>' + domain_full + '</title>' + \
|
|
|
|
' <link>' + http_prefix + '://' + domain_full + \
|
2020-10-13 17:17:17 +00:00
|
|
|
'/blog/rss.xml' + '</link>'
|
2020-10-04 12:29:07 +00:00
|
|
|
else:
|
2022-01-03 12:37:09 +00:00
|
|
|
rss_str += \
|
2021-07-04 09:50:09 +00:00
|
|
|
' <title>' + translate[title] + '</title>' + \
|
2021-08-14 11:13:39 +00:00
|
|
|
' <link>' + \
|
2021-12-26 10:19:59 +00:00
|
|
|
local_actor_url(http_prefix, nickname, domain_full) + \
|
2021-08-14 11:13:39 +00:00
|
|
|
'/rss.xml' + '</link>'
|
2022-01-03 12:37:09 +00:00
|
|
|
return rss_str
|
2020-10-04 12:29:07 +00:00
|
|
|
|
|
|
|
|
2021-12-29 21:55:09 +00:00
|
|
|
def rss2footer() -> str:
|
2020-10-06 09:22:23 +00:00
|
|
|
"""Footer for an RSS 2.0 feed
|
|
|
|
"""
|
2022-01-03 12:37:09 +00:00
|
|
|
rss_str = '</channel></rss>'
|
|
|
|
return rss_str
|
2020-10-04 12:29:07 +00:00
|
|
|
|
|
|
|
|
2022-01-03 12:37:09 +00:00
|
|
|
def get_newswire_tags(text: str, max_tags: int) -> []:
|
2020-10-16 19:49:34 +00:00
|
|
|
"""Returns a list of hashtags found in the given text
|
|
|
|
"""
|
2020-10-16 20:46:34 +00:00
|
|
|
if '#' not in text:
|
|
|
|
return []
|
2020-10-16 19:49:34 +00:00
|
|
|
if ' ' not in text:
|
|
|
|
return []
|
2022-01-03 12:37:09 +00:00
|
|
|
text_simplified = \
|
2020-10-16 19:49:34 +00:00
|
|
|
text.replace(',', ' ').replace(';', ' ').replace('- ', ' ')
|
2022-01-03 12:37:09 +00:00
|
|
|
text_simplified = text_simplified.replace('. ', ' ').strip()
|
|
|
|
if text_simplified.endswith('.'):
|
|
|
|
text_simplified = text_simplified[:len(text_simplified)-1]
|
|
|
|
words = text_simplified.split(' ')
|
2020-10-16 19:49:34 +00:00
|
|
|
tags = []
|
|
|
|
for wrd in words:
|
2021-07-07 14:00:53 +00:00
|
|
|
if not wrd.startswith('#'):
|
|
|
|
continue
|
|
|
|
if len(wrd) <= 1:
|
|
|
|
continue
|
|
|
|
if wrd in tags:
|
|
|
|
continue
|
|
|
|
tags.append(wrd)
|
2022-01-03 12:37:09 +00:00
|
|
|
if len(tags) >= max_tags:
|
2021-07-07 14:00:53 +00:00
|
|
|
break
|
2020-10-16 19:49:34 +00:00
|
|
|
return tags
|
|
|
|
|
|
|
|
|
2022-06-09 14:58:47 +00:00
|
|
|
def limit_word_lengths(text: str, max_word_length: int) -> str:
|
2021-07-07 14:00:53 +00:00
|
|
|
"""Limits the maximum length of words so that the newswire
|
|
|
|
column cannot become too wide
|
|
|
|
"""
|
|
|
|
if ' ' not in text:
|
|
|
|
return text
|
|
|
|
words = text.split(' ')
|
|
|
|
result = ''
|
|
|
|
for wrd in words:
|
2022-06-09 14:58:47 +00:00
|
|
|
if len(wrd) > max_word_length:
|
|
|
|
wrd = wrd[:max_word_length]
|
2021-07-07 14:00:53 +00:00
|
|
|
if result:
|
|
|
|
result += ' '
|
|
|
|
result += wrd
|
|
|
|
return result
|
|
|
|
|
|
|
|
|
2021-12-29 21:55:09 +00:00
|
|
|
def get_newswire_favicon_url(url: str) -> str:
|
2021-12-16 20:57:30 +00:00
|
|
|
"""Returns a favicon url from the given article link
|
|
|
|
"""
|
|
|
|
if '://' not in url:
|
|
|
|
return '/newswire_favicon.ico'
|
|
|
|
if url.startswith('http://'):
|
|
|
|
if not (url.endswith('.onion') or url.endswith('.i2p')):
|
|
|
|
return '/newswire_favicon.ico'
|
|
|
|
domain = url.split('://')[1]
|
|
|
|
if '/' not in domain:
|
|
|
|
return url + '/favicon.ico'
|
2022-01-03 12:37:09 +00:00
|
|
|
domain = domain.split('/')[0]
|
2021-12-16 20:57:30 +00:00
|
|
|
return url.split('://')[0] + '://' + domain + '/favicon.ico'
|
|
|
|
|
|
|
|
|
2021-12-29 21:55:09 +00:00
|
|
|
def _download_newswire_feed_favicon(session, base_dir: str,
|
|
|
|
link: str, debug: bool) -> bool:
|
2021-12-16 20:57:30 +00:00
|
|
|
"""Downloads the favicon for the given feed link
|
|
|
|
"""
|
2022-01-03 12:37:09 +00:00
|
|
|
fav_url = get_newswire_favicon_url(link)
|
2021-12-16 20:57:30 +00:00
|
|
|
if '://' not in link:
|
|
|
|
return False
|
2022-01-03 12:37:09 +00:00
|
|
|
timeout_sec = 10
|
|
|
|
image_data, mime_type = \
|
|
|
|
download_image_any_mime_type(session, fav_url, timeout_sec, debug)
|
|
|
|
if not image_data or not mime_type:
|
2021-12-16 23:59:53 +00:00
|
|
|
return False
|
2021-12-17 12:01:54 +00:00
|
|
|
|
|
|
|
# update the favicon url
|
2022-01-03 12:37:09 +00:00
|
|
|
extensions_to_mime = {
|
2021-12-17 12:01:54 +00:00
|
|
|
'ico': 'x-icon',
|
|
|
|
'png': 'png',
|
|
|
|
'jpg': 'jpeg',
|
2022-02-06 11:04:49 +00:00
|
|
|
'jxl': 'jxl',
|
2021-12-17 12:01:54 +00:00
|
|
|
'gif': 'gif',
|
|
|
|
'avif': 'avif',
|
|
|
|
'svg': 'svg+xml',
|
|
|
|
'webp': 'webp'
|
|
|
|
}
|
2022-01-03 12:37:09 +00:00
|
|
|
for ext, mime_ext in extensions_to_mime.items():
|
|
|
|
if 'image/' + mime_ext in mime_type:
|
|
|
|
fav_url = fav_url.replace('.ico', '.' + ext)
|
2021-12-17 12:01:54 +00:00
|
|
|
break
|
|
|
|
|
|
|
|
# create cached favicons directory if needed
|
2021-12-25 16:17:53 +00:00
|
|
|
if not os.path.isdir(base_dir + '/favicons'):
|
|
|
|
os.mkdir(base_dir + '/favicons')
|
2021-12-17 12:01:54 +00:00
|
|
|
|
2021-12-19 12:32:01 +00:00
|
|
|
# check svg for dubious scripts
|
2022-01-03 12:37:09 +00:00
|
|
|
if fav_url.endswith('.svg'):
|
|
|
|
image_data_str = str(image_data)
|
|
|
|
if dangerous_svg(image_data_str, False):
|
2021-12-19 12:32:01 +00:00
|
|
|
return False
|
|
|
|
|
2021-12-17 12:01:54 +00:00
|
|
|
# save to the cache
|
2022-01-03 12:37:09 +00:00
|
|
|
fav_filename = get_fav_filename_from_url(base_dir, fav_url)
|
|
|
|
if os.path.isfile(fav_filename):
|
2021-12-16 21:14:24 +00:00
|
|
|
return True
|
2021-12-16 20:57:30 +00:00
|
|
|
try:
|
2022-01-03 12:37:09 +00:00
|
|
|
with open(fav_filename, 'wb+') as fp_fav:
|
|
|
|
fp_fav.write(image_data)
|
2021-12-16 20:57:30 +00:00
|
|
|
except OSError:
|
2022-01-03 12:37:09 +00:00
|
|
|
print('EX: failed writing favicon ' + fav_filename)
|
2021-12-16 20:57:30 +00:00
|
|
|
return False
|
2021-12-19 12:32:01 +00:00
|
|
|
|
2021-12-16 20:57:30 +00:00
|
|
|
return True
|
|
|
|
|
|
|
|
|
2021-12-29 21:55:09 +00:00
|
|
|
def _add_newswire_dict_entry(base_dir: str, domain: str,
|
2022-01-03 12:37:09 +00:00
|
|
|
newswire: {}, date_str: str,
|
2021-12-29 21:55:09 +00:00
|
|
|
title: str, link: str,
|
2022-01-03 12:37:09 +00:00
|
|
|
votes_status: str, post_filename: str,
|
2021-12-29 21:55:09 +00:00
|
|
|
description: str, moderated: bool,
|
|
|
|
mirrored: bool,
|
|
|
|
tags: [],
|
2022-01-10 18:48:57 +00:00
|
|
|
max_tags: int, session, debug: bool,
|
|
|
|
podcast_properties: {}) -> None:
|
2020-10-16 19:25:55 +00:00
|
|
|
"""Update the newswire dictionary
|
|
|
|
"""
|
2020-12-12 15:44:43 +00:00
|
|
|
# remove any markup
|
2021-12-27 15:43:22 +00:00
|
|
|
title = remove_html(title)
|
|
|
|
description = remove_html(description)
|
2020-12-12 15:44:43 +00:00
|
|
|
|
2022-01-03 12:37:09 +00:00
|
|
|
all_text = title + ' ' + description
|
2020-10-25 10:17:12 +00:00
|
|
|
|
|
|
|
# check that none of the text is filtered against
|
2022-01-03 12:37:09 +00:00
|
|
|
if is_filtered(base_dir, None, None, all_text):
|
2020-10-17 16:08:07 +00:00
|
|
|
return
|
2020-10-25 10:17:12 +00:00
|
|
|
|
2021-12-29 21:55:09 +00:00
|
|
|
title = limit_word_lengths(title, 13)
|
2021-07-07 14:00:53 +00:00
|
|
|
|
2020-10-25 10:17:12 +00:00
|
|
|
if tags is None:
|
|
|
|
tags = []
|
|
|
|
|
|
|
|
# extract hashtags from the text of the feed post
|
2022-01-03 12:37:09 +00:00
|
|
|
post_tags = get_newswire_tags(all_text, max_tags)
|
2020-10-25 10:17:12 +00:00
|
|
|
|
2022-01-13 15:15:47 +00:00
|
|
|
# Include tags from podcast categories
|
|
|
|
if podcast_properties:
|
2022-01-13 23:06:04 +00:00
|
|
|
if podcast_properties.get('explicit'):
|
2022-01-14 13:15:43 +00:00
|
|
|
if '#nsfw' not in post_tags:
|
|
|
|
post_tags.append('#nsfw')
|
2022-01-13 23:06:04 +00:00
|
|
|
|
2022-01-13 15:15:47 +00:00
|
|
|
post_tags += podcast_properties['categories']
|
|
|
|
|
2020-10-25 10:17:12 +00:00
|
|
|
# combine the tags into a single list
|
2020-10-25 12:57:14 +00:00
|
|
|
for tag in tags:
|
2022-01-03 12:37:09 +00:00
|
|
|
if tag in post_tags:
|
2021-07-07 14:00:53 +00:00
|
|
|
continue
|
2022-01-03 12:37:09 +00:00
|
|
|
if len(post_tags) < max_tags:
|
|
|
|
post_tags.append(tag)
|
2020-10-25 10:17:12 +00:00
|
|
|
|
|
|
|
# check that no tags are blocked
|
2022-01-03 12:37:09 +00:00
|
|
|
for tag in post_tags:
|
2021-12-28 21:55:38 +00:00
|
|
|
if is_blocked_hashtag(base_dir, tag):
|
2020-10-25 10:18:07 +00:00
|
|
|
return
|
2020-10-25 10:17:12 +00:00
|
|
|
|
2021-12-29 21:55:09 +00:00
|
|
|
_download_newswire_feed_favicon(session, base_dir, link, debug)
|
2021-12-16 20:57:30 +00:00
|
|
|
|
2022-01-03 12:37:09 +00:00
|
|
|
newswire[date_str] = [
|
2020-10-25 10:17:12 +00:00
|
|
|
title,
|
|
|
|
link,
|
2022-01-03 12:37:09 +00:00
|
|
|
votes_status,
|
2021-12-26 23:41:34 +00:00
|
|
|
post_filename,
|
2020-10-25 10:17:12 +00:00
|
|
|
description,
|
|
|
|
moderated,
|
2022-01-03 12:37:09 +00:00
|
|
|
post_tags,
|
2022-01-10 18:48:57 +00:00
|
|
|
mirrored,
|
|
|
|
podcast_properties
|
2020-10-25 10:17:12 +00:00
|
|
|
]
|
2020-10-16 19:25:55 +00:00
|
|
|
|
|
|
|
|
2022-01-03 12:37:09 +00:00
|
|
|
def _valid_feed_date(pub_date: str, debug: bool = False) -> bool:
|
2020-12-21 12:11:45 +00:00
|
|
|
# convert from YY-MM-DD HH:MM:SS+00:00 to
|
|
|
|
# YY-MM-DDTHH:MM:SSZ
|
2022-01-03 12:37:09 +00:00
|
|
|
post_date = pub_date.replace(' ', 'T').replace('+00:00', 'Z')
|
2022-03-30 18:13:40 +00:00
|
|
|
if '.' in post_date:
|
|
|
|
ending = post_date.split('.')[1]
|
|
|
|
timezone_str = ''
|
|
|
|
for ending_char in ending:
|
|
|
|
if not ending_char.isdigit():
|
|
|
|
timezone_str += ending_char
|
|
|
|
if timezone_str:
|
|
|
|
post_date = post_date.split('.')[0] + timezone_str
|
2022-01-03 12:37:09 +00:00
|
|
|
return valid_post_date(post_date, 90, debug)
|
2020-12-21 12:11:45 +00:00
|
|
|
|
|
|
|
|
2022-01-27 10:42:46 +00:00
|
|
|
def parse_feed_date(pub_date: str, unique_string_identifier: str) -> str:
|
2020-11-22 19:01:18 +00:00
|
|
|
"""Returns a UTC date string based on the given date string
|
2020-11-22 18:14:40 +00:00
|
|
|
This tries a number of formats to see which work
|
|
|
|
"""
|
2022-01-27 10:42:46 +00:00
|
|
|
|
|
|
|
if ':00:00' in pub_date:
|
|
|
|
# If this was published exactly on the hour then assign a
|
|
|
|
# random minute and second to make this item relatively unique
|
|
|
|
randgen = random.Random(unique_string_identifier)
|
|
|
|
rand_min = randgen.randint(0, 59)
|
|
|
|
rand_sec = randgen.randint(0, 59)
|
|
|
|
replace_time_str = \
|
|
|
|
':' + str(rand_min).zfill(2) + ':' + str(rand_sec).zfill(2)
|
|
|
|
pub_date = pub_date.replace(':00:00', replace_time_str)
|
|
|
|
|
2020-11-22 18:14:40 +00:00
|
|
|
formats = ("%a, %d %b %Y %H:%M:%S %z",
|
2021-10-17 14:17:42 +00:00
|
|
|
"%a, %d %b %Y %H:%M:%S Z",
|
2021-09-07 19:09:41 +00:00
|
|
|
"%a, %d %b %Y %H:%M:%S GMT",
|
2021-10-17 14:24:21 +00:00
|
|
|
"%a, %d %b %Y %H:%M:%S EST",
|
|
|
|
"%a, %d %b %Y %H:%M:%S PST",
|
|
|
|
"%a, %d %b %Y %H:%M:%S AST",
|
|
|
|
"%a, %d %b %Y %H:%M:%S CST",
|
|
|
|
"%a, %d %b %Y %H:%M:%S MST",
|
|
|
|
"%a, %d %b %Y %H:%M:%S AKST",
|
|
|
|
"%a, %d %b %Y %H:%M:%S HST",
|
|
|
|
"%a, %d %b %Y %H:%M:%S UT",
|
2020-11-22 18:14:40 +00:00
|
|
|
"%Y-%m-%dT%H:%M:%SZ",
|
|
|
|
"%Y-%m-%dT%H:%M:%S%z")
|
2022-01-03 12:37:09 +00:00
|
|
|
published_date = None
|
|
|
|
for date_format in formats:
|
|
|
|
if ',' in pub_date and ',' not in date_format:
|
2020-11-22 18:14:40 +00:00
|
|
|
continue
|
2022-01-03 12:37:09 +00:00
|
|
|
if ',' not in pub_date and ',' in date_format:
|
2020-11-22 18:14:40 +00:00
|
|
|
continue
|
2022-01-03 12:37:09 +00:00
|
|
|
if 'Z' in pub_date and 'Z' not in date_format:
|
2020-11-22 18:14:40 +00:00
|
|
|
continue
|
2022-01-03 12:37:09 +00:00
|
|
|
if 'Z' not in pub_date and 'Z' in date_format:
|
2020-11-22 18:14:40 +00:00
|
|
|
continue
|
2022-01-03 12:37:09 +00:00
|
|
|
if 'EST' not in pub_date and 'EST' in date_format:
|
2020-11-22 18:14:40 +00:00
|
|
|
continue
|
2022-01-03 12:37:09 +00:00
|
|
|
if 'GMT' not in pub_date and 'GMT' in date_format:
|
2021-09-07 19:09:41 +00:00
|
|
|
continue
|
2022-01-03 12:37:09 +00:00
|
|
|
if 'EST' in pub_date and 'EST' not in date_format:
|
2020-11-22 18:14:40 +00:00
|
|
|
continue
|
2022-01-03 12:37:09 +00:00
|
|
|
if 'UT' not in pub_date and 'UT' in date_format:
|
2020-11-22 18:14:40 +00:00
|
|
|
continue
|
2022-01-03 12:37:09 +00:00
|
|
|
if 'UT' in pub_date and 'UT' not in date_format:
|
2020-11-22 18:14:40 +00:00
|
|
|
continue
|
|
|
|
|
2022-03-30 18:13:40 +00:00
|
|
|
# remove any fraction of a second
|
|
|
|
if '.' in pub_date:
|
|
|
|
ending = pub_date.split('.')[1]
|
|
|
|
timezone_str = ''
|
|
|
|
for ending_char in ending:
|
|
|
|
if not ending_char.isdigit():
|
|
|
|
timezone_str += ending_char
|
|
|
|
if timezone_str:
|
|
|
|
pub_date = pub_date.split('.')[0] + timezone_str
|
2020-11-22 18:14:40 +00:00
|
|
|
try:
|
2022-01-03 12:37:09 +00:00
|
|
|
published_date = datetime.strptime(pub_date, date_format)
|
2020-11-22 18:14:40 +00:00
|
|
|
except BaseException:
|
|
|
|
continue
|
|
|
|
|
2022-01-03 12:37:09 +00:00
|
|
|
if published_date:
|
|
|
|
if pub_date.endswith(' EST'):
|
|
|
|
hours_added = timedelta(hours=5)
|
|
|
|
published_date = published_date + hours_added
|
2020-11-22 18:14:40 +00:00
|
|
|
break
|
2020-11-22 19:01:18 +00:00
|
|
|
|
2022-01-03 12:37:09 +00:00
|
|
|
pub_date_str = None
|
|
|
|
if published_date:
|
|
|
|
offset = published_date.utcoffset()
|
2020-11-22 20:37:08 +00:00
|
|
|
if offset:
|
2022-01-03 12:37:09 +00:00
|
|
|
published_date = published_date - offset
|
2020-11-22 19:09:35 +00:00
|
|
|
# convert local date to UTC
|
2022-01-03 12:37:09 +00:00
|
|
|
published_date = published_date.replace(tzinfo=timezone.utc)
|
|
|
|
pub_date_str = str(published_date)
|
|
|
|
if not pub_date_str.endswith('+00:00'):
|
|
|
|
pub_date_str += '+00:00'
|
2021-09-07 19:33:27 +00:00
|
|
|
else:
|
2022-01-03 12:37:09 +00:00
|
|
|
print('WARN: unrecognized date format: ' + pub_date)
|
2020-11-22 19:01:18 +00:00
|
|
|
|
2022-01-03 12:37:09 +00:00
|
|
|
return pub_date_str
|
2020-11-22 18:14:40 +00:00
|
|
|
|
|
|
|
|
2021-12-29 21:55:09 +00:00
|
|
|
def load_hashtag_categories(base_dir: str, language: str) -> None:
|
2020-12-05 13:38:07 +00:00
|
|
|
"""Loads an rss file containing hashtag categories
|
|
|
|
"""
|
2022-01-03 12:37:09 +00:00
|
|
|
hashtag_categories_filename = base_dir + '/categories.xml'
|
|
|
|
if not os.path.isfile(hashtag_categories_filename):
|
|
|
|
hashtag_categories_filename = \
|
2021-12-25 16:17:53 +00:00
|
|
|
base_dir + '/defaultcategories/' + language + '.xml'
|
2022-01-03 12:37:09 +00:00
|
|
|
if not os.path.isfile(hashtag_categories_filename):
|
2020-12-05 13:38:07 +00:00
|
|
|
return
|
|
|
|
|
2022-06-09 14:46:30 +00:00
|
|
|
with open(hashtag_categories_filename, 'r', encoding='utf-8') as fp_cat:
|
2022-01-03 12:37:09 +00:00
|
|
|
xml_str = fp_cat.read()
|
|
|
|
_xml2str_to_hashtag_categories(base_dir, xml_str, 1024, True)
|
2020-12-05 13:38:07 +00:00
|
|
|
|
|
|
|
|
2022-01-03 12:37:09 +00:00
|
|
|
def _xml2str_to_hashtag_categories(base_dir: str, xml_str: str,
|
2022-06-09 14:58:47 +00:00
|
|
|
max_categories_feed_item_size_kb: int,
|
2021-12-29 21:55:09 +00:00
|
|
|
force: bool = False) -> None:
|
2020-12-02 16:18:36 +00:00
|
|
|
"""Updates hashtag categories based upon an rss feed
|
|
|
|
"""
|
2022-01-03 12:37:09 +00:00
|
|
|
rss_items = xml_str.split('<item>')
|
2022-06-09 14:58:47 +00:00
|
|
|
max_bytes = max_categories_feed_item_size_kb * 1024
|
2022-01-03 12:37:09 +00:00
|
|
|
for rss_item in rss_items:
|
|
|
|
if not rss_item:
|
2020-12-02 16:18:36 +00:00
|
|
|
continue
|
2022-01-03 12:37:09 +00:00
|
|
|
if len(rss_item) > max_bytes:
|
2020-12-02 16:18:36 +00:00
|
|
|
print('WARN: rss categories feed item is too big')
|
|
|
|
continue
|
2022-01-03 12:37:09 +00:00
|
|
|
if '<title>' not in rss_item:
|
2020-12-02 16:18:36 +00:00
|
|
|
continue
|
2022-01-03 12:37:09 +00:00
|
|
|
if '</title>' not in rss_item:
|
2020-12-02 16:18:36 +00:00
|
|
|
continue
|
2022-01-03 12:37:09 +00:00
|
|
|
if '<description>' not in rss_item:
|
2020-12-02 16:18:36 +00:00
|
|
|
continue
|
2022-01-03 12:37:09 +00:00
|
|
|
if '</description>' not in rss_item:
|
2020-12-02 16:18:36 +00:00
|
|
|
continue
|
2022-01-03 12:37:09 +00:00
|
|
|
category_str = rss_item.split('<title>')[1]
|
|
|
|
category_str = category_str.split('</title>')[0].strip()
|
|
|
|
if not category_str:
|
2020-12-02 16:18:36 +00:00
|
|
|
continue
|
2022-01-03 12:37:09 +00:00
|
|
|
if 'CDATA' in category_str:
|
2020-12-03 10:12:09 +00:00
|
|
|
continue
|
2022-01-03 12:37:09 +00:00
|
|
|
hashtag_list_str = rss_item.split('<description>')[1]
|
|
|
|
hashtag_list_str = hashtag_list_str.split('</description>')[0].strip()
|
|
|
|
if not hashtag_list_str:
|
2020-12-02 16:18:36 +00:00
|
|
|
continue
|
2022-01-03 12:37:09 +00:00
|
|
|
if 'CDATA' in hashtag_list_str:
|
2020-12-03 10:12:09 +00:00
|
|
|
continue
|
2022-01-03 12:37:09 +00:00
|
|
|
hashtag_list = hashtag_list_str.split(' ')
|
|
|
|
if not is_blocked_hashtag(base_dir, category_str):
|
|
|
|
for hashtag in hashtag_list:
|
|
|
|
set_hashtag_category(base_dir, hashtag, category_str,
|
2021-12-29 21:55:09 +00:00
|
|
|
False, force)
|
2020-12-02 16:18:36 +00:00
|
|
|
|
|
|
|
|
2022-01-13 15:30:55 +00:00
|
|
|
def _get_podcast_categories(xml_item: str, xml_str: str) -> str:
|
|
|
|
""" get podcast categories if they exist. These can be turned into hashtags
|
|
|
|
"""
|
|
|
|
podcast_categories = []
|
|
|
|
episode_category_tags = ['<itunes:category', '<category']
|
|
|
|
|
|
|
|
for category_tag in episode_category_tags:
|
|
|
|
item_str = xml_item
|
|
|
|
if category_tag not in xml_item:
|
|
|
|
if category_tag not in xml_str:
|
|
|
|
continue
|
|
|
|
item_str = xml_str
|
|
|
|
|
2022-01-13 16:04:14 +00:00
|
|
|
category_list = item_str.split(category_tag)
|
|
|
|
first_category = True
|
2022-01-13 16:12:55 +00:00
|
|
|
for episode_category in category_list:
|
2022-01-13 16:04:14 +00:00
|
|
|
if first_category:
|
|
|
|
first_category = False
|
|
|
|
continue
|
|
|
|
|
|
|
|
if 'text="' in episode_category:
|
|
|
|
episode_category = episode_category.split('text="')[1]
|
|
|
|
if '"' in episode_category:
|
|
|
|
episode_category = episode_category.split('"')[0]
|
|
|
|
episode_category = \
|
|
|
|
episode_category.lower().replace(' ', '')
|
|
|
|
episode_category = episode_category.replace('#', '')
|
|
|
|
if episode_category not in podcast_categories:
|
|
|
|
if valid_hash_tag(episode_category):
|
|
|
|
podcast_categories.append('#' + episode_category)
|
|
|
|
continue
|
2022-01-13 15:30:55 +00:00
|
|
|
|
2022-01-13 16:04:14 +00:00
|
|
|
if '>' in episode_category:
|
|
|
|
episode_category = episode_category.split('>')[1]
|
|
|
|
if '<' in episode_category:
|
|
|
|
episode_category = episode_category.split('<')[0]
|
|
|
|
episode_category = \
|
|
|
|
episode_category.lower().replace(' ', '')
|
|
|
|
episode_category = episode_category.replace('#', '')
|
|
|
|
if episode_category not in podcast_categories:
|
|
|
|
if valid_hash_tag(episode_category):
|
|
|
|
podcast_categories.append('#' + episode_category)
|
2022-01-13 15:30:55 +00:00
|
|
|
|
|
|
|
return podcast_categories
|
|
|
|
|
|
|
|
|
2022-02-12 20:37:15 +00:00
|
|
|
def _valid_podcast_entry(base_dir: str, key: str, entry: {}) -> bool:
|
|
|
|
"""Is the given podcast namespace entry valid?
|
|
|
|
https://github.com/Podcastindex-org/podcast-namespace/
|
|
|
|
blob/main/proposal-docs/social/social.md#socialinteract-element
|
|
|
|
"""
|
2022-05-30 15:15:17 +00:00
|
|
|
if key in ('socialInteract', 'discussion'):
|
2022-02-12 20:37:15 +00:00
|
|
|
if not entry.get('protocol'):
|
|
|
|
return False
|
2022-04-21 09:21:25 +00:00
|
|
|
if not entry.get('uri'):
|
|
|
|
if not entry.get('text'):
|
|
|
|
if not entry.get('url'):
|
|
|
|
return False
|
2022-02-12 20:37:15 +00:00
|
|
|
if entry['protocol'].tolower() != 'activitypub':
|
|
|
|
return False
|
2022-04-21 09:21:25 +00:00
|
|
|
if entry.get('uri'):
|
|
|
|
post_url = entry['uri']
|
|
|
|
elif entry.get('url'):
|
|
|
|
post_url = entry['uri']
|
|
|
|
else:
|
|
|
|
post_url = entry['text']
|
2022-02-12 20:37:15 +00:00
|
|
|
if '://' not in post_url:
|
|
|
|
return False
|
2022-06-09 14:58:47 +00:00
|
|
|
post_domain, _ = get_domain_from_actor(post_url)
|
2022-02-12 20:37:15 +00:00
|
|
|
if not post_domain:
|
|
|
|
return False
|
|
|
|
if is_blocked_domain(base_dir, post_domain):
|
|
|
|
return False
|
|
|
|
return True
|
|
|
|
|
|
|
|
|
|
|
|
def xml_podcast_to_dict(base_dir: str, xml_item: str, xml_str: str) -> {}:
|
2022-01-10 18:48:57 +00:00
|
|
|
"""podcasting extensions for RSS feeds
|
2022-01-10 19:07:16 +00:00
|
|
|
See https://github.com/Podcastindex-org/podcast-namespace/
|
|
|
|
blob/main/docs/1.0.md
|
2022-02-12 15:40:55 +00:00
|
|
|
https://github.com/Podcastindex-org/podcast-namespace/
|
|
|
|
blob/main/proposal-docs/social/social.md#socialinteract-element
|
2022-01-10 18:48:57 +00:00
|
|
|
"""
|
2022-01-13 12:37:08 +00:00
|
|
|
if '<podcast:' not in xml_item:
|
|
|
|
if '<itunes:' not in xml_item:
|
2022-01-14 18:05:29 +00:00
|
|
|
if '<media:thumbnail' not in xml_item:
|
|
|
|
return {}
|
2022-01-10 18:48:57 +00:00
|
|
|
|
|
|
|
podcast_properties = {
|
|
|
|
"locations": [],
|
|
|
|
"persons": [],
|
|
|
|
"soundbites": [],
|
|
|
|
"transcripts": [],
|
2022-01-10 19:07:16 +00:00
|
|
|
"valueRecipients": [],
|
2022-02-12 15:38:35 +00:00
|
|
|
"trailers": [],
|
2022-05-03 16:38:16 +00:00
|
|
|
"chapters": [],
|
2022-04-29 13:54:13 +00:00
|
|
|
"discussion": [],
|
|
|
|
"episode": '',
|
|
|
|
"socialInteract": [],
|
2022-01-10 18:48:57 +00:00
|
|
|
}
|
|
|
|
|
2022-01-13 12:37:08 +00:00
|
|
|
pod_lines = xml_item.split('<podcast:')
|
2022-01-10 18:48:57 +00:00
|
|
|
ctr = 0
|
|
|
|
for pod_line in pod_lines:
|
|
|
|
if ctr == 0 or '>' not in pod_line:
|
|
|
|
ctr += 1
|
|
|
|
continue
|
|
|
|
if ' ' not in pod_line.split('>')[0]:
|
|
|
|
pod_key = pod_line.split('>')[0].strip()
|
|
|
|
pod_val = pod_line.split('>', 1)[1].strip()
|
|
|
|
if '<' in pod_val:
|
|
|
|
pod_val = pod_val.split('<')[0]
|
2022-04-29 13:54:13 +00:00
|
|
|
if pod_key in podcast_properties:
|
|
|
|
podcast_properties[pod_key] = pod_val
|
2022-01-10 18:48:57 +00:00
|
|
|
ctr += 1
|
|
|
|
continue
|
|
|
|
pod_key = pod_line.split(' ')[0]
|
|
|
|
|
|
|
|
pod_fields = (
|
|
|
|
'url', 'geo', 'osm', 'type', 'method', 'group',
|
|
|
|
'owner', 'srcset', 'img', 'role', 'address', 'suggested',
|
2022-01-10 19:07:16 +00:00
|
|
|
'startTime', 'duration', 'href', 'name', 'pubdate',
|
2022-02-12 15:38:35 +00:00
|
|
|
'length', 'season', 'email', 'platform', 'protocol',
|
2022-02-16 13:27:11 +00:00
|
|
|
'accountId', 'priority', 'podcastAccountId',
|
|
|
|
'podcastAccountUrl'
|
2022-01-10 18:48:57 +00:00
|
|
|
)
|
|
|
|
pod_entry = {}
|
|
|
|
for pod_field in pod_fields:
|
|
|
|
if pod_field + '="' not in pod_line:
|
|
|
|
continue
|
|
|
|
pod_str = pod_line.split(pod_field + '="')[1]
|
|
|
|
if '"' not in pod_str:
|
|
|
|
continue
|
|
|
|
pod_val = pod_str.split('"')[0]
|
|
|
|
pod_entry[pod_field] = pod_val
|
|
|
|
|
|
|
|
pod_text = pod_line.split('>')[1]
|
|
|
|
if '<' in pod_text:
|
|
|
|
pod_text = pod_text.split('<')[0].strip()
|
|
|
|
if pod_text:
|
|
|
|
pod_entry['text'] = pod_text
|
|
|
|
|
2022-02-12 16:00:45 +00:00
|
|
|
appended = False
|
2022-01-10 18:48:57 +00:00
|
|
|
if pod_key + 's' in podcast_properties:
|
|
|
|
if isinstance(podcast_properties[pod_key + 's'], list):
|
|
|
|
podcast_properties[pod_key + 's'].append(pod_entry)
|
2022-02-12 16:00:45 +00:00
|
|
|
appended = True
|
|
|
|
if not appended:
|
2022-02-12 16:05:44 +00:00
|
|
|
# if there are repeated keys then only use the first one
|
|
|
|
if not podcast_properties.get(pod_key):
|
2022-02-12 20:37:15 +00:00
|
|
|
if _valid_podcast_entry(base_dir, pod_key, pod_entry):
|
|
|
|
podcast_properties[pod_key] = pod_entry
|
2022-01-10 18:48:57 +00:00
|
|
|
ctr += 1
|
|
|
|
|
2022-01-11 18:25:13 +00:00
|
|
|
# get the image for the podcast, if it exists
|
2022-01-12 17:44:49 +00:00
|
|
|
podcast_episode_image = None
|
2022-01-14 17:40:42 +00:00
|
|
|
episode_image_tags = ['<itunes:image', '<media:thumbnail']
|
2022-01-11 18:25:13 +00:00
|
|
|
for image_tag in episode_image_tags:
|
2022-01-13 12:37:08 +00:00
|
|
|
item_str = xml_item
|
|
|
|
if image_tag not in xml_item:
|
|
|
|
if image_tag not in xml_str:
|
|
|
|
continue
|
|
|
|
item_str = xml_str
|
|
|
|
|
|
|
|
episode_image = item_str.split(image_tag)[1]
|
2022-01-14 17:55:56 +00:00
|
|
|
if image_tag + ' ' in item_str and '>' in episode_image:
|
|
|
|
episode_image = episode_image.split('>')[0]
|
|
|
|
|
2022-01-11 18:25:13 +00:00
|
|
|
if 'href="' in episode_image:
|
|
|
|
episode_image = episode_image.split('href="')[1]
|
|
|
|
if '"' in episode_image:
|
|
|
|
episode_image = episode_image.split('"')[0]
|
2022-01-12 17:44:49 +00:00
|
|
|
podcast_episode_image = episode_image
|
2022-01-11 18:25:13 +00:00
|
|
|
break
|
2022-01-14 17:40:42 +00:00
|
|
|
elif 'url="' in episode_image:
|
|
|
|
episode_image = episode_image.split('url="')[1]
|
|
|
|
if '"' in episode_image:
|
|
|
|
episode_image = episode_image.split('"')[0]
|
|
|
|
podcast_episode_image = episode_image
|
|
|
|
break
|
2022-01-14 18:05:29 +00:00
|
|
|
elif '>' in episode_image:
|
|
|
|
episode_image = episode_image.split('>')[1]
|
|
|
|
if '<' in episode_image:
|
|
|
|
episode_image = episode_image.split('<')[0]
|
|
|
|
if '://' in episode_image and '.' in episode_image:
|
|
|
|
podcast_episode_image = episode_image
|
|
|
|
break
|
2022-01-12 17:44:49 +00:00
|
|
|
|
2022-01-13 15:10:41 +00:00
|
|
|
# get categories if they exist. These can be turned into hashtags
|
2022-01-13 15:30:55 +00:00
|
|
|
podcast_categories = _get_podcast_categories(xml_item, xml_str)
|
2022-01-13 15:10:41 +00:00
|
|
|
|
2022-01-12 17:44:49 +00:00
|
|
|
if podcast_episode_image:
|
|
|
|
podcast_properties['image'] = podcast_episode_image
|
2022-01-13 15:10:41 +00:00
|
|
|
podcast_properties['categories'] = podcast_categories
|
2022-01-12 17:44:49 +00:00
|
|
|
|
2022-01-13 12:37:08 +00:00
|
|
|
if '<itunes:explicit>Y' in xml_item or \
|
|
|
|
'<itunes:explicit>T' in xml_item or \
|
|
|
|
'<itunes:explicit>1' in xml_item:
|
2022-01-12 17:44:49 +00:00
|
|
|
podcast_properties['explicit'] = True
|
|
|
|
else:
|
|
|
|
podcast_properties['explicit'] = False
|
|
|
|
else:
|
2022-01-13 12:37:08 +00:00
|
|
|
if '<podcast:' not in xml_item:
|
2022-01-12 17:44:49 +00:00
|
|
|
return {}
|
|
|
|
|
2022-01-10 18:48:57 +00:00
|
|
|
return podcast_properties
|
|
|
|
|
|
|
|
|
2022-04-22 12:53:34 +00:00
|
|
|
def get_link_from_rss_item(rss_item: str,
|
2022-04-22 13:46:42 +00:00
|
|
|
preferred_mime_types: [],
|
|
|
|
proxy_type: str) -> (str, str):
|
2022-01-12 14:23:07 +00:00
|
|
|
"""Extracts rss link from rss item string
|
|
|
|
"""
|
2022-01-12 18:35:15 +00:00
|
|
|
mime_type = None
|
|
|
|
|
2022-04-22 12:53:34 +00:00
|
|
|
if preferred_mime_types and '<podcast:alternateEnclosure ' in rss_item:
|
|
|
|
enclosures = rss_item.split('<podcast:alternateEnclosure ')
|
|
|
|
ctr = 0
|
|
|
|
for enclosure in enclosures:
|
|
|
|
if ctr == 0:
|
|
|
|
ctr += 1
|
|
|
|
continue
|
|
|
|
ctr += 1
|
|
|
|
if '</podcast:alternateEnclosure' not in enclosure:
|
|
|
|
continue
|
|
|
|
enclosure = enclosure.split('</podcast:alternateEnclosure')[0]
|
|
|
|
if 'type="' not in enclosure:
|
|
|
|
continue
|
|
|
|
mime_type = enclosure.split('type="')[1]
|
|
|
|
if '"' in mime_type:
|
|
|
|
mime_type = mime_type.split('"')[0]
|
|
|
|
if mime_type not in preferred_mime_types:
|
|
|
|
continue
|
|
|
|
if 'uri="' not in enclosure:
|
|
|
|
continue
|
|
|
|
uris = enclosure.split('uri="')
|
|
|
|
ctr2 = 0
|
|
|
|
for uri in uris:
|
|
|
|
if ctr2 == 0:
|
|
|
|
ctr2 += 1
|
|
|
|
continue
|
|
|
|
ctr2 += 1
|
|
|
|
if '"' not in uri:
|
|
|
|
continue
|
|
|
|
link = uri.split('"')[0]
|
|
|
|
if '://' not in link:
|
|
|
|
continue
|
|
|
|
if proxy_type:
|
|
|
|
if proxy_type == 'tor' and \
|
|
|
|
'.onion/' not in link:
|
|
|
|
continue
|
|
|
|
if proxy_type == 'onion' and \
|
|
|
|
'.onion/' not in link:
|
|
|
|
continue
|
|
|
|
if proxy_type == 'i2p' and \
|
|
|
|
'.i2p/' not in link:
|
|
|
|
continue
|
|
|
|
return link, mime_type
|
2022-05-30 15:15:17 +00:00
|
|
|
if '.onion/' not in link and \
|
|
|
|
'.i2p/' not in link:
|
|
|
|
return link, mime_type
|
2022-04-22 12:53:34 +00:00
|
|
|
|
2022-01-12 14:23:07 +00:00
|
|
|
if '<enclosure ' in rss_item:
|
|
|
|
# get link from audio or video enclosure
|
|
|
|
enclosure = rss_item.split('<enclosure ')[1]
|
|
|
|
if '>' in enclosure:
|
|
|
|
enclosure = enclosure.split('>')[0]
|
2022-01-12 18:35:15 +00:00
|
|
|
if ' type="' in enclosure:
|
|
|
|
mime_type = enclosure.split(' type="')[1]
|
|
|
|
if '"' in mime_type:
|
|
|
|
mime_type = mime_type.split('"')[0]
|
2022-01-12 14:23:07 +00:00
|
|
|
if 'url="' in enclosure and \
|
|
|
|
('"audio/' in enclosure or '"video/' in enclosure):
|
|
|
|
link_str = enclosure.split('url="')[1]
|
|
|
|
if '"' in link_str:
|
2022-01-12 16:18:54 +00:00
|
|
|
link = link_str.split('"')[0]
|
|
|
|
if '://' in link:
|
2022-01-12 18:35:15 +00:00
|
|
|
return link, mime_type
|
2022-01-12 16:18:54 +00:00
|
|
|
|
2022-01-13 22:26:01 +00:00
|
|
|
if '<link>' in rss_item and '</link>' in rss_item:
|
|
|
|
link = rss_item.split('<link>')[1]
|
|
|
|
link = link.split('</link>')[0]
|
|
|
|
if '://' not in link:
|
|
|
|
return None, None
|
|
|
|
elif '<link ' in rss_item:
|
|
|
|
link_str = rss_item.split('<link ')[1]
|
|
|
|
if '>' in link_str:
|
|
|
|
link_str = link_str.split('>')[0]
|
|
|
|
if 'href="' in link_str:
|
|
|
|
link_str = link_str.split('href="')[1]
|
|
|
|
if '"' in link_str:
|
|
|
|
link = link_str.split('"')[0]
|
|
|
|
|
2022-01-12 18:35:15 +00:00
|
|
|
return link, mime_type
|
2022-01-12 14:23:07 +00:00
|
|
|
|
|
|
|
|
2022-01-03 12:37:09 +00:00
|
|
|
def _xml2str_to_dict(base_dir: str, domain: str, xml_str: str,
|
2021-12-29 21:55:09 +00:00
|
|
|
moderated: bool, mirrored: bool,
|
2022-01-03 12:37:09 +00:00
|
|
|
max_posts_per_source: int,
|
2021-12-29 21:55:09 +00:00
|
|
|
max_feed_item_size_kb: int,
|
2022-06-09 14:58:47 +00:00
|
|
|
max_categories_feed_item_size_kb: int,
|
2022-04-22 13:46:42 +00:00
|
|
|
session, debug: bool,
|
|
|
|
preferred_podcast_formats: []) -> {}:
|
2020-12-14 14:22:44 +00:00
|
|
|
"""Converts an xml RSS 2.0 string to a dictionary
|
2020-10-04 09:51:12 +00:00
|
|
|
"""
|
2022-01-03 12:37:09 +00:00
|
|
|
if '<item>' not in xml_str:
|
2020-10-04 09:51:12 +00:00
|
|
|
return {}
|
|
|
|
result = {}
|
2020-12-09 10:38:09 +00:00
|
|
|
|
|
|
|
# is this an rss feed containing hashtag categories?
|
2022-01-03 12:37:09 +00:00
|
|
|
if '<title>#categories</title>' in xml_str:
|
|
|
|
_xml2str_to_hashtag_categories(base_dir, xml_str,
|
2022-06-09 14:58:47 +00:00
|
|
|
max_categories_feed_item_size_kb)
|
2020-12-02 16:18:36 +00:00
|
|
|
return {}
|
2020-12-09 10:38:09 +00:00
|
|
|
|
2022-01-03 12:37:09 +00:00
|
|
|
rss_items = xml_str.split('<item>')
|
|
|
|
post_ctr = 0
|
|
|
|
max_bytes = max_feed_item_size_kb * 1024
|
2022-01-13 12:16:42 +00:00
|
|
|
first_item = True
|
2022-01-03 12:37:09 +00:00
|
|
|
for rss_item in rss_items:
|
2022-01-13 12:16:42 +00:00
|
|
|
if first_item:
|
2022-01-13 12:19:35 +00:00
|
|
|
first_item = False
|
2022-01-13 12:16:42 +00:00
|
|
|
continue
|
2022-01-03 12:37:09 +00:00
|
|
|
if not rss_item:
|
2020-11-27 22:43:34 +00:00
|
|
|
continue
|
2022-01-03 12:37:09 +00:00
|
|
|
if len(rss_item) > max_bytes:
|
2020-11-03 16:04:25 +00:00
|
|
|
print('WARN: rss feed item is too big')
|
|
|
|
continue
|
2022-01-03 12:37:09 +00:00
|
|
|
if '<title>' not in rss_item:
|
2020-10-04 09:51:12 +00:00
|
|
|
continue
|
2022-01-03 12:37:09 +00:00
|
|
|
if '</title>' not in rss_item:
|
2020-10-04 09:51:12 +00:00
|
|
|
continue
|
2022-01-13 22:26:01 +00:00
|
|
|
if '<link' not in rss_item:
|
2020-10-04 09:51:12 +00:00
|
|
|
continue
|
2022-01-03 12:37:09 +00:00
|
|
|
if '<pubDate>' not in rss_item:
|
2020-10-04 09:51:12 +00:00
|
|
|
continue
|
2022-01-03 12:37:09 +00:00
|
|
|
if '</pubDate>' not in rss_item:
|
2020-10-04 09:51:12 +00:00
|
|
|
continue
|
2022-01-11 18:25:13 +00:00
|
|
|
|
2022-01-03 12:37:09 +00:00
|
|
|
title = rss_item.split('<title>')[1]
|
2021-12-29 21:55:09 +00:00
|
|
|
title = _remove_cdata(title.split('</title>')[0])
|
2021-12-27 15:43:22 +00:00
|
|
|
title = remove_html(title)
|
2022-01-11 18:25:13 +00:00
|
|
|
|
2020-10-07 12:05:49 +00:00
|
|
|
description = ''
|
2022-01-03 12:37:09 +00:00
|
|
|
if '<description>' in rss_item and '</description>' in rss_item:
|
|
|
|
description = rss_item.split('<description>')[1]
|
2021-12-27 15:43:22 +00:00
|
|
|
description = remove_html(description.split('</description>')[0])
|
2020-11-21 23:18:34 +00:00
|
|
|
else:
|
2022-01-03 12:37:09 +00:00
|
|
|
if '<media:description>' in rss_item and \
|
|
|
|
'</media:description>' in rss_item:
|
|
|
|
description = rss_item.split('<media:description>')[1]
|
2020-11-21 23:18:34 +00:00
|
|
|
description = description.split('</media:description>')[0]
|
2021-12-27 15:43:22 +00:00
|
|
|
description = remove_html(description)
|
2022-01-11 18:25:13 +00:00
|
|
|
|
2022-04-22 13:46:42 +00:00
|
|
|
proxy_type = None
|
|
|
|
if domain.endswith('.onion'):
|
|
|
|
proxy_type = 'tor'
|
|
|
|
elif domain.endswith('.i2p'):
|
|
|
|
proxy_type = 'i2p'
|
|
|
|
|
|
|
|
link, link_mime_type = \
|
|
|
|
get_link_from_rss_item(rss_item, preferred_podcast_formats,
|
|
|
|
proxy_type)
|
2022-01-12 14:02:47 +00:00
|
|
|
if not link:
|
2022-01-12 14:23:07 +00:00
|
|
|
continue
|
2022-01-11 18:25:13 +00:00
|
|
|
|
2022-01-12 14:02:47 +00:00
|
|
|
item_domain = link.split('://')[1]
|
2022-01-03 12:37:09 +00:00
|
|
|
if '/' in item_domain:
|
|
|
|
item_domain = item_domain.split('/')[0]
|
2022-01-12 14:02:47 +00:00
|
|
|
|
2022-01-03 12:37:09 +00:00
|
|
|
if is_blocked_domain(base_dir, item_domain):
|
2020-10-16 11:58:31 +00:00
|
|
|
continue
|
2022-01-03 12:37:09 +00:00
|
|
|
pub_date = rss_item.split('<pubDate>')[1]
|
|
|
|
pub_date = pub_date.split('</pubDate>')[0]
|
2020-11-22 18:14:40 +00:00
|
|
|
|
2022-01-27 10:42:46 +00:00
|
|
|
unique_string_identifier = title + ' ' + link
|
|
|
|
pub_date_str = parse_feed_date(pub_date, unique_string_identifier)
|
2022-01-03 12:37:09 +00:00
|
|
|
if pub_date_str:
|
|
|
|
if _valid_feed_date(pub_date_str):
|
2021-12-26 23:41:34 +00:00
|
|
|
post_filename = ''
|
2022-01-03 12:37:09 +00:00
|
|
|
votes_status = []
|
2022-02-12 20:37:15 +00:00
|
|
|
podcast_properties = \
|
|
|
|
xml_podcast_to_dict(base_dir, rss_item, xml_str)
|
2022-01-12 18:35:15 +00:00
|
|
|
if podcast_properties:
|
|
|
|
podcast_properties['linkMimeType'] = link_mime_type
|
2021-12-29 21:55:09 +00:00
|
|
|
_add_newswire_dict_entry(base_dir, domain,
|
2022-01-03 12:37:09 +00:00
|
|
|
result, pub_date_str,
|
2021-12-29 21:55:09 +00:00
|
|
|
title, link,
|
2022-01-03 12:37:09 +00:00
|
|
|
votes_status, post_filename,
|
2021-12-29 21:55:09 +00:00
|
|
|
description, moderated,
|
2022-01-10 18:48:57 +00:00
|
|
|
mirrored, [], 32, session, debug,
|
|
|
|
podcast_properties)
|
2022-01-03 12:37:09 +00:00
|
|
|
post_ctr += 1
|
|
|
|
if post_ctr >= max_posts_per_source:
|
2020-12-21 12:11:45 +00:00
|
|
|
break
|
2022-01-03 12:37:09 +00:00
|
|
|
if post_ctr > 0:
|
2022-01-12 14:31:04 +00:00
|
|
|
print('Added ' + str(post_ctr) + ' rss 2.0 feed items to newswire')
|
2020-10-04 09:51:12 +00:00
|
|
|
return result
|
|
|
|
|
|
|
|
|
2022-01-03 12:37:09 +00:00
|
|
|
def _xml1str_to_dict(base_dir: str, domain: str, xml_str: str,
|
2021-12-29 21:55:09 +00:00
|
|
|
moderated: bool, mirrored: bool,
|
2022-01-03 12:37:09 +00:00
|
|
|
max_posts_per_source: int,
|
2021-12-29 21:55:09 +00:00
|
|
|
max_feed_item_size_kb: int,
|
2022-06-09 14:58:47 +00:00
|
|
|
max_categories_feed_item_size_kb: int,
|
2022-04-22 13:46:42 +00:00
|
|
|
session, debug: bool,
|
|
|
|
preferred_podcast_formats: []) -> {}:
|
2020-12-14 14:22:44 +00:00
|
|
|
"""Converts an xml RSS 1.0 string to a dictionary
|
|
|
|
https://validator.w3.org/feed/docs/rss1.html
|
|
|
|
"""
|
2022-01-03 12:37:09 +00:00
|
|
|
item_str = '<item'
|
|
|
|
if item_str not in xml_str:
|
2020-12-14 14:22:44 +00:00
|
|
|
return {}
|
|
|
|
result = {}
|
|
|
|
|
|
|
|
# is this an rss feed containing hashtag categories?
|
2022-01-03 12:37:09 +00:00
|
|
|
if '<title>#categories</title>' in xml_str:
|
|
|
|
_xml2str_to_hashtag_categories(base_dir, xml_str,
|
2022-06-09 14:58:47 +00:00
|
|
|
max_categories_feed_item_size_kb)
|
2020-12-14 14:22:44 +00:00
|
|
|
return {}
|
|
|
|
|
2022-01-03 12:37:09 +00:00
|
|
|
rss_items = xml_str.split(item_str)
|
|
|
|
post_ctr = 0
|
|
|
|
max_bytes = max_feed_item_size_kb * 1024
|
2022-01-13 12:16:42 +00:00
|
|
|
first_item = True
|
2022-01-03 12:37:09 +00:00
|
|
|
for rss_item in rss_items:
|
2022-01-13 12:16:42 +00:00
|
|
|
if first_item:
|
2022-01-13 12:19:35 +00:00
|
|
|
first_item = False
|
2022-01-13 12:16:42 +00:00
|
|
|
continue
|
2022-01-03 12:37:09 +00:00
|
|
|
if not rss_item:
|
2020-12-14 14:22:44 +00:00
|
|
|
continue
|
2022-01-03 12:37:09 +00:00
|
|
|
if len(rss_item) > max_bytes:
|
2020-12-14 17:18:16 +00:00
|
|
|
print('WARN: rss 1.0 feed item is too big')
|
|
|
|
continue
|
2022-01-03 12:37:09 +00:00
|
|
|
if rss_item.startswith('s>'):
|
2020-12-14 14:22:44 +00:00
|
|
|
continue
|
2022-01-03 12:37:09 +00:00
|
|
|
if '<title>' not in rss_item:
|
2020-12-14 14:22:44 +00:00
|
|
|
continue
|
2022-01-03 12:37:09 +00:00
|
|
|
if '</title>' not in rss_item:
|
2020-12-14 14:22:44 +00:00
|
|
|
continue
|
2022-01-13 22:26:01 +00:00
|
|
|
if '<link' not in rss_item:
|
2020-12-14 14:22:44 +00:00
|
|
|
continue
|
2022-01-03 12:37:09 +00:00
|
|
|
if '<dc:date>' not in rss_item:
|
2020-12-14 14:22:44 +00:00
|
|
|
continue
|
2022-01-03 12:37:09 +00:00
|
|
|
if '</dc:date>' not in rss_item:
|
2020-12-14 14:22:44 +00:00
|
|
|
continue
|
2022-01-03 12:37:09 +00:00
|
|
|
title = rss_item.split('<title>')[1]
|
2021-12-29 21:55:09 +00:00
|
|
|
title = _remove_cdata(title.split('</title>')[0])
|
2021-12-27 15:43:22 +00:00
|
|
|
title = remove_html(title)
|
2020-12-14 14:22:44 +00:00
|
|
|
description = ''
|
2022-01-03 12:37:09 +00:00
|
|
|
if '<description>' in rss_item and '</description>' in rss_item:
|
|
|
|
description = rss_item.split('<description>')[1]
|
2021-12-27 15:43:22 +00:00
|
|
|
description = remove_html(description.split('</description>')[0])
|
2020-12-14 14:22:44 +00:00
|
|
|
else:
|
2022-01-03 12:37:09 +00:00
|
|
|
if '<media:description>' in rss_item and \
|
|
|
|
'</media:description>' in rss_item:
|
|
|
|
description = rss_item.split('<media:description>')[1]
|
2020-12-14 14:22:44 +00:00
|
|
|
description = description.split('</media:description>')[0]
|
2021-12-27 15:43:22 +00:00
|
|
|
description = remove_html(description)
|
2022-01-12 14:02:47 +00:00
|
|
|
|
2022-04-22 13:46:42 +00:00
|
|
|
proxy_type = None
|
|
|
|
if domain.endswith('.onion'):
|
|
|
|
proxy_type = 'tor'
|
|
|
|
elif domain.endswith('.i2p'):
|
|
|
|
proxy_type = 'i2p'
|
|
|
|
|
|
|
|
link, link_mime_type = \
|
|
|
|
get_link_from_rss_item(rss_item, preferred_podcast_formats,
|
|
|
|
proxy_type)
|
2022-01-12 14:02:47 +00:00
|
|
|
if not link:
|
2022-01-12 14:23:07 +00:00
|
|
|
continue
|
2022-01-12 14:02:47 +00:00
|
|
|
|
2022-01-03 12:37:09 +00:00
|
|
|
item_domain = link.split('://')[1]
|
|
|
|
if '/' in item_domain:
|
|
|
|
item_domain = item_domain.split('/')[0]
|
2022-01-12 14:02:47 +00:00
|
|
|
|
2022-01-03 12:37:09 +00:00
|
|
|
if is_blocked_domain(base_dir, item_domain):
|
2020-12-14 14:22:44 +00:00
|
|
|
continue
|
2022-01-03 12:37:09 +00:00
|
|
|
pub_date = rss_item.split('<dc:date>')[1]
|
|
|
|
pub_date = pub_date.split('</dc:date>')[0]
|
2020-12-14 14:22:44 +00:00
|
|
|
|
2022-01-27 10:42:46 +00:00
|
|
|
unique_string_identifier = title + ' ' + link
|
|
|
|
pub_date_str = parse_feed_date(pub_date, unique_string_identifier)
|
2022-01-03 12:37:09 +00:00
|
|
|
if pub_date_str:
|
|
|
|
if _valid_feed_date(pub_date_str):
|
2021-12-26 23:41:34 +00:00
|
|
|
post_filename = ''
|
2022-01-03 12:37:09 +00:00
|
|
|
votes_status = []
|
2022-02-12 20:37:15 +00:00
|
|
|
podcast_properties = \
|
|
|
|
xml_podcast_to_dict(base_dir, rss_item, xml_str)
|
2022-01-12 18:35:15 +00:00
|
|
|
if podcast_properties:
|
|
|
|
podcast_properties['linkMimeType'] = link_mime_type
|
2021-12-29 21:55:09 +00:00
|
|
|
_add_newswire_dict_entry(base_dir, domain,
|
2022-01-03 12:37:09 +00:00
|
|
|
result, pub_date_str,
|
2021-12-29 21:55:09 +00:00
|
|
|
title, link,
|
2022-01-03 12:37:09 +00:00
|
|
|
votes_status, post_filename,
|
2021-12-29 21:55:09 +00:00
|
|
|
description, moderated,
|
2022-01-10 18:48:57 +00:00
|
|
|
mirrored, [], 32, session, debug,
|
|
|
|
podcast_properties)
|
2022-01-03 12:37:09 +00:00
|
|
|
post_ctr += 1
|
|
|
|
if post_ctr >= max_posts_per_source:
|
2020-12-21 12:11:45 +00:00
|
|
|
break
|
2022-01-03 12:37:09 +00:00
|
|
|
if post_ctr > 0:
|
2022-01-12 14:31:04 +00:00
|
|
|
print('Added ' + str(post_ctr) + ' rss 1.0 feed items to newswire')
|
2020-12-14 14:22:44 +00:00
|
|
|
return result
|
|
|
|
|
|
|
|
|
2022-01-03 12:37:09 +00:00
|
|
|
def _atom_feed_to_dict(base_dir: str, domain: str, xml_str: str,
|
2021-12-29 21:55:09 +00:00
|
|
|
moderated: bool, mirrored: bool,
|
2022-01-03 12:37:09 +00:00
|
|
|
max_posts_per_source: int,
|
2021-12-29 21:55:09 +00:00
|
|
|
max_feed_item_size_kb: int,
|
2022-04-22 13:46:42 +00:00
|
|
|
session, debug: bool,
|
|
|
|
preferred_podcast_formats: []) -> {}:
|
2020-10-10 12:24:14 +00:00
|
|
|
"""Converts an atom feed string to a dictionary
|
|
|
|
"""
|
2022-01-03 12:37:09 +00:00
|
|
|
if '<entry>' not in xml_str:
|
2020-10-10 12:24:14 +00:00
|
|
|
return {}
|
|
|
|
result = {}
|
2022-01-03 12:37:09 +00:00
|
|
|
atom_items = xml_str.split('<entry>')
|
|
|
|
post_ctr = 0
|
|
|
|
max_bytes = max_feed_item_size_kb * 1024
|
2022-01-13 12:16:42 +00:00
|
|
|
first_item = True
|
2022-01-03 12:37:09 +00:00
|
|
|
for atom_item in atom_items:
|
2022-01-13 12:16:42 +00:00
|
|
|
if first_item:
|
2022-01-13 12:19:35 +00:00
|
|
|
first_item = False
|
2022-01-13 12:16:42 +00:00
|
|
|
continue
|
2022-01-03 12:37:09 +00:00
|
|
|
if not atom_item:
|
2020-11-27 22:43:34 +00:00
|
|
|
continue
|
2022-01-03 12:37:09 +00:00
|
|
|
if len(atom_item) > max_bytes:
|
2020-11-03 16:04:25 +00:00
|
|
|
print('WARN: atom feed item is too big')
|
|
|
|
continue
|
2022-01-03 12:37:09 +00:00
|
|
|
if '<title>' not in atom_item:
|
2020-10-10 12:24:14 +00:00
|
|
|
continue
|
2022-01-03 12:37:09 +00:00
|
|
|
if '</title>' not in atom_item:
|
2020-10-10 12:24:14 +00:00
|
|
|
continue
|
2022-01-13 22:26:01 +00:00
|
|
|
if '<link' not in atom_item:
|
2020-10-10 12:24:14 +00:00
|
|
|
continue
|
2022-01-03 12:37:09 +00:00
|
|
|
if '<updated>' not in atom_item:
|
2020-10-10 12:24:14 +00:00
|
|
|
continue
|
2022-01-03 12:37:09 +00:00
|
|
|
if '</updated>' not in atom_item:
|
2020-10-10 12:24:14 +00:00
|
|
|
continue
|
2022-01-03 12:37:09 +00:00
|
|
|
title = atom_item.split('<title>')[1]
|
2021-12-29 21:55:09 +00:00
|
|
|
title = _remove_cdata(title.split('</title>')[0])
|
2021-12-27 15:43:22 +00:00
|
|
|
title = remove_html(title)
|
2020-10-10 12:24:14 +00:00
|
|
|
description = ''
|
2022-01-03 12:37:09 +00:00
|
|
|
if '<summary>' in atom_item and '</summary>' in atom_item:
|
|
|
|
description = atom_item.split('<summary>')[1]
|
2021-12-27 15:43:22 +00:00
|
|
|
description = remove_html(description.split('</summary>')[0])
|
2020-11-21 23:29:46 +00:00
|
|
|
else:
|
2022-01-03 12:37:09 +00:00
|
|
|
if '<media:description>' in atom_item and \
|
|
|
|
'</media:description>' in atom_item:
|
|
|
|
description = atom_item.split('<media:description>')[1]
|
2020-11-21 23:29:46 +00:00
|
|
|
description = description.split('</media:description>')[0]
|
2021-12-27 15:43:22 +00:00
|
|
|
description = remove_html(description)
|
2022-01-12 14:02:47 +00:00
|
|
|
|
2022-04-22 13:46:42 +00:00
|
|
|
proxy_type = None
|
|
|
|
if domain.endswith('.onion'):
|
|
|
|
proxy_type = 'tor'
|
|
|
|
elif domain.endswith('.i2p'):
|
|
|
|
proxy_type = 'i2p'
|
|
|
|
|
|
|
|
link, link_mime_type = \
|
|
|
|
get_link_from_rss_item(atom_item, preferred_podcast_formats,
|
|
|
|
proxy_type)
|
2022-01-12 14:02:47 +00:00
|
|
|
if not link:
|
2022-01-12 14:23:07 +00:00
|
|
|
continue
|
2022-01-12 14:02:47 +00:00
|
|
|
|
2022-01-03 12:37:09 +00:00
|
|
|
item_domain = link.split('://')[1]
|
|
|
|
if '/' in item_domain:
|
|
|
|
item_domain = item_domain.split('/')[0]
|
2022-01-12 14:02:47 +00:00
|
|
|
|
2022-01-03 12:37:09 +00:00
|
|
|
if is_blocked_domain(base_dir, item_domain):
|
2020-10-16 11:58:31 +00:00
|
|
|
continue
|
2022-01-03 12:37:09 +00:00
|
|
|
pub_date = atom_item.split('<updated>')[1]
|
|
|
|
pub_date = pub_date.split('</updated>')[0]
|
2020-11-22 18:14:40 +00:00
|
|
|
|
2022-01-27 10:42:46 +00:00
|
|
|
unique_string_identifier = title + ' ' + link
|
|
|
|
pub_date_str = parse_feed_date(pub_date, unique_string_identifier)
|
2022-01-03 12:37:09 +00:00
|
|
|
if pub_date_str:
|
|
|
|
if _valid_feed_date(pub_date_str):
|
2021-12-26 23:41:34 +00:00
|
|
|
post_filename = ''
|
2022-01-03 12:37:09 +00:00
|
|
|
votes_status = []
|
2022-02-12 20:37:15 +00:00
|
|
|
podcast_properties = \
|
|
|
|
xml_podcast_to_dict(base_dir, atom_item, xml_str)
|
2022-01-12 18:35:15 +00:00
|
|
|
if podcast_properties:
|
|
|
|
podcast_properties['linkMimeType'] = link_mime_type
|
2021-12-29 21:55:09 +00:00
|
|
|
_add_newswire_dict_entry(base_dir, domain,
|
2022-01-03 12:37:09 +00:00
|
|
|
result, pub_date_str,
|
2021-12-29 21:55:09 +00:00
|
|
|
title, link,
|
2022-01-03 12:37:09 +00:00
|
|
|
votes_status, post_filename,
|
2021-12-29 21:55:09 +00:00
|
|
|
description, moderated,
|
2022-01-10 18:48:57 +00:00
|
|
|
mirrored, [], 32, session, debug,
|
|
|
|
podcast_properties)
|
2022-01-03 12:37:09 +00:00
|
|
|
post_ctr += 1
|
|
|
|
if post_ctr >= max_posts_per_source:
|
2021-02-12 11:28:00 +00:00
|
|
|
break
|
2022-01-03 12:37:09 +00:00
|
|
|
if post_ctr > 0:
|
2022-01-12 14:31:04 +00:00
|
|
|
print('Added ' + str(post_ctr) + ' atom feed items to newswire')
|
2021-02-12 11:28:00 +00:00
|
|
|
return result
|
|
|
|
|
|
|
|
|
2022-01-03 12:37:09 +00:00
|
|
|
def _json_feed_v1to_dict(base_dir: str, domain: str, xml_str: str,
|
2021-12-29 21:55:09 +00:00
|
|
|
moderated: bool, mirrored: bool,
|
2022-01-03 12:37:09 +00:00
|
|
|
max_posts_per_source: int,
|
2021-12-29 21:55:09 +00:00
|
|
|
max_feed_item_size_kb: int,
|
|
|
|
session, debug: bool) -> {}:
|
2021-02-12 11:28:00 +00:00
|
|
|
"""Converts a json feed string to a dictionary
|
2021-02-12 11:46:26 +00:00
|
|
|
See https://jsonfeed.org/version/1.1
|
2021-02-12 11:28:00 +00:00
|
|
|
"""
|
2022-01-03 12:37:09 +00:00
|
|
|
if '"items"' not in xml_str:
|
2021-02-12 11:28:00 +00:00
|
|
|
return {}
|
|
|
|
try:
|
2022-01-03 12:37:09 +00:00
|
|
|
feed_json = json.loads(xml_str)
|
2021-02-12 11:28:00 +00:00
|
|
|
except BaseException:
|
2022-01-03 12:37:09 +00:00
|
|
|
print('EX: _json_feed_v1to_dict unable to load json ' + str(xml_str))
|
2021-02-12 11:28:00 +00:00
|
|
|
return {}
|
2022-01-03 12:37:09 +00:00
|
|
|
max_bytes = max_feed_item_size_kb * 1024
|
|
|
|
if not feed_json.get('version'):
|
2021-02-12 11:28:00 +00:00
|
|
|
return {}
|
2022-01-03 12:37:09 +00:00
|
|
|
if not feed_json['version'].startswith('https://jsonfeed.org/version/1'):
|
2021-02-12 11:28:00 +00:00
|
|
|
return {}
|
2022-01-03 12:37:09 +00:00
|
|
|
if not feed_json.get('items'):
|
2021-02-12 11:28:00 +00:00
|
|
|
return {}
|
2022-01-03 12:37:09 +00:00
|
|
|
if not isinstance(feed_json['items'], list):
|
2021-02-12 11:28:00 +00:00
|
|
|
return {}
|
2022-01-03 12:37:09 +00:00
|
|
|
post_ctr = 0
|
2021-02-12 11:47:49 +00:00
|
|
|
result = {}
|
2022-01-03 12:37:09 +00:00
|
|
|
for json_feed_item in feed_json['items']:
|
|
|
|
if not json_feed_item:
|
2021-02-12 11:28:00 +00:00
|
|
|
continue
|
2022-01-03 12:37:09 +00:00
|
|
|
if not isinstance(json_feed_item, dict):
|
2021-02-12 11:28:00 +00:00
|
|
|
continue
|
2022-01-03 12:37:09 +00:00
|
|
|
if not json_feed_item.get('url'):
|
2021-02-12 11:28:00 +00:00
|
|
|
continue
|
2022-01-03 12:37:09 +00:00
|
|
|
if not isinstance(json_feed_item['url'], str):
|
2021-02-12 11:28:00 +00:00
|
|
|
continue
|
2022-01-03 12:37:09 +00:00
|
|
|
if not json_feed_item.get('date_published'):
|
|
|
|
if not json_feed_item.get('date_modified'):
|
2021-02-12 11:28:00 +00:00
|
|
|
continue
|
2022-01-03 12:37:09 +00:00
|
|
|
if not json_feed_item.get('content_text'):
|
|
|
|
if not json_feed_item.get('content_html'):
|
2021-02-12 11:28:00 +00:00
|
|
|
continue
|
2022-01-03 12:37:09 +00:00
|
|
|
if json_feed_item.get('content_html'):
|
|
|
|
if not isinstance(json_feed_item['content_html'], str):
|
2021-02-12 11:28:00 +00:00
|
|
|
continue
|
2022-01-03 12:37:09 +00:00
|
|
|
title = remove_html(json_feed_item['content_html'])
|
2021-02-12 11:28:00 +00:00
|
|
|
else:
|
2022-01-03 12:37:09 +00:00
|
|
|
if not isinstance(json_feed_item['content_text'], str):
|
2021-02-12 11:28:00 +00:00
|
|
|
continue
|
2022-01-03 12:37:09 +00:00
|
|
|
title = remove_html(json_feed_item['content_text'])
|
|
|
|
if len(title) > max_bytes:
|
2021-02-12 11:28:00 +00:00
|
|
|
print('WARN: json feed title is too long')
|
|
|
|
continue
|
|
|
|
description = ''
|
2022-01-03 12:37:09 +00:00
|
|
|
if json_feed_item.get('description'):
|
|
|
|
if not isinstance(json_feed_item['description'], str):
|
2021-02-12 11:28:00 +00:00
|
|
|
continue
|
2022-01-03 12:37:09 +00:00
|
|
|
description = remove_html(json_feed_item['description'])
|
|
|
|
if len(description) > max_bytes:
|
2021-02-12 11:28:00 +00:00
|
|
|
print('WARN: json feed description is too long')
|
|
|
|
continue
|
2022-01-03 12:37:09 +00:00
|
|
|
if json_feed_item.get('tags'):
|
|
|
|
if isinstance(json_feed_item['tags'], list):
|
|
|
|
for tag_name in json_feed_item['tags']:
|
|
|
|
if not isinstance(tag_name, str):
|
2021-02-12 12:09:16 +00:00
|
|
|
continue
|
2022-01-03 12:37:09 +00:00
|
|
|
if ' ' in tag_name:
|
2021-02-12 12:09:16 +00:00
|
|
|
continue
|
2022-01-03 12:37:09 +00:00
|
|
|
if not tag_name.startswith('#'):
|
|
|
|
tag_name = '#' + tag_name
|
|
|
|
if tag_name not in description:
|
|
|
|
description += ' ' + tag_name
|
2021-02-12 12:09:16 +00:00
|
|
|
|
2022-01-03 12:37:09 +00:00
|
|
|
link = json_feed_item['url']
|
2021-02-12 11:28:00 +00:00
|
|
|
if '://' not in link:
|
|
|
|
continue
|
2022-01-03 12:37:09 +00:00
|
|
|
if len(link) > max_bytes:
|
2021-02-12 11:28:00 +00:00
|
|
|
print('WARN: json feed link is too long')
|
|
|
|
continue
|
2022-01-03 12:37:09 +00:00
|
|
|
item_domain = link.split('://')[1]
|
|
|
|
if '/' in item_domain:
|
|
|
|
item_domain = item_domain.split('/')[0]
|
|
|
|
if is_blocked_domain(base_dir, item_domain):
|
2021-02-12 11:28:00 +00:00
|
|
|
continue
|
2022-01-03 12:37:09 +00:00
|
|
|
if json_feed_item.get('date_published'):
|
|
|
|
if not isinstance(json_feed_item['date_published'], str):
|
2021-02-12 11:28:00 +00:00
|
|
|
continue
|
2022-01-03 12:37:09 +00:00
|
|
|
pub_date = json_feed_item['date_published']
|
2021-02-12 11:28:00 +00:00
|
|
|
else:
|
2022-01-03 12:37:09 +00:00
|
|
|
if not isinstance(json_feed_item['date_modified'], str):
|
2021-02-12 11:28:00 +00:00
|
|
|
continue
|
2022-01-03 12:37:09 +00:00
|
|
|
pub_date = json_feed_item['date_modified']
|
2021-02-12 11:28:00 +00:00
|
|
|
|
2022-01-27 10:42:46 +00:00
|
|
|
unique_string_identifier = title + ' ' + link
|
|
|
|
pub_date_str = parse_feed_date(pub_date, unique_string_identifier)
|
2022-01-03 12:37:09 +00:00
|
|
|
if pub_date_str:
|
|
|
|
if _valid_feed_date(pub_date_str):
|
2021-12-26 23:41:34 +00:00
|
|
|
post_filename = ''
|
2022-01-03 12:37:09 +00:00
|
|
|
votes_status = []
|
2021-12-29 21:55:09 +00:00
|
|
|
_add_newswire_dict_entry(base_dir, domain,
|
2022-01-03 12:37:09 +00:00
|
|
|
result, pub_date_str,
|
2021-12-29 21:55:09 +00:00
|
|
|
title, link,
|
2022-01-03 12:37:09 +00:00
|
|
|
votes_status, post_filename,
|
2021-12-29 21:55:09 +00:00
|
|
|
description, moderated,
|
2022-01-10 18:48:57 +00:00
|
|
|
mirrored, [], 32, session, debug,
|
|
|
|
None)
|
2022-01-03 12:37:09 +00:00
|
|
|
post_ctr += 1
|
|
|
|
if post_ctr >= max_posts_per_source:
|
2020-12-21 12:11:45 +00:00
|
|
|
break
|
2022-01-03 12:37:09 +00:00
|
|
|
if post_ctr > 0:
|
|
|
|
print('Added ' + str(post_ctr) +
|
2021-02-12 11:28:00 +00:00
|
|
|
' json feed items to newswire')
|
2020-10-10 12:24:14 +00:00
|
|
|
return result
|
|
|
|
|
|
|
|
|
2022-01-03 12:37:09 +00:00
|
|
|
def _atom_feed_yt_to_dict(base_dir: str, domain: str, xml_str: str,
|
2021-12-29 21:55:09 +00:00
|
|
|
moderated: bool, mirrored: bool,
|
2022-01-03 12:37:09 +00:00
|
|
|
max_posts_per_source: int,
|
2021-12-29 21:55:09 +00:00
|
|
|
max_feed_item_size_kb: int,
|
|
|
|
session, debug: bool) -> {}:
|
2020-11-22 10:34:42 +00:00
|
|
|
"""Converts an atom-style YouTube feed string to a dictionary
|
|
|
|
"""
|
2022-01-03 12:37:09 +00:00
|
|
|
if '<entry>' not in xml_str:
|
2020-11-22 10:34:42 +00:00
|
|
|
return {}
|
2021-12-28 21:55:38 +00:00
|
|
|
if is_blocked_domain(base_dir, 'www.youtube.com'):
|
2020-11-22 10:34:42 +00:00
|
|
|
return {}
|
|
|
|
result = {}
|
2022-01-03 12:37:09 +00:00
|
|
|
atom_items = xml_str.split('<entry>')
|
|
|
|
post_ctr = 0
|
|
|
|
max_bytes = max_feed_item_size_kb * 1024
|
2022-01-13 22:57:16 +00:00
|
|
|
first_entry = True
|
2022-01-03 12:37:09 +00:00
|
|
|
for atom_item in atom_items:
|
2022-01-13 22:57:16 +00:00
|
|
|
if first_entry:
|
|
|
|
first_entry = False
|
|
|
|
continue
|
2022-01-03 12:37:09 +00:00
|
|
|
if not atom_item:
|
2020-11-27 22:43:34 +00:00
|
|
|
continue
|
2022-01-03 12:37:09 +00:00
|
|
|
if not atom_item.strip():
|
2020-11-27 22:43:34 +00:00
|
|
|
continue
|
2022-01-03 12:37:09 +00:00
|
|
|
if len(atom_item) > max_bytes:
|
2020-11-22 10:34:42 +00:00
|
|
|
print('WARN: atom feed item is too big')
|
|
|
|
continue
|
2022-01-03 12:37:09 +00:00
|
|
|
if '<title>' not in atom_item:
|
2020-11-22 10:34:42 +00:00
|
|
|
continue
|
2022-01-03 12:37:09 +00:00
|
|
|
if '</title>' not in atom_item:
|
2020-11-22 10:34:42 +00:00
|
|
|
continue
|
2022-01-03 12:37:09 +00:00
|
|
|
if '<published>' not in atom_item:
|
2020-11-22 10:34:42 +00:00
|
|
|
continue
|
2022-01-03 12:37:09 +00:00
|
|
|
if '</published>' not in atom_item:
|
2020-11-22 10:34:42 +00:00
|
|
|
continue
|
2022-01-03 12:37:09 +00:00
|
|
|
if '<yt:videoId>' not in atom_item:
|
2020-11-22 10:34:42 +00:00
|
|
|
continue
|
2022-01-03 12:37:09 +00:00
|
|
|
if '</yt:videoId>' not in atom_item:
|
2020-11-22 10:34:42 +00:00
|
|
|
continue
|
2022-01-03 12:37:09 +00:00
|
|
|
title = atom_item.split('<title>')[1]
|
2021-12-29 21:55:09 +00:00
|
|
|
title = _remove_cdata(title.split('</title>')[0])
|
2020-11-22 10:34:42 +00:00
|
|
|
description = ''
|
2022-01-03 12:37:09 +00:00
|
|
|
if '<media:description>' in atom_item and \
|
|
|
|
'</media:description>' in atom_item:
|
|
|
|
description = atom_item.split('<media:description>')[1]
|
2020-11-22 10:34:42 +00:00
|
|
|
description = description.split('</media:description>')[0]
|
2021-12-27 15:43:22 +00:00
|
|
|
description = remove_html(description)
|
2022-01-03 12:37:09 +00:00
|
|
|
elif '<summary>' in atom_item and '</summary>' in atom_item:
|
|
|
|
description = atom_item.split('<summary>')[1]
|
2020-11-22 10:34:42 +00:00
|
|
|
description = description.split('</summary>')[0]
|
2021-12-27 15:43:22 +00:00
|
|
|
description = remove_html(description)
|
2022-01-14 17:40:42 +00:00
|
|
|
|
2022-04-22 13:46:42 +00:00
|
|
|
link, _ = get_link_from_rss_item(atom_item, None, None)
|
2022-01-14 17:40:42 +00:00
|
|
|
if not link:
|
|
|
|
link = atom_item.split('<yt:videoId>')[1]
|
|
|
|
link = link.split('</yt:videoId>')[0]
|
|
|
|
link = 'https://www.youtube.com/watch?v=' + link.strip()
|
|
|
|
if not link:
|
|
|
|
continue
|
|
|
|
|
2022-01-03 12:37:09 +00:00
|
|
|
pub_date = atom_item.split('<published>')[1]
|
|
|
|
pub_date = pub_date.split('</published>')[0]
|
2020-11-22 18:14:40 +00:00
|
|
|
|
2022-01-27 10:42:46 +00:00
|
|
|
unique_string_identifier = title + ' ' + link
|
|
|
|
pub_date_str = parse_feed_date(pub_date, unique_string_identifier)
|
2022-01-03 12:37:09 +00:00
|
|
|
if pub_date_str:
|
|
|
|
if _valid_feed_date(pub_date_str):
|
2021-12-26 23:41:34 +00:00
|
|
|
post_filename = ''
|
2022-01-03 12:37:09 +00:00
|
|
|
votes_status = []
|
2022-02-12 20:37:15 +00:00
|
|
|
podcast_properties = \
|
|
|
|
xml_podcast_to_dict(base_dir, atom_item, xml_str)
|
2022-01-14 17:40:42 +00:00
|
|
|
if podcast_properties:
|
2022-01-14 18:48:43 +00:00
|
|
|
podcast_properties['linkMimeType'] = 'video/youtube'
|
2021-12-29 21:55:09 +00:00
|
|
|
_add_newswire_dict_entry(base_dir, domain,
|
2022-01-03 12:37:09 +00:00
|
|
|
result, pub_date_str,
|
2021-12-29 21:55:09 +00:00
|
|
|
title, link,
|
2022-01-03 12:37:09 +00:00
|
|
|
votes_status, post_filename,
|
2021-12-29 21:55:09 +00:00
|
|
|
description, moderated, mirrored,
|
2022-01-10 18:48:57 +00:00
|
|
|
[], 32, session, debug,
|
2022-01-14 17:40:42 +00:00
|
|
|
podcast_properties)
|
2022-01-03 12:37:09 +00:00
|
|
|
post_ctr += 1
|
|
|
|
if post_ctr >= max_posts_per_source:
|
2020-12-21 12:11:45 +00:00
|
|
|
break
|
2022-01-03 12:37:09 +00:00
|
|
|
if post_ctr > 0:
|
|
|
|
print('Added ' + str(post_ctr) + ' YouTube feed items to newswire')
|
2020-11-22 10:34:42 +00:00
|
|
|
return result
|
|
|
|
|
|
|
|
|
2022-01-03 12:37:09 +00:00
|
|
|
def _xml_str_to_dict(base_dir: str, domain: str, xml_str: str,
|
2021-12-29 21:55:09 +00:00
|
|
|
moderated: bool, mirrored: bool,
|
2022-01-03 12:37:09 +00:00
|
|
|
max_posts_per_source: int,
|
2021-12-29 21:55:09 +00:00
|
|
|
max_feed_item_size_kb: int,
|
2022-06-09 14:58:47 +00:00
|
|
|
max_categories_feed_item_size_kb: int,
|
2022-04-22 13:46:42 +00:00
|
|
|
session, debug: bool,
|
|
|
|
preferred_podcast_formats: []) -> {}:
|
2020-10-04 09:51:12 +00:00
|
|
|
"""Converts an xml string to a dictionary
|
|
|
|
"""
|
2022-01-03 12:37:09 +00:00
|
|
|
if '<yt:videoId>' in xml_str and '<yt:channelId>' in xml_str:
|
2020-11-22 16:10:58 +00:00
|
|
|
print('YouTube feed: reading')
|
2021-12-29 21:55:09 +00:00
|
|
|
return _atom_feed_yt_to_dict(base_dir, domain,
|
2022-01-03 12:37:09 +00:00
|
|
|
xml_str, moderated, mirrored,
|
|
|
|
max_posts_per_source,
|
|
|
|
max_feed_item_size_kb,
|
2021-12-29 21:55:09 +00:00
|
|
|
session, debug)
|
2022-01-10 22:30:06 +00:00
|
|
|
if 'rss version="2.0"' in xml_str:
|
2021-12-29 21:55:09 +00:00
|
|
|
return _xml2str_to_dict(base_dir, domain,
|
2022-01-03 12:37:09 +00:00
|
|
|
xml_str, moderated, mirrored,
|
|
|
|
max_posts_per_source, max_feed_item_size_kb,
|
2022-06-09 14:58:47 +00:00
|
|
|
max_categories_feed_item_size_kb,
|
2022-04-22 13:46:42 +00:00
|
|
|
session, debug,
|
|
|
|
preferred_podcast_formats)
|
2022-01-10 22:30:06 +00:00
|
|
|
if '<?xml version="1.0"' in xml_str:
|
2021-12-29 21:55:09 +00:00
|
|
|
return _xml1str_to_dict(base_dir, domain,
|
2022-01-03 12:37:09 +00:00
|
|
|
xml_str, moderated, mirrored,
|
|
|
|
max_posts_per_source, max_feed_item_size_kb,
|
2022-06-09 14:58:47 +00:00
|
|
|
max_categories_feed_item_size_kb,
|
2022-04-22 13:46:42 +00:00
|
|
|
session, debug, preferred_podcast_formats)
|
2022-01-10 22:30:06 +00:00
|
|
|
if 'xmlns="http://www.w3.org/2005/Atom"' in xml_str:
|
2021-12-29 21:55:09 +00:00
|
|
|
return _atom_feed_to_dict(base_dir, domain,
|
2022-01-03 12:37:09 +00:00
|
|
|
xml_str, moderated, mirrored,
|
|
|
|
max_posts_per_source, max_feed_item_size_kb,
|
2022-04-22 13:46:42 +00:00
|
|
|
session, debug, preferred_podcast_formats)
|
2022-01-10 22:30:06 +00:00
|
|
|
if 'https://jsonfeed.org/version/1' in xml_str:
|
2021-12-29 21:55:09 +00:00
|
|
|
return _json_feed_v1to_dict(base_dir, domain,
|
2022-01-03 12:37:09 +00:00
|
|
|
xml_str, moderated, mirrored,
|
|
|
|
max_posts_per_source,
|
|
|
|
max_feed_item_size_kb,
|
2021-12-29 21:55:09 +00:00
|
|
|
session, debug)
|
2020-10-04 09:51:12 +00:00
|
|
|
return {}
|
|
|
|
|
|
|
|
|
2022-01-03 12:37:09 +00:00
|
|
|
def _yt_channel_to_atom_feed(url: str) -> str:
|
2020-11-22 10:46:54 +00:00
|
|
|
"""Converts a YouTube channel url into an atom feed url
|
|
|
|
"""
|
|
|
|
if 'youtube.com/channel/' not in url:
|
|
|
|
return url
|
2022-01-03 12:37:09 +00:00
|
|
|
channel_id = url.split('youtube.com/channel/')[1].strip()
|
|
|
|
channel_url = \
|
|
|
|
'https://www.youtube.com/feeds/videos.xml?channel_id=' + channel_id
|
|
|
|
print('YouTube feed: ' + channel_url)
|
|
|
|
return channel_url
|
2020-11-22 10:46:54 +00:00
|
|
|
|
|
|
|
|
2021-12-29 21:55:09 +00:00
|
|
|
def get_rss(base_dir: str, domain: str, session, url: str,
|
|
|
|
moderated: bool, mirrored: bool,
|
2022-01-03 12:37:09 +00:00
|
|
|
max_posts_per_source: int, max_feed_size_kb: int,
|
2021-12-29 21:55:09 +00:00
|
|
|
max_feed_item_size_kb: int,
|
2022-06-09 14:58:47 +00:00
|
|
|
max_categories_feed_item_size_kb: int, debug: bool,
|
2022-04-24 19:03:02 +00:00
|
|
|
preferred_podcast_formats: [],
|
|
|
|
timeout_sec: int) -> {}:
|
2020-10-04 09:51:12 +00:00
|
|
|
"""Returns an RSS url as a dict
|
|
|
|
"""
|
|
|
|
if not isinstance(url, str):
|
|
|
|
print('url: ' + str(url))
|
2021-12-29 21:55:09 +00:00
|
|
|
print('ERROR: get_rss url should be a string')
|
2020-10-04 09:51:12 +00:00
|
|
|
return None
|
|
|
|
headers = {
|
2020-12-14 20:22:05 +00:00
|
|
|
'Accept': 'text/xml, application/xml; charset=UTF-8'
|
2020-10-04 09:51:12 +00:00
|
|
|
}
|
|
|
|
params = None
|
2022-01-03 12:37:09 +00:00
|
|
|
session_params = {}
|
|
|
|
session_headers = {}
|
2020-10-04 09:51:12 +00:00
|
|
|
if headers:
|
2022-01-03 12:37:09 +00:00
|
|
|
session_headers = headers
|
2020-10-04 09:51:12 +00:00
|
|
|
if params:
|
2022-01-03 12:37:09 +00:00
|
|
|
session_params = params
|
|
|
|
session_headers['User-Agent'] = \
|
2020-10-04 09:51:12 +00:00
|
|
|
'Mozilla/5.0 (X11; Linux x86_64; rv:81.0) Gecko/20100101 Firefox/81.0'
|
|
|
|
if not session:
|
2021-12-29 21:55:09 +00:00
|
|
|
print('WARN: no session specified for get_rss')
|
2022-01-03 12:37:09 +00:00
|
|
|
url = _yt_channel_to_atom_feed(url)
|
2020-10-04 09:51:12 +00:00
|
|
|
try:
|
2022-01-03 12:37:09 +00:00
|
|
|
result = \
|
2022-04-24 19:03:02 +00:00
|
|
|
session.get(url, headers=session_headers,
|
|
|
|
params=session_params,
|
2022-04-24 20:33:07 +00:00
|
|
|
timeout=timeout_sec,
|
|
|
|
allow_redirects=False)
|
2020-10-16 11:40:01 +00:00
|
|
|
if result:
|
2022-01-03 12:37:09 +00:00
|
|
|
if int(len(result.text) / 1024) < max_feed_size_kb and \
|
2021-12-27 17:53:41 +00:00
|
|
|
not contains_invalid_chars(result.text):
|
2021-12-29 21:55:09 +00:00
|
|
|
return _xml_str_to_dict(base_dir, domain, result.text,
|
|
|
|
moderated, mirrored,
|
2022-01-03 12:37:09 +00:00
|
|
|
max_posts_per_source,
|
2021-12-29 21:55:09 +00:00
|
|
|
max_feed_item_size_kb,
|
2022-06-09 14:58:47 +00:00
|
|
|
max_categories_feed_item_size_kb,
|
2022-04-22 13:46:42 +00:00
|
|
|
session, debug,
|
|
|
|
preferred_podcast_formats)
|
2022-05-30 15:15:17 +00:00
|
|
|
print('WARN: feed is too large, ' +
|
|
|
|
'or contains invalid characters: ' + url)
|
2020-11-22 13:04:58 +00:00
|
|
|
else:
|
|
|
|
print('WARN: no result returned for feed ' + url)
|
2021-12-25 15:28:52 +00:00
|
|
|
except requests.exceptions.RequestException as ex:
|
2021-12-29 21:55:09 +00:00
|
|
|
print('WARN: get_rss failed\nurl: ' + str(url) + ', ' +
|
2022-01-03 12:37:09 +00:00
|
|
|
'headers: ' + str(session_headers) + ', ' +
|
|
|
|
'params: ' + str(session_params) + ', ' + str(ex))
|
2021-12-25 15:28:52 +00:00
|
|
|
except ValueError as ex:
|
2021-12-29 21:55:09 +00:00
|
|
|
print('WARN: get_rss failed\nurl: ' + str(url) + ', ' +
|
2022-01-03 12:37:09 +00:00
|
|
|
'headers: ' + str(session_headers) + ', ' +
|
|
|
|
'params: ' + str(session_params) + ', ' + str(ex))
|
2021-12-25 15:28:52 +00:00
|
|
|
except SocketError as ex:
|
|
|
|
if ex.errno == errno.ECONNRESET:
|
2021-12-29 21:55:09 +00:00
|
|
|
print('WARN: connection was reset during get_rss ' + str(ex))
|
2021-05-20 12:52:13 +00:00
|
|
|
else:
|
2021-12-29 21:55:09 +00:00
|
|
|
print('WARN: get_rss, ' + str(ex))
|
2020-10-04 09:51:12 +00:00
|
|
|
return None
|
|
|
|
|
|
|
|
|
2021-12-29 21:55:09 +00:00
|
|
|
def get_rs_sfrom_dict(base_dir: str, newswire: {},
|
|
|
|
http_prefix: str, domain_full: str,
|
|
|
|
title: str, translate: {}) -> str:
|
2020-10-04 12:29:07 +00:00
|
|
|
"""Returns an rss feed from the current newswire dict.
|
|
|
|
This allows other instances to subscribe to the same newswire
|
|
|
|
"""
|
2022-01-03 12:37:09 +00:00
|
|
|
rss_str = rss2header(http_prefix,
|
|
|
|
None, domain_full,
|
|
|
|
'Newswire', translate)
|
2020-11-03 14:41:28 +00:00
|
|
|
if not newswire:
|
|
|
|
return ''
|
2020-10-04 12:29:07 +00:00
|
|
|
for published, fields in newswire.items():
|
2020-10-20 12:22:52 +00:00
|
|
|
if '+00:00' in published:
|
|
|
|
published = published.replace('+00:00', 'Z').strip()
|
|
|
|
published = published.replace(' ', 'T')
|
|
|
|
else:
|
2022-01-03 12:37:09 +00:00
|
|
|
published_with_offset = \
|
2020-10-20 12:37:32 +00:00
|
|
|
datetime.strptime(published, "%Y-%m-%d %H:%M:%S%z")
|
2022-01-03 12:37:09 +00:00
|
|
|
published = published_with_offset.strftime("%Y-%m-%dT%H:%M:%SZ")
|
2020-10-04 22:08:13 +00:00
|
|
|
try:
|
2022-01-03 12:37:09 +00:00
|
|
|
pub_date = datetime.strptime(published, "%Y-%m-%dT%H:%M:%SZ")
|
2022-02-03 10:39:52 +00:00
|
|
|
except BaseException as ex:
|
2021-12-25 15:28:52 +00:00
|
|
|
print('WARN: Unable to convert date ' + published + ' ' + str(ex))
|
2020-10-04 22:08:13 +00:00
|
|
|
continue
|
2022-01-03 12:37:09 +00:00
|
|
|
rss_str += \
|
2021-07-04 11:02:08 +00:00
|
|
|
'<item>\n' + \
|
|
|
|
' <title>' + fields[0] + '</title>\n'
|
2021-12-27 15:52:08 +00:00
|
|
|
description = remove_html(first_paragraph_from_string(fields[4]))
|
2022-01-03 12:37:09 +00:00
|
|
|
rss_str += ' <description>' + description + '</description>\n'
|
2020-10-08 15:07:06 +00:00
|
|
|
url = fields[1]
|
2020-11-08 11:04:52 +00:00
|
|
|
if '://' not in url:
|
2021-12-26 10:00:46 +00:00
|
|
|
if domain_full not in url:
|
|
|
|
url = http_prefix + '://' + domain_full + url
|
2022-01-03 12:37:09 +00:00
|
|
|
rss_str += ' <link>' + url + '</link>\n'
|
2020-10-04 22:12:27 +00:00
|
|
|
|
2022-01-03 12:37:09 +00:00
|
|
|
rss_date_str = pub_date.strftime("%a, %d %b %Y %H:%M:%S UT")
|
|
|
|
rss_str += \
|
|
|
|
' <pubDate>' + rss_date_str + '</pubDate>\n' + \
|
2021-07-04 11:02:08 +00:00
|
|
|
'</item>\n'
|
2022-01-03 12:37:09 +00:00
|
|
|
rss_str += rss2footer()
|
|
|
|
return rss_str
|
2020-10-04 12:29:07 +00:00
|
|
|
|
|
|
|
|
2021-12-29 21:55:09 +00:00
|
|
|
def _is_newswire_blog_post(post_json_object: {}) -> bool:
|
2020-10-06 11:28:32 +00:00
|
|
|
"""Is the given object a blog post?
|
2020-10-25 10:47:39 +00:00
|
|
|
There isn't any difference between a blog post and a newswire blog post
|
|
|
|
but we may here need to check for different properties than
|
2021-12-28 13:49:44 +00:00
|
|
|
is_blog_post does
|
2020-10-06 11:28:32 +00:00
|
|
|
"""
|
2021-12-25 22:09:19 +00:00
|
|
|
if not post_json_object:
|
2020-10-06 11:28:32 +00:00
|
|
|
return False
|
2021-12-26 10:57:03 +00:00
|
|
|
if not has_object_dict(post_json_object):
|
2020-10-06 11:28:32 +00:00
|
|
|
return False
|
2021-12-25 22:09:19 +00:00
|
|
|
if post_json_object['object'].get('summary') and \
|
|
|
|
post_json_object['object'].get('url') and \
|
|
|
|
post_json_object['object'].get('content') and \
|
|
|
|
post_json_object['object'].get('published'):
|
2021-12-28 14:41:10 +00:00
|
|
|
return is_public_post(post_json_object)
|
2020-10-06 11:28:32 +00:00
|
|
|
return False
|
|
|
|
|
|
|
|
|
2021-12-29 21:55:09 +00:00
|
|
|
def _get_hashtags_from_post(post_json_object: {}) -> []:
|
2020-10-16 20:13:23 +00:00
|
|
|
"""Returns a list of any hashtags within a post
|
|
|
|
"""
|
2021-12-26 10:57:03 +00:00
|
|
|
if not has_object_dict(post_json_object):
|
2020-10-16 20:13:23 +00:00
|
|
|
return []
|
2021-12-25 22:09:19 +00:00
|
|
|
if not post_json_object['object'].get('tag'):
|
2020-10-16 20:13:23 +00:00
|
|
|
return []
|
2021-12-25 22:09:19 +00:00
|
|
|
if not isinstance(post_json_object['object']['tag'], list):
|
2020-10-16 20:13:23 +00:00
|
|
|
return []
|
|
|
|
tags = []
|
2022-01-03 12:37:09 +00:00
|
|
|
for tgname in post_json_object['object']['tag']:
|
|
|
|
if not isinstance(tgname, dict):
|
2020-10-16 20:13:23 +00:00
|
|
|
continue
|
2022-01-03 12:37:09 +00:00
|
|
|
if not tgname.get('name'):
|
2020-10-16 20:13:23 +00:00
|
|
|
continue
|
2022-01-03 12:37:09 +00:00
|
|
|
if not tgname.get('type'):
|
2020-10-16 20:13:23 +00:00
|
|
|
continue
|
2022-01-03 12:37:09 +00:00
|
|
|
if tgname['type'] != 'Hashtag':
|
2020-10-16 20:13:23 +00:00
|
|
|
continue
|
2022-01-03 12:37:09 +00:00
|
|
|
if tgname['name'] not in tags:
|
|
|
|
tags.append(tgname['name'])
|
2020-10-16 20:13:23 +00:00
|
|
|
return tags
|
|
|
|
|
|
|
|
|
2021-12-29 21:55:09 +00:00
|
|
|
def _add_account_blogs_to_newswire(base_dir: str, nickname: str, domain: str,
|
|
|
|
newswire: {},
|
2022-01-03 12:37:09 +00:00
|
|
|
max_blogs_per_account: int,
|
|
|
|
index_filename: str,
|
|
|
|
max_tags: int, system_language: str,
|
2021-12-29 21:55:09 +00:00
|
|
|
session, debug: bool) -> None:
|
2020-10-05 11:11:48 +00:00
|
|
|
"""Adds blogs for the given account to the newswire
|
|
|
|
"""
|
2022-01-03 12:37:09 +00:00
|
|
|
if not os.path.isfile(index_filename):
|
2020-10-05 11:11:48 +00:00
|
|
|
return
|
2020-10-09 10:33:06 +00:00
|
|
|
# local blog entries are unmoderated by default
|
|
|
|
moderated = False
|
|
|
|
|
|
|
|
# local blogs can potentially be moderated
|
2022-01-03 12:37:09 +00:00
|
|
|
moderated_filename = \
|
2021-12-26 12:02:29 +00:00
|
|
|
acct_dir(base_dir, nickname, domain) + '/.newswiremoderated'
|
2022-01-03 12:37:09 +00:00
|
|
|
if os.path.isfile(moderated_filename):
|
2020-10-09 10:33:06 +00:00
|
|
|
moderated = True
|
|
|
|
|
2022-06-09 14:58:47 +00:00
|
|
|
with open(index_filename, 'r', encoding='utf-8') as index_file:
|
2021-12-26 23:41:34 +00:00
|
|
|
post_filename = 'start'
|
2020-10-05 11:11:48 +00:00
|
|
|
ctr = 0
|
2021-12-26 23:41:34 +00:00
|
|
|
while post_filename:
|
2022-01-03 12:37:09 +00:00
|
|
|
post_filename = index_file.readline()
|
2021-12-26 23:41:34 +00:00
|
|
|
if post_filename:
|
2020-10-05 11:11:48 +00:00
|
|
|
# if this is a full path then remove the directories
|
2021-12-26 23:41:34 +00:00
|
|
|
if '/' in post_filename:
|
|
|
|
post_filename = post_filename.split('/')[-1]
|
2020-10-05 11:11:48 +00:00
|
|
|
|
|
|
|
# filename of the post without any extension or path
|
|
|
|
# This should also correspond to any index entry in
|
|
|
|
# the posts cache
|
2022-01-03 12:37:09 +00:00
|
|
|
post_url = \
|
2021-12-26 23:41:34 +00:00
|
|
|
post_filename.replace('\n', '').replace('\r', '')
|
2022-01-03 12:37:09 +00:00
|
|
|
post_url = post_url.replace('.json', '').strip()
|
2020-10-05 11:11:48 +00:00
|
|
|
|
|
|
|
# read the post from file
|
2022-01-03 12:37:09 +00:00
|
|
|
full_post_filename = \
|
2021-12-26 20:36:08 +00:00
|
|
|
locate_post(base_dir, nickname,
|
2022-01-03 12:37:09 +00:00
|
|
|
domain, post_url, False)
|
|
|
|
if not full_post_filename:
|
|
|
|
print('Unable to locate post for newswire ' + post_url)
|
2020-10-06 13:05:15 +00:00
|
|
|
ctr += 1
|
2022-01-03 12:37:09 +00:00
|
|
|
if ctr >= max_blogs_per_account:
|
2020-10-06 13:05:15 +00:00
|
|
|
break
|
2020-10-06 13:34:04 +00:00
|
|
|
continue
|
2020-10-06 13:05:15 +00:00
|
|
|
|
2021-12-25 22:09:19 +00:00
|
|
|
post_json_object = None
|
2022-01-03 12:37:09 +00:00
|
|
|
if full_post_filename:
|
|
|
|
post_json_object = load_json(full_post_filename)
|
2021-12-29 21:55:09 +00:00
|
|
|
if _is_newswire_blog_post(post_json_object):
|
2021-12-25 22:09:19 +00:00
|
|
|
published = post_json_object['object']['published']
|
2020-10-06 11:28:32 +00:00
|
|
|
published = published.replace('T', ' ')
|
|
|
|
published = published.replace('Z', '+00:00')
|
2020-10-06 20:17:34 +00:00
|
|
|
votes = []
|
2022-01-03 12:37:09 +00:00
|
|
|
if os.path.isfile(full_post_filename + '.votes'):
|
|
|
|
votes = load_json(full_post_filename + '.votes')
|
2021-07-18 14:15:16 +00:00
|
|
|
content = \
|
2021-12-26 11:29:40 +00:00
|
|
|
get_base_content_from_post(post_json_object,
|
|
|
|
system_language)
|
2021-12-27 15:52:08 +00:00
|
|
|
description = first_paragraph_from_string(content)
|
2021-12-27 15:43:22 +00:00
|
|
|
description = remove_html(description)
|
2022-01-03 12:37:09 +00:00
|
|
|
tags_from_post = _get_hashtags_from_post(post_json_object)
|
2021-12-25 22:09:19 +00:00
|
|
|
summary = post_json_object['object']['summary']
|
2021-12-29 21:55:09 +00:00
|
|
|
_add_newswire_dict_entry(base_dir, domain,
|
|
|
|
newswire, published,
|
|
|
|
summary,
|
|
|
|
post_json_object['object']['url'],
|
2022-01-03 12:37:09 +00:00
|
|
|
votes, full_post_filename,
|
2021-12-29 21:55:09 +00:00
|
|
|
description, moderated, False,
|
2022-01-03 12:37:09 +00:00
|
|
|
tags_from_post,
|
2022-01-10 18:48:57 +00:00
|
|
|
max_tags, session, debug,
|
|
|
|
None)
|
2020-10-05 11:11:48 +00:00
|
|
|
|
|
|
|
ctr += 1
|
2022-01-03 12:37:09 +00:00
|
|
|
if ctr >= max_blogs_per_account:
|
2020-10-05 11:11:48 +00:00
|
|
|
break
|
|
|
|
|
|
|
|
|
2021-12-29 21:55:09 +00:00
|
|
|
def _add_blogs_to_newswire(base_dir: str, domain: str, newswire: {},
|
2022-01-03 12:37:09 +00:00
|
|
|
max_blogs_per_account: int,
|
|
|
|
max_tags: int, system_language: str,
|
2021-12-29 21:55:09 +00:00
|
|
|
session, debug: bool) -> None:
|
2020-10-06 09:47:58 +00:00
|
|
|
"""Adds blogs from each user account into the newswire
|
2020-10-06 09:37:22 +00:00
|
|
|
"""
|
2022-01-03 12:37:09 +00:00
|
|
|
moderation_dict = {}
|
2020-10-06 10:34:56 +00:00
|
|
|
|
2020-10-05 11:11:48 +00:00
|
|
|
# go through each account
|
2022-01-03 12:37:09 +00:00
|
|
|
for _, dirs, _ in os.walk(base_dir + '/accounts'):
|
2020-10-05 11:11:48 +00:00
|
|
|
for handle in dirs:
|
2021-12-26 18:46:43 +00:00
|
|
|
if not is_account_dir(handle):
|
2020-10-05 11:11:48 +00:00
|
|
|
continue
|
2020-10-06 10:34:56 +00:00
|
|
|
|
2020-10-06 09:37:22 +00:00
|
|
|
nickname = handle.split('@')[0]
|
2020-10-05 11:30:11 +00:00
|
|
|
|
|
|
|
# has this account been suspended?
|
2021-12-27 15:37:31 +00:00
|
|
|
if is_suspended(base_dir, nickname):
|
2020-10-06 08:58:44 +00:00
|
|
|
continue
|
2020-10-05 11:30:11 +00:00
|
|
|
|
2021-12-25 16:17:53 +00:00
|
|
|
if os.path.isfile(base_dir + '/accounts/' + handle +
|
2020-10-06 21:28:40 +00:00
|
|
|
'/.nonewswire'):
|
|
|
|
continue
|
|
|
|
|
2020-10-05 11:11:48 +00:00
|
|
|
# is there a blogs timeline for this account?
|
2022-01-03 12:37:09 +00:00
|
|
|
account_dir = os.path.join(base_dir + '/accounts', handle)
|
|
|
|
blogs_index = account_dir + '/tlblogs.index'
|
|
|
|
if os.path.isfile(blogs_index):
|
2020-10-05 11:11:48 +00:00
|
|
|
domain = handle.split('@')[1]
|
2021-12-29 21:55:09 +00:00
|
|
|
_add_account_blogs_to_newswire(base_dir, nickname, domain,
|
2022-01-03 12:37:09 +00:00
|
|
|
newswire, max_blogs_per_account,
|
|
|
|
blogs_index, max_tags,
|
2021-12-29 21:55:09 +00:00
|
|
|
system_language, session,
|
|
|
|
debug)
|
2020-12-13 22:13:45 +00:00
|
|
|
break
|
2020-10-05 11:11:48 +00:00
|
|
|
|
2020-10-06 11:28:32 +00:00
|
|
|
# sort the moderation dict into chronological order, latest first
|
2022-01-03 12:37:09 +00:00
|
|
|
sorted_moderation_dict = \
|
|
|
|
OrderedDict(sorted(moderation_dict.items(), reverse=True))
|
2020-10-06 12:15:35 +00:00
|
|
|
# save the moderation queue details for later display
|
2022-01-03 12:37:09 +00:00
|
|
|
newswire_moderation_filename = \
|
|
|
|
base_dir + '/accounts/newswiremoderation.txt'
|
|
|
|
if sorted_moderation_dict:
|
|
|
|
save_json(sorted_moderation_dict, newswire_moderation_filename)
|
2020-10-06 14:32:53 +00:00
|
|
|
else:
|
|
|
|
# remove the file if there is nothing to moderate
|
2022-01-03 12:37:09 +00:00
|
|
|
if os.path.isfile(newswire_moderation_filename):
|
2021-09-05 10:17:43 +00:00
|
|
|
try:
|
2022-01-03 12:37:09 +00:00
|
|
|
os.remove(newswire_moderation_filename)
|
2021-11-25 18:42:38 +00:00
|
|
|
except OSError:
|
2021-12-29 21:55:09 +00:00
|
|
|
print('EX: _add_blogs_to_newswire unable to delete ' +
|
2022-01-03 12:37:09 +00:00
|
|
|
str(newswire_moderation_filename))
|
2020-10-06 11:28:32 +00:00
|
|
|
|
2020-10-05 11:11:48 +00:00
|
|
|
|
2021-12-29 21:55:09 +00:00
|
|
|
def get_dict_from_newswire(session, base_dir: str, domain: str,
|
2022-01-03 12:37:09 +00:00
|
|
|
max_posts_per_source: int, max_feed_size_kb: int,
|
|
|
|
max_tags: int, max_feed_item_size_kb: int,
|
2021-12-29 21:55:09 +00:00
|
|
|
max_newswire_posts: int,
|
2022-06-09 14:58:47 +00:00
|
|
|
max_categories_feed_item_size_kb: int,
|
2022-04-22 13:46:42 +00:00
|
|
|
system_language: str, debug: bool,
|
2022-04-24 19:03:02 +00:00
|
|
|
preferred_podcast_formats: [],
|
|
|
|
timeout_sec: int) -> {}:
|
2020-10-04 09:59:55 +00:00
|
|
|
"""Gets rss feeds as a dictionary from newswire file
|
2020-10-04 09:51:12 +00:00
|
|
|
"""
|
2022-01-03 12:37:09 +00:00
|
|
|
subscriptions_filename = base_dir + '/accounts/newswire.txt'
|
|
|
|
if not os.path.isfile(subscriptions_filename):
|
2020-10-04 09:51:12 +00:00
|
|
|
return {}
|
|
|
|
|
2022-01-03 12:37:09 +00:00
|
|
|
max_posts_per_source = 5
|
2020-10-16 10:13:14 +00:00
|
|
|
|
2020-10-05 11:11:48 +00:00
|
|
|
# add rss feeds
|
2022-01-03 12:37:09 +00:00
|
|
|
rss_feed = []
|
2022-06-09 14:58:47 +00:00
|
|
|
with open(subscriptions_filename, 'r', encoding='utf-8') as fp_sub:
|
2022-01-03 12:37:09 +00:00
|
|
|
rss_feed = fp_sub.readlines()
|
2020-10-04 09:51:12 +00:00
|
|
|
result = {}
|
2022-01-03 12:37:09 +00:00
|
|
|
for url in rss_feed:
|
2020-10-04 09:51:12 +00:00
|
|
|
url = url.strip()
|
2020-10-09 10:33:06 +00:00
|
|
|
|
|
|
|
# Does this contain a url?
|
2020-10-04 09:51:12 +00:00
|
|
|
if '://' not in url:
|
|
|
|
continue
|
2020-10-09 10:33:06 +00:00
|
|
|
|
|
|
|
# is this a comment?
|
2020-10-04 09:51:12 +00:00
|
|
|
if url.startswith('#'):
|
|
|
|
continue
|
2020-10-09 10:33:06 +00:00
|
|
|
|
|
|
|
# should this feed be moderated?
|
|
|
|
moderated = False
|
|
|
|
if '*' in url:
|
|
|
|
moderated = True
|
|
|
|
url = url.replace('*', '').strip()
|
|
|
|
|
2020-10-19 14:37:17 +00:00
|
|
|
# should this feed content be mirrored?
|
|
|
|
mirrored = False
|
|
|
|
if '!' in url:
|
|
|
|
mirrored = True
|
|
|
|
url = url.replace('!', '').strip()
|
|
|
|
|
2022-01-03 12:37:09 +00:00
|
|
|
items_list = get_rss(base_dir, domain, session, url,
|
|
|
|
moderated, mirrored,
|
|
|
|
max_posts_per_source, max_feed_size_kb,
|
|
|
|
max_feed_item_size_kb,
|
2022-06-09 14:58:47 +00:00
|
|
|
max_categories_feed_item_size_kb, debug,
|
2022-04-24 19:03:02 +00:00
|
|
|
preferred_podcast_formats,
|
|
|
|
timeout_sec)
|
2022-01-03 12:37:09 +00:00
|
|
|
if items_list:
|
|
|
|
for date_str, item in items_list.items():
|
|
|
|
result[date_str] = item
|
2022-04-27 17:12:25 +00:00
|
|
|
time.sleep(4)
|
2020-10-05 11:11:48 +00:00
|
|
|
|
2020-10-06 09:47:58 +00:00
|
|
|
# add blogs from each user account
|
2021-12-29 21:55:09 +00:00
|
|
|
_add_blogs_to_newswire(base_dir, domain, result,
|
2022-01-03 12:37:09 +00:00
|
|
|
max_posts_per_source, max_tags, system_language,
|
2021-12-29 21:55:09 +00:00
|
|
|
session, debug)
|
2020-10-05 11:11:48 +00:00
|
|
|
|
|
|
|
# sort into chronological order, latest first
|
2022-01-03 12:37:09 +00:00
|
|
|
sorted_result = OrderedDict(sorted(result.items(), reverse=True))
|
2020-11-22 11:48:53 +00:00
|
|
|
|
|
|
|
# are there too many posts? If so then remove the oldest ones
|
2022-01-03 12:37:09 +00:00
|
|
|
no_of_posts = len(sorted_result.items())
|
|
|
|
if no_of_posts > max_newswire_posts:
|
2020-11-22 12:05:15 +00:00
|
|
|
ctr = 0
|
|
|
|
removals = []
|
2022-01-03 12:37:09 +00:00
|
|
|
for date_str, item in sorted_result.items():
|
2020-11-22 12:05:15 +00:00
|
|
|
ctr += 1
|
2021-12-25 18:49:19 +00:00
|
|
|
if ctr > max_newswire_posts:
|
2022-01-03 12:37:09 +00:00
|
|
|
removals.append(date_str)
|
|
|
|
for remov in removals:
|
|
|
|
sorted_result.pop(remov)
|
2020-11-22 11:48:53 +00:00
|
|
|
|
2022-01-03 12:37:09 +00:00
|
|
|
return sorted_result
|