Yawarakai Core

A running core that support many neural network models which works for NLP or providing solution
Based on Telegram Bot API and CLI for the main interfaces

Installation

git clone https://github.com/hanamiyuna/yawarakai.git
npm install # or yarn install

Configuration

cp config.json.example config.json
# Edit the info inside config.json

Start an instance

npm start
# or
node app.js start
# add --debug flag if you want it to be debug mode
node app.js start --debug

Component API

Yawarakai provides a dynamical and powerful API for users to control the message flows and the way the users could do through the core system

Credits

Query feature of Wiki and Component loader by BlueCocoa zh-CN, zh-TW, zh-HK, jp-JP Translation by KagurazakaIzumi

Change log

1.4.3

Mood Update

1. Code organization according to Codacy
2. Code reform as a new directory for processor script
3. Added a new API feature as Session
4. Added a new handler as callbackquery for System
5. Added many new command for admin
6. Added a new feature for admin as Broadcast to broadcast message to all users
7. Fixed several typos and prompts for commands
8. Optimized the inlineQuery functions
9. Fixed MusicShare command /netease function returning message while still WIP
10. Fixed the broadcast function is not working propertly
11. Fixed the tos message might try to send to group but permission is not exist

1.4.2

Scene Update

1. Code organization according to Codacy
2. Database seperated into different groups
3. Feature: Scene added and component API adapted
   Scene is designed for the stage control, scene, which is the meaning itself, the situation. Based on different situation, messages or commands may be processed differently, this won't effect the inline, channel post, callbackquery.
   Scene will only take place when the message or the trigger command has been executed, when a user is inside the scene, Yawarakai will ignore the other message distributor or the processor to the other components, the user's message will be redirected to the scene registered function to process in its own. This made a great experience where you need to make a wizard or make a short or 'question' based setup while in the interaction with user.
4. Formal documentation added
5. Two of the official component updated with Scene feature

1.4.1

Data Update Discontinued the Redis server support of this project. We now use NeDB for the local database storage.

1. Redis is now not used in this project and has been removed and related localization and config settings
2. Logs now will be saved into logs according to the regulation of Node.js project
3. Settings of database files is inside of the config file, update before startup
4. Fixed the flight component when the String of flight number cannot be matched in AA-0000 pattern
5. Fixed the musicshare component where the tag of song #s12345678 missing with letter s
6. Fixed the NLP display error after using NeDB

1.4.0

Wonderful Update

1. Added base command /info, /help, /start
2. Diagnostic log updated with new component feature
3. Message log format has been changed
4. Admin settings now available
5. Added commands for official component MusicShare
6. Fixed sendChatAction error
7. Fixed NLP processing issue
8. Fixed many issue

1.3.0

1. Added ja-JP as Japanese support
2. Fixed the issue about flight and wiki inline query

1.2.6

1. Added zh-TW as Chinese (Taiwan) support
2. Added zh-HK as Chinese (Hongkong) support
3. Added the feature that locale files will be automatically loaded
4. Added some detail for components
5. Added a new component flight as official component with flight search and track
6. Fixed MusicShare hashtag, commands, and data processing issue
7. Fixed several processing issue with commands and messages