Setting up a minimal example of an application using the emubot framework) is simple. But to leverage the framework’s advantages you might want to use some of the functionality (e.g. to pseudonymize the user) and connect your application to a database. This repository aims at providing exactly that.
This repository uses the emubot npm module to provide functionality for real use cases. Some of the examples are not useful if you use the default behavior, but require (any) working Chat- and NlpAdapter and the respective configuration. For more information, please take a look at the documentation
We provide some examplary setup files for different messengers and NLP platforms. What you need is the following minimal structure:
root
│ main.ts
|
└───configuration/
│ │ config.ts
│
└───interceptors
│ ....
│ ...
A valid configuration file is required to run your bot, more information on their formatcan be found in the documentation. Most of the presented functionalities are separated by using distinctive interceptors. These interceptors can be combined to e.g. include a database as well as calling external APIs in a single interceptor.
Make sure that you have installed npm
and install the dependencies using npm -i
.
Compile the typescript code using npm run tsc
and start the bot with npm run start
. This will use the default configuration: make sure to change it in advance.
A rudimentary database implementation is provided with this repo. The implementation is based on the TypeORM package and should be used with a postgres database. In order to use the database you will need to setup a postgres database and provide a valid configuration for the database in the src/configuration/databasePostgresConfig.ts
file. If you do not have an existing postgres database, an easy way to set one up is by using the postgres docker image. If you do not have docker installed please refer to this tutorial. After you have installed docker you can run the following commands to start the postgres docker image
docker pull postgres
docker run --name some-postgres-container-name -p 5432:5432 -e
POSTGRES_PASSWORD=password_set_as_environmentvar -d postgres
This will start the postgres docker container listening for connections on port 5432. The next step will be to create the database inside the docker container. You can log into the docker container with the command
docker exec -it some-postgres-container-name /bin/bash
where you replaced some-postgres-container-name
with the name of your container. Inside the docker container, you will need to create database:
su postgres
psql
CREATE DATABASE databaseName;
Afterwards a \l
command should list your database
postgres=# \l
List of databases
Name | Owner | Encoding | Collate | Ctype | Access privileges
-----------+----------+----------+------------+------------+-----------------------
postgres | postgres | UTF8 | en_US.utf8 | en_US.utf8 |
template0 | postgres | UTF8 | en_US.utf8 | en_US.utf8 | =c/postgres +
| | | | | postgres=CTc/postgres
template1 | postgres | UTF8 | en_US.utf8 | en_US.utf8 | =c/postgres +
| | | | | postgres=CTc/postgres
databaseName | postgres | UTF8 | en_US.utf8 | en_US.utf8 |
If all looks good and you can see your database in the list you have to provide the database configuration to the framework:
const databasePostgresConfig: DatabaseConfig = {
host: 'server-hostname', // <- your host URL
name: 'name-of-database', // <- your database name
password: 'password_set_as_environmentvar', // <- your database password
port: 5432, // <- your database port
type: 'postgres',
user: 'postgres', // <- your database user name
entities: [string]; // Directory of the entities
};
Pseudonymization (using a database to store the identifiers) is implemented with the DatabasepseudonymizationInterceptor
and DatabaseDepseudonymizationInterceptor
. In order to use those interceptors simply set them up in the configuration file (Chat to Core and Nlp to Core interceptors).
In order to stop a bot a NlpControlledPauseInterceptor
is provided. This interceptor requires a database to save the state of the conversation (paused or unpaused). The pause will be triggered by a fallback intent and resume if an intent with the "unpause" action is detected. For every user the current state of the interaction is saved in the database. This interceptor could serve as baseline for some clever logic to inform your staff that one of your bot conversations has some issues that need to be addressed (fix the bot, contact the user personally).
Please send a mail to emubot@e-mundo.de if you have any further questions/inquiries.