10 KiB
TelegramBotAPI Util Extensions
Table of contents generated with markdown-toc
What is it?
It is wrapper library for TelegramBotAPI Core. Currently, this library contains some usefull filters for commands, updates types and different others.
How to implement library?
Common ways to implement this library are presented here. In some cases it will require additional steps
like inserting of additional libraries (like kotlin stdlib
). In the examples will be used variable
telegrambotapi-extensions-utils_version
, which must be set up by developer. Available versions are presented on
bintray, next version is last published:
Maven
Dependency config presented here:
<dependency>
<groupId>dev.inmo</groupId>
<artifactId>tgbotapi.extensions.utils</artifactId>
<version>${telegrambotapi-extensions-utils_version}</version>
</dependency>
Gradle
To use last versions you will need to add one line in repositories block of your build.gradle
:
mavenCentral()
And add next line to your dependencies block:
implementation "dev.inmo:tgbotapi.extensions.utils:$telegrambotapi-extensions-utils_version"
or for old gradle:
compile "dev.inmo:tgbotapi.extensions.utils:$telegrambotapi-extensions-utils_version"
How to use?
Here will be presented several examples of usage. In all cases it is expected that you have created your bot and filter:
val bot: RequestsExecutor = KtorRequestsExecutor(
TelegramAPIUrlsKeeper(BOT_TOKEN)
)
val filter = FlowsUpdatesFilter(64)
Alternative way to use the things below:
val filter = bot.startGettingFlowsUpdatesByLongPolling(
scope = CoroutineScope(Dispatchers.Default)
) {
// place code from examples here with replacing of `filter` by `this`
}
Updates
As mentioned in Telegram Bot API reference, there are two ways for updates retrieving:
- Webhooks
- Long Polling
Both of them you could use in your project using TelegramBotAPI Core, but here there are several useful extensions for both of them.
Anyway, in both of ways it will be useful to know that it is possible to create UpdateReceiver
object using function
flowsUpdatesFilter
:
val internalChannelsSizes = 128
flowsUpdatesFilter(internalChannelsSizes/* default is 64 */) {
textMessages().onEach {
println("I have received text message: ${it.content}")
}.launchIn(someCoroutineScope)
/* ... */
}
Long polling
The most simple way is Long Polling and one of the usages was mentioned above:
val filter = bot.startGettingFlowsUpdatesByLongPolling(
scope = CoroutineScope(Dispatchers.Default)
) {
// place code from examples here with replacing of `filter` by `this`
}
Extension startGettingFlowsUpdatesByLongPolling
was used in this example, but there are a lot of variations of
startGettingOfUpdatesByLongPolling
and others for getting the same result. Usually, it is supposed that you already
have created filter
object (or something like this) and will pass it into extension:
val filter = FlowsUpdatesFilter(64)
bot.startGettingOfUpdatesByLongPolling(
filter
)
But also there are extensions which allow to pass lambdas directly:
bot.startGettingOfUpdatesByLongPolling(
{
println("Received message update: $it")
}
)
Anyway, it is strictly recommended to pass your CoroutineScope
object to this method at least for more comfortable
management of updates.
WebHooks (currently JVM-only)
For webhooks there are less number of functions and extensions than for Long Polling (but it is still fully automated):
startListenWebhooks(
8081,
CIO // require to implement this engine dependency
) {
// here will be all updates one by one in $it
}
Besides, there are two additional opportunities:
- Extension
Route#includeWebhookHandlingInRoute
, which allow you to include webhook processing inside your ktor application without creating of new one server (as it is happening instartListenWebhooks
)- Also, you can use
Route#includeWebhookHandlingInRouteWithFlows
to use it likeflowUpdatesFilter
fun, but applyFlowsUpdatesFilter
to the block
- Also, you can use
- Extension
RequestsExecutor#setWebhookInfoAndStartListenWebhooks
. It is allow to set up full server (in fact, withstartListenWebhooks
), but also sendSetWebhook
request before and check that it was successful
Filters
There are several filters for flows.
Updates
In the next table it is supposed that you are using some Flow
with type from Base type of update
and apply
extension Extension
and will get Flow
with type from Result type of update
column.
Base type of update | Extension | Result type of update |
---|---|---|
Update |
onlyBaseMessageUpdates |
BaseMessageUpdate |
BaseMessageUpdate |
onlySentMessageUpdates |
BaseSentMessageUpdate |
BaseMessageUpdate |
onlyEditMessageUpdates |
BaseEditMessageUpdate |
BaseMessageUpdate |
onlyMediaGroupsUpdates |
MediaGroupUpdate |
MediaGroupUpdate |
onlySentMediaGroupUpdates |
SentMediaGroupUpdate |
MediaGroupUpdate |
onlyEditMediaGroupUpdates |
EditMediaGroupUpdate |
All of these extensions was made for more simple work with the others:
val flow: Flow<BaseMessageUpdate> = ...; // here we are getting flow from somewhere,
// for example, FlowsUpdatesFilter#messageFlow
flow.onlySentMessageUpdates().filterExactCommands(Regex("start"))
Here we have used filter filterExactCommands
which will pass only ContentMessage
with only one command start
Sent messages
All sent messages can be filtered for three types:
Type | Description | Flow extension |
---|---|---|
Common messages | Simple messages with text, media, location, etc. | asContentMessagesFlow |
Chat actions | New chat member, rename of chat, etc. | asChatEventsFlow |
Unknown events | Any other messages, that contain unsupported data | asUnknownMessagesFlow |
Common messages
Unfortunately, due to the erasing of generic types, when you are using asContentMessagesFlow
you will retrieve
data with type ContentMessage<*>
. For correct filtering of content type for retrieved objects, was created special
filters:
Content type | Result type | Flow extension |
---|---|---|
Animation | ContentMessage<AnimationContent> |
onlyAnimationContentMessages |
Audio | ContentMessage<AudioContent> |
onlyAudioContentMessages |
Contact | ContentMessage<ContactContent> |
onlyContactContentMessages |
Dice | ContentMessage<DiceContent> |
onlyDiceContentMessages |
Document | ContentMessage<DocumentContent> |
onlyDocumentContentMessages |
Game | ContentMessage<GameContent> |
onlyGameContentMessages |
Invoice | ContentMessage<InvoiceContent> |
onlyInvoiceContentMessages |
Location | ContentMessage<LocationContent> |
onlyLocationContentMessages |
Photo | ContentMessage<PhotoContent> |
onlyPhotoContentMessages |
Poll | ContentMessage<PollContent> |
onlyPollContentMessages |
Sticker | ContentMessage<StickerContent> |
onlyStickerContentMessages |
Text | ContentMessage<TextContent> |
onlyTextContentMessages |
Venue | ContentMessage<VenueContent> |
onlyVenueContentMessages |
Video | ContentMessage<VideoContent> |
onlyVideoContentMessages |
VideoNote | ContentMessage<VideoNoteContent> |
onlyVideoNoteContentMessages |
Voice | ContentMessage<VoiceContent> |
onlyVoiceContentMessages |
For example, if you wish to get only photo messages from private chats of groups, you should call next code:
filter.messageFlow.asContentMessagesFlow().onlyPhotoContentMessages().onEach {
println(it.content)
}.launchIn(
CoroutineScope(Dispatchers.Default)
)
Chat actions
Chat actions can be divided for three types of events source:
Type | Flow extension |
---|---|
Channel events | onlyChannelEvents |
Group events | onlyGroupEvents |
Supergroup events | onlySupergroupEvents |
According to this table, if you want to add filtering by supergroup events, you will use code like this:
filter.messageFlow.asChatEventsFlow().onlySupergroupEvents().onEach {
println(it.chatEvent)
}.launchIn(
CoroutineScope(Dispatchers.Default)
)
Shortcuts
With shortcuts you are able to use simple factories for several things.
ScheduledCloseInfo
In case if you are creating some poll, you able to use next shortcuts.
Next sample will use info with closing at the 10 seconds after now:
closePollExactAt(DateTime.now() + TimeSpan(10000.0))
In this example we will do the same, but in another way:
closePollExactAfter(10)
Here we have passed 10
seconds and will get the same result object.
In opposite to previous shortcuts, the next one will create approximate
closing schedule:
closePollAfter(10)
The main difference here is that the last one will be closed after 10 seconds since the sending. With first samples will be created exact time for closing of poll