def export(update, context):
if update.edited_message or update.edited_channel_post:
return
msg = update.effective_message
if msg.chat_id < 0 and ('source' in msg.text) and ('[source]'
in msg.text_markdown):
return
if msg.chat.username == 'web_record':
if (matchKey(msg.text_markdown,
['twitter', 'weibo', 'douban', 't.me/'])
and not matchKey(msg.text_markdown, ['article', 'note'])):
tryDelete(msg)
return
try:
r = msg.chat.send_message('received')
except:
return
try:
exportImp(msg)
if msg.chat.username == 'web_record':
tryDelete(msg)
except Exception as e:
msg.chat.send_message(str(e))
if not matchKey(str(e), ['Content is too big.']):
raise e
finally:
r.delete()
def veryBadMsg(msg, has_similar_log):
if msg.forward_from_chat:
if matchKey(msg.forward_from_chat.title, [
'新闻频道', '新闻网', '我的频道', '点我有惊喜', '引流推广', '自由之声🌈', '业务咨询', '大家好',
'信息'
]):
return True
if badTextScore(msg.forward_from_chat.title)[0]:
return True
if badText(msg.text):
return True
if ((not msg.from_user.last_name) and (not msg.from_user.username)
and len(msg.from_user.first_name) <= 3):
return True
if badText(msg.caption):
return True
if matchKey(msg.text, [
'群发需要滴滴', 'joinchat', 'Pua把妹教程五百多套', '前一百名加我的兄弟', '联系我', '+v',
'Louisville_fx', 'LY11618', '滴我', 'Binance'
]):
return True
if hasSeq(msg.text, ['原生话费收粮', '啦裙發']):
return True
if msg.contact:
return True
if has_similar_log:
return True
return False
def _formatImgUrl(raw, domain):
parts = raw.split('/')
success = False
for index, part in enumerate(parts):
if part == 'max':
try:
if int(parts[index + 1]) > 0:
success = True
break
except:
pass
if success and 'guim' not in raw:
parts[index + 1] = '1300'
raw = '/'.join(parts)
if not matchKey(raw, ['guim']):
raw = re.sub('width=\d\d*', 'width=1300', raw)
if matchKey(raw, ['condecdn']):
raw = re.sub('/\d\d*/', '/1300/', raw)
if matchKey(raw, ['twreport']) and matchKey(raw, ['-tiny']):
raw = raw.replace('-tiny', '-desktop')
if raw.startswith('//'):
return 'https:' + raw
if raw.startswith('/'):
return domain + raw
return raw
def _decompseAds(soup):
for item in soup.find_all("div", class_="article-paragraph"):
if matchKey(item.text, DIV_AD_WORDS):
_decompose(item)
for item in soup.find_all("p"):
if matchKey(item.text, P_AD_WORDS) or item.text in ['广告']:
_decompose(item)
def export(update, context):
if update.edited_message or update.edited_channel_post:
return
msg = update.effective_message
if msg.chat_id < 0 and 'source</a>' in msg.text_html_urled:
return
if msg.chat.username == 'web_record':
if (matchKey(msg.text_markdown,
['twitter', 'weibo', 'douban', 't.me/'])
and not matchKey(msg.text_markdown, ['article', 'note'])):
return
try:
tmp_msg_1 = msg.chat.send_message('received')
except:
return
error = ''
result = []
try:
result = list(exportImp(msg))
if str(msg.chat.id) in remove_origin._db.items:
tryDelete(msg)
except Exception as e:
tmp_msg_2 = msg.chat.send_message(str(e))
autoDestroy(tmp_msg_2, 0.05)
error = ' error: ' + str(e)
finally:
info_log.send_message(getBasicLog(msg) + error + ' result: ' +
' '.join(result),
parse_mode='html',
disable_web_page_preview=True)
tmp_msg_1.delete()
def command(update, context): msg = update.message if matchKey(msg.text, ['auth', 'token']): return msgTelegraphToken(msg) if matchKey(msg.text, ['toggle', 'source']): return toggleSourceLink(msg) if msg.chat_id > 0: msg.reply_text(help_message)
def commandInternal(msg): command, text = splitCommand(msg.text) if matchKey(command, ['/abl', '/d_ba', 'blocklist_ba']): return blocklist.add(text) if matchKey(command, ['/d_br', 'blocklist_br']): return blocklist.remove(text) if matchKey(command, ['/d_bl', 'blocklist_list']): return 'blocklist:\n' + '\n'.join(blocklist.items())
def shouldSend(card):
if matchKey(str(card), db.whitelist.items):
return True
if matchKey(str(card), db.blacklist.items):
return False
if matchKey(str(card), db.preferlist.items):
return getCount(card.get('mblog')) > 300
if matchKey(str(card), db.popularlist.items):
return getCount(card.get('mblog')) > 10000
return getCount(card.get('mblog')) > 1000
def shouldFlipFirst(key):
channel = key.split('/')[0]
if channels.get(channel) == -1:
return False
if 0 <= channels.get(channel) <= 2:
return True
if len(index.get(key)) < 20 and not matchKey(index.get(key),
['hasFile', 'hasLink']):
return False
return not matchKey(index.get(key), blocklist.items())
def command(update, context):
msg = update.message or update.channel_post
if matchKey(msg.text, ['auth', 'token']):
return msgTelegraphToken(msg)
if matchKey(msg.text, ['source', 'tnsl', 'toggle_no_source_link']):
return toggleSourceLink(msg)
if matchKey(msg.text, ['origin', 'trmo', 'toggle_remove_origin']):
return toggleRemoveOrigin(msg)
if msg.chat_id > 0:
msg.reply_text(help_message)
def command(update, context):
msg = update.message
if matchKey(msg.text, ['auth', 'token']):
return get_telegraph_token(msg)
if matchKey(msg.text, ['source']):
return switch_source_flag(msg)
if matchKey(msg.text, ['simplify']):
return switch_simplify_flag(msg)
if msg.chat_id > 0: # from private
msg.reply_text(help_message)
def getUrl(msg):
if matchKey(msg.text_html_urled, ['source</a>']):
return
if (matchKey(msg.text_html_urled,
['mp.weixin.qq.com', 'telegra.ph'])
and msg.chat.username == 'web_record'):
return
soup = BeautifulSoup(msg.text_html_urled, 'html.parser')
for item in soup.find_all('a'):
if 'http' in item.get('href'):
return item.get('href')
def wantSee(item, page): if matchKey(str(item), ['people/gyz', '4898454']): return True if matchKey(str(item), blocklist.items()): return False require = 120 + page if 'people/renjiananhuo' in str(item.parent): require *= 4 # 这人太火，发什么都有人点赞。。。 if sum(list(dataCount(item))[:3]) > require: return True return False
def getDoubanId(link):
if not matchKey(link, ['note', 'group/topic', 'status', 'album']):
return
if matchKey(link, ['notes', 'statuses']):
return
if 'http' not in link:
return
parts = link.split('/')
for part in parts[:-1]:
try:
int(part)
return part
except:
...
def getCnLink(link):
link = getRawLink(link)
if not link:
return
if not matchKey(
link,
['douban.', 'thepaper', 'weixin', 'zhihu.', 'cnpoliti', 'weibo']):
return
link = clearUrl(link)
# getTitle by default will cache
if matchKey(export_to_telegraph.getTitle(link),
['链接已过期', '仅限男生', '男生350分', '常玮平', '【来自投稿】20世纪初', '做了套打拳入门']):
return False
return link
def _findOrgName(soup):
head = str(soup.find('head'))
if matchKey(head, ['bbc.com']):
return 'BBC', True
if matchKey(head, ['nyt.com', 'new york times']):
return 'NYT', True
if matchKey(head, ['stackoverflow']):
return 'StackOverflow', False
if matchKey(head, ['medium.com']):
return 'Medium', False
if matchKey(head, ['dw.come']):
return 'DW', True
r = _findPossibleRawContent(_yieldPossibleOrgItem(soup))
if r:
return r, False
return 'Source', False
def log(url, card, key, channels, sent):
if weiboo.getCount(card) < 20:
return
whash = weiboo.getHash(card)
if not log_existing.add(whash):
return
additional_info = weibo_2_album.getAdditionalInfo(card['mblog'])
if additional_info:
additional_info += ' '
disable_web_page_preview = not matchKey(additional_info,
['imgs:', 'video:'])
if sent:
sent = ' weibo_bot_sent'
else:
sent = ''
if set([channel.id for channel in channels]) & core_channels_ids:
mark = ''
else:
mark = ' weibo_channel_ignore'
message = '%s\n\n%skey: %s channel_id: %s %s%s%s %s <a href="%s">source</a>' % (
weibo_2_album.getCap(card['mblog']), additional_info, key, ' '.join([
str(channel.id) for channel in channels
]), getChannelsLog(channels), sent, mark, url, url)
try:
logger.send_message(message,
parse_mode='html',
disable_web_page_preview=disable_web_page_preview)
except Exception as e:
print('log failed', str(e), message)
time.sleep(5)
def cleanupCap(text):
text = getPrintableForProd(text).strip()
soup = BeautifulSoup(text, features="html.parser")
result = []
for item in soup:
if item.name == 'br':
result.append('\n')
continue
if item.name == None:
result.append(str(item))
continue
if item.name == 'a':
link = item.get('href')
if urlShouldRemove(link):
continue
if (not link or matchKey(
link, ['weibo.cn/p', 'weibo.cn/search', 'weibo.com/show'])
or item.text[:1] == '@'):
result.append(item.text)
continue
result.append('<a href="%s">%s</a>' % (link, item.text))
text = ''.join(result).strip()
text = text.replace('\n', '\n\n')
for _ in range(5):
text = text.replace('\n\n\n', '\n\n')
return text.strip()
def handleCommand(update, context):
global ban
usr = update.effective_user
msg = update.effective_message
msg.forward(debug_group.id)
usr = usr.username
command, text = splitCommand(msg.text)
if matchKey(command, ['get', 'search']):
keys = text.split()
usrs = [x for x in db.usrs() if matchAll(db.getRaw(x), keys)]
usrs = [x for x in usrs if x != usr and x not in ban]
random.shuffle(usrs)
if usr != test_usr:
usrs = usrs[:LIMIT]
if not usrs:
return msg.reply_text(strings['e4'])
for x in usrs:
sendUsr(x, msg)
return
if not usr:
return msg.reply_text(strings['e0'])
if 'start' in command:
msg.reply_text(strings['h1'])
return askNext(usr, msg)
if 'question' in command:
for q in db.questions:
msg.reply_text(strings['q' + q])
return
if 'update' in command:
db.save(usr, 'key', text)
if not checkProfileFinish(usr, msg):
return
if 'preview' in command:
sendUsr(usr, msg)
return msg.reply_text(strings['h2'])
def processSite(site):
try:
links = link_extractor.getLinks(site)
except Exception as e:
print('web_bot, getLinks fail', str(e), site)
return
count = 0
if 'douban' in site and 'people' not in site:
limit = 1 # may change later
elif matchKey(site, ['tempo', 'kompas', 'nature.com']):
limit = 1
else:
limit = 20
for link in links:
if db.existing.contain(link):
continue
title = ''.join(export_to_telegraph.getTitle(link).split())
if db.existing.contain(title):
continue
success = sendLink(site, link)
db.existing.add(link)
db.existing.add(title)
count += 1
if (not success) or count >= limit:
return
def _isOffTopic(attrs):
if not attrs:
return
r = []
for k, v in attrs.items():
if k != 'data-component' and matchKey(
k, ['href', 'src', 'url', 'alt', 'data', 'xmlns:fb']):
continue
r.append(str(k) + ' : ' + str(v))
r = '\n'.join(r)
for att in OFFTOPIC_ATT:
if att in r:
return att
for att in OFFTOPIC_ATT_WITH_EXCEPTION:
if att in r and not matchKey(r, OFFTOPIC_ATT_WITH_EXCEPTION[att]):
return att
return
def getShortLink(link):
if matchKey(link, ['weibo.', 'twitter.', 't.me/']):
return
raw_link = getRawLink(link)
if isCN(export_to_telegraph.getTitle(raw_link)):
return shorter(raw_link, link)
if isCN(export_to_telegraph.getTitle(link)):
return link
def getVideo(b):
for video in b.find_all('video'):
if not video.parent or not video.parent.parent:
continue
wrapper = video.parent.parent
if not matchKey(str(wrapper.get('id')), ['video_info']):
continue
return video['src']
def search(keywords):
pool = Pool()
while pool.pool:
for name, pos in pool.items():
if matchKey(getSoup(name, pos).text, keywords):
print(getRootUrl(name, post))
next_pos = getNextPos(name, pos)
pool.update(name, next_pos)
def shouldKick(self, user):
if (self.kick_if_name_longer_than and
len(user.first_name or '') + len(user.last_name or '') >
self.kick_if_name_longer_than):
return True
if matchKey(getDisplayUser(user), self.kick_if_name_contains):
return True
return False
def remindIfNecessary(msg):
if not msg.text:
return
if matchKey(msg.text,
better_avoid_words) and not matchKey(msg.text, quotes):
reminder = '建议避免使用带有强烈主观判断的词哦，比如：' + ', '.join(better_avoid_words) + \
'。 谢谢啦！'
autoDestroy(msg.reply_text(reminder), 10)
emotional_words = ['意淫', '凭什么']
if matchKey(
msg.text,
emotional_words) or msg.text.count('?') + msg.text.count('？') >= 3:
reminder = '反问，反讽不利于友好交流哦，建议您换成大家更容易理解的表达哦。谢谢啦！'
autoDestroy(msg.reply_text(reminder), 10)
attacking_words = ['太low']
if matchKey(msg.text, attacking_words):
reminder = '请友好交流，争取互相理解。谢谢啦！'
autoDestroy(msg.reply_text(reminder), 10)
def decorate(text):
if 'http' not in text:
text = 'https://' + text
text = getRawLink(text)
if matchKey(text, [
'www.douban.com/people/', 'twitter.com', 'facebook.com',
'm.weibo.cn'
]):
return '\n\n' + text
return '\n\n【%s】 %s' % (export_to_telegraph.getTitle(text), text)
def getSrc(img):
src = img.get('src') and img.get('src').strip()
if not src:
return
if not img.parent or not img.parent.parent:
return
wrapper = img.parent.parent
if matchKey(str(wrapper.get('class')) or '', ['f-m-img', 'group-pic']):
return src
return
def getMatches(text):
if not text:
return []
if matchKey(text, ['/', '@']) or isInt(text):
user_id = getUserId(text)
if user_id and isInt(text):
return [user_id]
if user_id:
return [user_id, text]
return [text]
def handleCommand(update, context, dbs):
msg = update.effective_message
autoDestroy(msg, 0.1)
if msg.from_user and matchKey(msg.from_user.first_name, ['telegram']):
# don't deal with group message auto forwarded linked channel
return
command, text = splitCommand(msg.text)
if 's3_l' in command:
subscriptions = dbs.getList(msg.chat_id)
subscriptions = [str(index) + ': ' + \
formatChat(context.bot, x['id']) for \
index, x in enumerate(subscriptions)]
r = msg.reply_text('subscription list: \n\n' +
'\n'.join(subscriptions),
quote=False,
parse_mode='Markdown',
disable_web_page_preview=True)
autoDestroy(r)
return
if 's3_un' in command:
try:
index = int(text)
except:
r = msg.reply_text('please give index')
autoDestroy(r)
return
r = dbs.deleteIndex(msg.chat_id, index)
autoDestroy(msg.reply_text(r, quote=False))
return
if 's3_s' in command:
chat = getChat(context.bot, text)
if not chat:
return
r = dbs.add(msg.chat, chat)
autoDestroy(msg.reply_text(r, quote=False))
return
if not msg.from_user or msg.from_user.id != debug_group.id:
return
if 'repeat' in command:
msg.bot.send_message(msg.chat.id,
msg.reply_to_message.text_markdown,
parse_mode='Markdown',
disable_web_page_preview=True)
# guard this feature
# if 'all' in command:
# sendAll(msg, dbs)
# return
if 'delete' in command:
to_delete = msg.reply_to_message
key = (to_delete.chat_id, to_delete.message_id)
for r in forward_all_record[key]:
try:
r.delete()
except:
pass
