vvzvlad/pyrogram-bridge
1
0
Fork 0
Code Issues 6 Pull Requests Actions 8 Packages Projects Releases Wiki Activity

add only_link flag

Browse Source
This commit is contained in:
vvzvlad
2025-04-13 18:59:57 +03:00
parent 7199a955cd
commit 09d3fcf5a1
2 changed files with 57 additions and 15 deletions
Download Patch File Download Diff File Expand all files Collapse all files
post_parser.py
+38 -15
View File
@@ -288,7 +288,10 @@ class PostParser:
return None
def _extract_flags(self, message: Message) -> List[str]:
message_text = self._generate_html_body(message)
# Use raw text/caption for some checks before HTML processing
message_text_str = str(message.text or message.caption or '')
# Use HTML body for checks involving formatted text or links within HTML
message_body_html = self._generate_html_body(message)
flags = []
# Add "fwd" flag for forwarded messages
@@ -310,19 +313,19 @@ class PostParser:
# Check if the message text contains variations of the word "стрим", "вебинар"
# or "онлайн-лекция" in a case-insensitive manner.
if re.search(r'(?i)\b(стрим\w*|livestream|онлайн-лекци[яю]|вебинар\w*)\b', message_text):
if re.search(r'(?i)\b(стрим\w*|livestream|онлайн-лекци[яю]|вебинар\w*)\b', message_text_str):
flags.append("stream")
# Check if the message text contains the word "донат" in a case-insensitive manner.
if re.search(r'(?i)\bдонат\w*\b', message_text):
if re.search(r'(?i)\bдонат\w*\b', message_text_str):
flags.append("donat")
# Check for pay.cloudtips.ru links and add donat flag
if re.search(r'(?i)pay\.cloudtips\.ru', message_text):
if re.search(r'(?i)pay\.cloudtips\.ru', message_text_str):
flags.append("donat")
# Check for t.me/boost links and add donat flag
if re.search(r'https?://(?:www\.)?t\.me/boost/', message_text):
if re.search(r'https?://(?:www\.)?t\.me/boost/', message_text_str):
flags.append("donat")
# Check if the post's reactions contain more clown emojis (🤡) or poo emojis (💩).
@@ -336,30 +339,50 @@ class PostParser:
break
# Check if the message text contains "#реклама", "Партнерский пост", "по промокоду", "скидка на курс", "регистрируйтесь тут" in a case-insensitive manner.
if re.search(r'(?i)(#реклама|#промо|О\s+рекламодателе|партнерский\s+пост|по\s+промокоду|erid|скидка\s+на\s+курс|регистрируйтесь\s+тут)', message_text):
if re.search(r'(?i)(#реклама|#промо|О\s+рекламодателе|партнерский\s+пост|по\s+промокоду|erid|скидка\s+на\s+курс|регистрируйтесь\s+тут)', message_text_str):
flags.append("advert")
# Check for paywall-related words and tags
if re.search(r'(?i)(Дзен\.Премиум|Sponsr|Бусти|Boosty)', message_text):
if re.search(r'(?i)(Дзен\.Премиум|Sponsr|Бусти|Boosty)', message_text_str):
flags.append("paywall")
# Check if the message contains any http/https links in text or href attributes (excluding t.me links)
if (re.search(r'https?://(?!(?:www\.)?t\.me)[^\s<>"\']+', message_text) or
re.search(r'href=["\']https?://(?!(?:www\.)?t\.me)[^"\']+["\']', message_text)):
flags.append("link")
# --- Link Flags ---
# Check if the message consists ONLY of a link (text/caption) or only has a webpage preview
is_only_link = False
if message.web_page and not message_text_str.strip():
# Case 1: Only a web_page preview, no text/caption
is_only_link = True
elif message_text_str.strip():
# Case 2: Text/caption exists, check if it's just a single non-t.me URL
# Regex matches string starting and ending with a non-t.me URL
url_pattern = r'^https?://(?!(?:www\.)?t\.me)[^\s<>"\']+$'
if re.fullmatch(url_pattern, message_text_str.strip()):
is_only_link = True
if is_only_link:
flags.append("only_link")
# Check if the message contains any http/https links (excluding t.me)
# Only add 'link' if 'only_link' isn't already set
if not is_only_link:
# Search within the generated HTML body to catch links in hrefs as well
if (re.search(r'https?://(?!(?:www\.)?t\.me)[^\s<>"\']+', message_body_html) or
re.search(r'href=[\"\']https?://(?!(?:www\.)?t\.me)[^\"\']+[\"\']', message_body_html)):
flags.append("link")
# --- End Link Flags ---
# Check if the message contains channel mentions in the format @name
if re.search(r'@[a-zA-Z][a-zA-Z0-9_]{3,}', message_text):
if re.search(r'@[a-zA-Z][a-zA-Z0-9_]{3,}', message_body_html):
flags.append("mention")
try:
# Find links with a '+' after t.me/ indicating a hidden channel link.
hidden_links = re.findall(r'https?://(?:www\.)?t\.me/\+([A-Za-z0-9]+)', message_text)
hidden_links = re.findall(r'https?://(?:www\.)?t\.me/\+([A-Za-z0-9]+)', message_body_html)
# Find links without a '+' after t.me/ indicating an open (foreign) channel link.
open_links = re.findall(r'https?://(?:www\.)?t\.me/(?!\+)([A-Za-z0-9_]+)', message_text)
open_links = re.findall(r'https?://(?:www\.)?t\.me/(?!\+)([A-Za-z0-9_]+)', message_body_html)
# Find links with pattern t.me/boost/channel_name
boost_links = re.findall(r'https?://(?:www\.)?t\.me/boost/([A-Za-z0-9_]+)', message_text)
boost_links = re.findall(r'https?://(?:www\.)?t\.me/boost/([A-Za-z0-9_]+)', message_body_html)
if hidden_links:
flags.append("hid_channel")
tests/test_post_parser_flags.py
+19
View File
@@ -39,6 +39,7 @@ class TestPostParserExtractFlags(unittest.TestCase):
message.media = media
message.id = 123 # Add a default message ID
message.forward_origin = None # Add missing attribute
message.web_page = None # Add missing attribute checked in _extract_flags
# Mock chat attribute
mock_chat = MagicMock(spec=Chat)
@@ -126,6 +127,7 @@ class TestPostParserExtractFlags(unittest.TestCase):
message.media = MessageMediaType.VIDEO
message.id = 123
message.forward_origin = None # Add missing attribute
message.web_page = None # Add missing attribute checked in _extract_flags
# Create a basic mock Chat
mock_chat = MagicMock(spec=Chat)
@@ -261,6 +263,23 @@ class TestPostParserExtractFlags(unittest.TestCase):
_ = self.parser._generate_html_body(message) # Call body generation which prepares text
self.assertIn("link", self.parser._extract_flags(message))
def test_flag_only_link_in_text(self):
message = self._create_mock_message(text="https://example.com")
# Use _generate_html_body to ensure link detection logic runs
_ = self.parser._generate_html_body(message) # Call body generation which prepares text
self.assertIn("only_link", self.parser._extract_flags(message))
def test_flag_only_webpage(self):
# Create a message with no text/caption first
message = self._create_mock_message()
# Manually set the web_page attribute to simulate a webpage preview
message.web_page = MagicMock() # Use MagicMock to simulate presence
# Ensure text and caption are None for the 'only_link' logic
message.text = None
message.caption = None
# The flag logic should now correctly identify this as only_link
self.assertIn("only_link", self.parser._extract_flags(message))
def test_flag_link_in_href(self):
# Need to mock html property correctly
message = self._create_mock_message(text="Check this <a href='http://test.com'>link</a>")