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

Add time-based media group merging for RSS and HTML generation

Browse Source
This commit is contained in:
vvzvlad
2025-02-09 21:44:55 +03:00
parent c554bd4c44
commit 67d52b507a
4 changed files with 116 additions and 27 deletions
Download Patch File Download Diff File Expand all files Collapse all files
api_server.py
+17 -3
View File
@@ -520,7 +520,13 @@ async def get_media(channel: str, post_id: int, file_unique_id: str, digest: str
@app.get("/rss/{channel}", response_class=Response)
@app.get("/rss/{channel}/{token}", response_class=Response)
async def get_rss_feed(channel: str, token: str | None = None, limit: int = 50, output_type: str = 'rss', exclude_flags: str | None = None):
async def get_rss_feed(channel: str,
token: str | None = None,
limit: int = 50,
output_type: str = 'rss',
exclude_flags: str | None = None,
merge_seconds: int = 5
):
if Config["token"]:
if token != Config["token"]:
logger.error(f"Invalid token for RSS feed: {token}, expected: {Config['token']}")
@@ -530,10 +536,18 @@ async def get_rss_feed(channel: str, token: str | None = None, limit: int = 50,
while True:
try:
if output_type == 'rss':
rss_content = await generate_channel_rss(channel, client=client.client, limit=limit, exclude_flags=exclude_flags)
rss_content = await generate_channel_rss(channel,
client=client.client,
limit=limit,
exclude_flags=exclude_flags,
merge_seconds=merge_seconds)
return Response(content=rss_content, media_type="application/xml")
elif output_type == 'html':
rss_content = await generate_channel_html(channel, client=client.client, limit=limit, exclude_flags=exclude_flags)
rss_content = await generate_channel_html(channel,
client=client.client,
limit=limit,
exclude_flags=exclude_flags,
merge_seconds=merge_seconds)
return Response(content=rss_content, media_type="text/html")
except ValueError as e:
error_message = f"Invalid parameters for RSS feed generation: {str(e)}"
config.py
+2 -1
View File
@@ -11,5 +11,6 @@ def get_settings():
"pyrogram_bridge_url": os.getenv("PYROGRAM_BRIDGE_URL", ""),
"log_level": os.getenv("LOG_LEVEL", "INFO"),
"debug": os.getenv("DEBUG", "False") == "True",
"token": os.getenv("TOKEN", "")
"token": os.getenv("TOKEN", ""),
"time_based_merge": os.getenv("TIME_BASED_MERGE", "False") == "True",
}
post_parser.py
+20 -12
View File
@@ -404,9 +404,9 @@ class PostParser:
def _generate_html_footer(self, message: Message) -> str:
content_footer = []
if reactions_views_html := self._reactions_views_links(message): # Add reactions, views and links
if reactions_views_html := self._reactions_views_links(message): # Add reactions, views, date and links
content_footer.append(reactions_views_html)
if flags_html := self._format_flags(message): # Add flags
if flags_html := self._format_flags(message): # Add flags
content_footer.append(flags_html)
html_footer = '\n'.join(content_footer)
html_footer = self._sanitize_html(html_footer)
@@ -475,14 +475,19 @@ class PostParser:
if views := getattr(message, "views", None):
views_html = f'<span class="views">{views} views</span>'
# If both reactions and views exist, insert a line break between them.
if reactions_html and views_html:
parts.append(reactions_html + "<br>" + views_html)
elif reactions_html:
parts.append(reactions_html)
elif views_html:
parts.append(views_html)
# Combine reactions and views on the same line
if reactions_html or views_html:
parts.append(f'{reactions_html}&nbsp;&nbsp;{views_html}')
# Add date and links on the same line
second_line_parts = []
# Format date as DD/MM/YY, HH:MM:SS
if message.date:
formatted_date = message.date.strftime("%d/%m/%y, %H:%M:%S")
second_line_parts.append(formatted_date)
# Add links
channel_identifier = self.get_channel_username(message)
if channel_identifier:
links = []
@@ -490,12 +495,15 @@ class PostParser:
channel_id = channel_identifier[4:] # Remove '-100' prefix for web links
links.append(f'<a href="tg://resolve?domain=c/{channel_id}&post={message.id}">Open in Telegram</a>')
links.append(f'<a href="https://t.me/c/{channel_id}/{message.id}">Open in Web</a>')
else: # For channels with username
else: # For channels with username
links.append(f'<a href="tg://resolve?domain={channel_identifier}&post={message.id}">Open in Telegram</a>')
links.append(f'<a href="https://t.me/{channel_identifier}/{message.id}">Open in Web</a>')
parts.append('&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;'.join(links))
second_line_parts.extend(links)
html = '&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;'.join(parts) if parts else None
if second_line_parts:
parts.append('&nbsp;&nbsp;&nbsp;&nbsp;'.join(second_line_parts))
html = '<br>'.join(parts) if parts else None
return f"<br>{html}" if html else None
except Exception as e:
rss_generator.py
+75 -9
View File
@@ -16,6 +16,49 @@ if not logger.handlers:
logger.addHandler(handler)
async def create_time_based_media_groups(messages, merge_seconds: int = 5):
messages_sorted = sorted(messages, key=lambda x: x.date)
cluster = []
last_msg_date = None
current_media_group_id = None
for msg in messages_sorted:
if not cluster:
cluster.append(msg)
last_msg_date = msg.date
current_media_group_id = getattr(msg, "media_group_id", None)
continue
time_diff = (msg.date - last_msg_date).total_seconds()
msg_media_group_id = getattr(msg, "media_group_id", None)
if time_diff <= merge_seconds:
if current_media_group_id:
msg.media_group_id = current_media_group_id
elif msg_media_group_id:
current_media_group_id = msg_media_group_id
for m in cluster:
m.media_group_id = current_media_group_id
cluster.append(msg)
last_msg_date = msg.date
else:
if len(cluster) >= 2 and not current_media_group_id:
new_group_id = f"time_{min(m.date for m in cluster)}"
for m in cluster:
m.media_group_id = new_group_id
cluster = [msg]
last_msg_date = msg.date
current_media_group_id = msg_media_group_id
if len(cluster) >= 2 and not current_media_group_id:
new_group_id = f"time_{min(m.date for m in cluster)}"
for m in cluster:
m.media_group_id = new_group_id
return messages
async def create_messages_groups(messages):
"""
Process messages into formatted posts, handling media groups
@@ -96,22 +139,29 @@ async def render_messages_groups(messages_groups, post_parser, exclude_flags: st
'author': message_data['author'],
'flags': message_data['flags']
})
else: # Multiple messages in group - need to merge
else: # Multiple messages in group - merge media, text and html body
processed_messages = [post_parser.process_message(msg) for msg in group]
main_message = next( # Find main message (one with text)
main_message = next( # Determine main message for header/footer/title
(msg for msg in processed_messages if msg['text']),
processed_messages[0] # fallback to first if none has text
processed_messages[0] # fallback if no message contains text
)
# Collect all media sections
# Collect all media sections from all messages in the group
all_media = [msg['html']['media'] for msg in processed_messages if msg['html']['media']]
combined_media = '\n'.join(all_media)
# Combine parts
# Merge text fields from all messages
all_texts = [msg['text'] for msg in processed_messages if msg['text']]
combined_text = '\n'.join(all_texts)
# Merge html body sections from all messages
all_html_bodies = [msg['html']['body'] for msg in processed_messages if msg['html']['body']]
combined_html_body = '\n'.join(all_html_bodies)
html_parts = [
f'<div class="message-header">{main_message["html"]["header"]}</div>',
f'<div class="message-media">{combined_media}</div>',
f'<div class="message-text">{main_message["html"]["body"]}</div>',
f'<div class="message-text">{combined_html_body}</div>',
f'<div class="message-footer">{main_message["html"]["footer"]}</div>'
]
@@ -120,7 +170,7 @@ async def render_messages_groups(messages_groups, post_parser, exclude_flags: st
'date': main_message['date'],
'message_id': main_message['message_id'],
'title': main_message['html']['title'],
'text': main_message['text'],
'text': combined_text,
'author': main_message['author'],
'flags': main_message['flags']
})
@@ -147,7 +197,13 @@ async def render_messages_groups(messages_groups, post_parser, exclude_flags: st
rendered_posts.sort(key=lambda x: x['date'], reverse=True)
return rendered_posts
async def generate_channel_rss(channel: str, post_parser: Optional[PostParser] = None, client = None, limit: int = 20, exclude_flags: str | None = None ) -> str:
async def generate_channel_rss(channel: str,
post_parser: Optional[PostParser] = None,
client = None,
limit: int = 20,
exclude_flags: str | None = None,
merge_seconds: int = 5
) -> str:
"""
Generate RSS feed for channel using actual messages
Args:
@@ -199,6 +255,8 @@ async def generate_channel_rss(channel: str, post_parser: Optional[PostParser] =
messages.append(message)
# Process messages into groups and render them
if Config['time_based_merge']:
messages = await create_time_based_media_groups(messages, merge_seconds)
message_groups = await create_messages_groups(messages)
message_groups = await trim_messages_groups(message_groups, limit)
final_posts = await render_messages_groups(message_groups, post_parser, exclude_flags)
@@ -231,7 +289,13 @@ async def generate_channel_rss(channel: str, post_parser: Optional[PostParser] =
raise
async def generate_channel_html(channel: str, post_parser: Optional[PostParser] = None, client = None, limit: int = 20, exclude_flags: str | None = None) -> str:
async def generate_channel_html(channel: str,
post_parser: Optional[PostParser] = None,
client = None,
limit: int = 20,
exclude_flags: str | None = None,
merge_seconds: int = 5
) -> str:
"""
Generate HTML feed for channel using actual messages
Args:
@@ -272,6 +336,8 @@ async def generate_channel_html(channel: str, post_parser: Optional[PostParser]
messages.append(message)
# Process messages into groups and render them
if Config['time_based_merge']:
messages = await create_time_based_media_groups(messages, merge_seconds)
message_groups = await create_messages_groups(messages)
message_groups = await trim_messages_groups(message_groups, limit)
final_posts = await render_messages_groups(message_groups, post_parser, exclude_flags)