1. Is it right that this is bot-api to client-api proxy for bots?
2. Can we extend this server api with methods that exist in client api and can be called by bots but does not exist in bot api?

Awesome! If someone would be run bot-api on cloud native environment pull requests with Dockerfiles will be accepted or users must implement it on their side?

My server has an 8-core CPU. Currently, only 1 CPU is used, and the other 7 are idle. I hope to reply, thank you!

DURATION inf 5sec 1min 1hour uptime 2482.520765 bot_count 4379 active_bot_count 150 rss 18GB vm 18GB rss_peak 18GB vm_peak 18GB total_cpu 118.269319% 113.913043% 114.552656% 118.269319% user_cpu 99.416224% 100.347826% 99.993907% 99.416224% system_cpu 18.853095% 13.565217% 14.558749% 18.853095% buffer_memory 3040MB active_webhook_connections 3281 active_requests 174706 active_network_queries 172347 request_count 614.006956 364.547365 294.479924 614.006956 request_bytes 271470.504437 166391.486635 129315.445543 271470.504468 request_file_count 0.000000 0.000000 0.000000 0.000000 request_files_bytes 0.000000 0.000000 0.000000 0.000000 request_max_bytes 0 0 0 0 response_count 543.640529 347.578607 347.736007 543.640529 response_count_ok 76.525494 107.880848 70.113676 76.525494 response_count_error 467.115035 239.697759 277.622331 467.115035 response_bytes 136281.333011 162937.389112 112554.637102 136281.333026 update_count 1348.788008 1905.333100 2052.825080 1348.788008

Hello! I am making small bot-api server based on MTProto. In some cases i use api.telegram.org/getme. When i revoke token and re-call api.telegram.org/getme its hangs my code i thought i have a problem with my code but when i check logs i found that its hangs on starting new https connection to api.telegram.org/bot../getme after some seconds its returns api.telegram.org/bot../getme 504 Gateway Timeout . Means if i have new Token and call api.telegram.org in the first request i get 504 Gateway Timeout in the second request i get 200 OK

Sorry for the bad explanation 😅

I tried to build for android on linux debian 10 intel 64 bit but there is a failure

I have installed all the required packages but when I type this command a log appears

cmake -DCMAKE_BUILD_TYPE=Release ..

error log

-- Could NOT find ZLIB (missing: ZLIB_LIBRARY) (found version "1.2.11")
-- Could NOT find ZLIB (missing: ZLIB_LIBRARY) (found version "1.2.11")
CMake Warning at td/CMakeLists.txt:181 (message):
  Can't find zlib: stop TDLib building

CMake Error: The following variables are used in this project, but they are set to NOTFOUND.
Please set them or make sure they are set and tested correctly in the CMake files:
ZLIB_LIBRARY
    linked by target "tdutils" in directory /home/azka/telegram-bot-api/td/tdutils

-- Configuring incomplete, errors occurred!
See also "/home/azka/telegram-bot-api/build/CMakeFiles/CMakeOutput.log".
See also "/home/azka/telegram-bot-api/build/CMakeFiles/CMakeError.log".

Hi, I have a question about https://api.telegram.org also tdlib.

I have my high loaded bot 100 update per second i am sure in weekend 100+ update per second Its group management bot

I want to rebuild the bot source but I'm confused what to use Tdlib or api.telegram.org webhook, Who better?

If i use Tdlib will use more memory and more disk space (i try to login using my bot in the first 5 minutes db.sqlite gets 55M size)

If i use api.telegram.org will be lot easier dealing with updates but i will lose access to some methods that not available on Bot-API like get user information by username etc and i need that .

I can use two instance api.telegram.org and MTProto Client (for other methods not available on Bot-API) I don't think is the right way to use that.

@levlam If it was you, what would you choose and why?

An EOF error occurs when I upload a file that is larger than about 1GB（local mode）, even though the file is still sent successfully. This doesn't happen when I send files around 500MB. The POST request returns to the EOF about halfway through the file, but the request has been interrupted. After the result is returned, the file is still sent by the API.

Context: https://github.com/teloxide/teloxide/issues/547

After some testing, we can use MessageEntity's text_mention type to mention users when sending messages, and the required fields of user are id only. But this behavior is undocumented, may we document it?

I receive "until_date": 0 in ChatMemberBanned object for banned forever users, which doesn't make sense. Is this intentional behaviour? Could this field be optional for such occurrences?

Hello, I've created a bot for interacting with my website (displaying flights data). after activating the bot, and running it successfully, I got many errors on my web servers' log: "HTTP Header Injection Attack via payload (CR/LF detected)"

I've done some digging and I believe the source is the "\n" here: "ARGS_NAMES:{"update_id":38xxxxxxx,\n"message":{"message_id":3xxx, ..." (after the "update_id" and before the "message") If I'm not mistaken, the LineFeed mark is generated in "telegram-bot-api/WebhookActor.h", line 217 : object << td::JsonRaw(",\n");

Is the "\n" a part of a mandatory args structure? Is there a way to remove it? Any other suggestion?

Hi. I've install telegram-bot-api on my server based on "Telegram Bot API server build instructions generator" part with clang compiler in /usr/local directory and run it with this command :

telegram-bot-api --api-id=... --api-hash=... --local --dir=/var/log/tg --max-webhook-connections=100000 --log=/var/log/tg/log.tg --max-connections=1000

(api id and api hash that i wroted is fake). so then i changed the telegram api of the source of my bot (php) to the ip address of my server. I loged out the bot with logOut method, too. now some part of my bot work fast and some part have 1 ~ 2 second delay for responsing. I tried to examine the source code and requests and i find out that the handling requests with sendmessage method that have short text are faster than longer one. How can i fix it? Is there any misconfiguration while running telegram-bot-api?

I use it on server with 3 Gb of ram and 3 core of cpu (3.07Ghz). operating system : ubuntu 18 webserver : apache

note that I don't use ssl with domain . I use ip address and port number for sending requests and setwebhook.

I setted up nginx server as in aiogram/telegram-bot-api/example repo. Without local mode files downloading without any problems but when I turn local mode on bot returns this error: aiohttp.client_exceptions.ClientResponseError: 404, message='Not Found', url=URL('http://localhost:80/file/bot<token>//var/lib/telegram-bot-api/<token>/documents/file_3.p12')

Hi, I have a problem with telegram bot server on my server. I have created a service in Linux Ubuntu for telegram bot api with this details:

`[Unit] Description=Telegram bot api on Ubuntu

[Service] #WorkingDirectory=/root/telegram-bot-api/bin ExecStart=/root/telegram-bot-api/bin/telegram-bot-api --local --api-id=12345 --api-hash=[my hash id] Restart=always RestartSec=10 # Restart service after 10 seconds if dotnet service crashes SyslogIdentifier=telegram-bot-api-service

[Install] WantedBy=multi-user.target

`

When I call getFile api with this data, I got Bad Request: file is too big.

Is there any solution for this problem?

Is it possible for you to add a option for customizing longer timeout at least for local servers? The default is 30 seconds, I find myself in some situations like encoding a long video for the user need longer than 30 seconds.

What's the problem

I ran into a problem with handling media group messages, in other words — messages with multiple medias attached. The current version of bot API sends each media in a separate message and adds mediaGroupID to it, which makes sense since each media can have their own caption.

However, it makes a problem when you try to handle a whole group and store it, for instance. That happens because the code doesn't know when sending media group stops, only which message belong to media group and which not.

The only way I see is to make a timer, like for a second, during which the bot will look for messages with specific mediaGroupID. After the expiration of time, if no other message with mediaGroupID sent, the timer stops. Obviously, this is a terrible solution, so there should be another way to do it (if you have some idea, feel free to write them below).

How to solve a problem

To solve this problem, just let the bot know how much messages a group contains. There are two ways how to do it:

• Put number of messages in group right into API response, next to the mediaGroupID for each message from the group;
• Create new method, that receives a mediaGroupID and returns a number of messages in group, like that: https://api.telegram.org/bot<token>/mediaGroupSize?media_group_id=<your id>