Pathways

Install dependencies

Install on your local machine (Docker and psql):

sudo apt install docker.io docker-compose postgresql-client

Running the setup

To run the app and set the passwords use (assuming a Linux computer the following is one command):

POSTGRES_PASSWORD="secretpassword" \
PGADMIN_DEFAULT_EMAIL=example@example.com PGADMIN_DEFAULT_PASSWORD=example \
SQLALCHEMY_DATABASE_URI=postgresql://postgres:${POSTGRES_PASSWORD}@db/postgres \
docker-compose -f docker-compose.yml up --build

The --build ensures that the image is rebuild from the Dockerfile.

Uploading the data

When the setup is running (i.e. after docker-compose up), you can connect from your local machine using psql. To do that, determine which Docker network the setup is using:

docker network ls

You should see something like:

d8b0289eebdc        pathways_default    bridge              local

Then inspect the network:

docker network inspect pathways_default

You will get JSON where you can find IP address of the database machine:

            ...
            "c0e2970e62a3a7c1abc23649a08b9317158e22edae84288ba02ce5b9c9ed676e": {
                "Name": "pathways_db_1",
                "EndpointID": "c230ed3e70738a699c924135719365e0d93dfde0fdb588cf27b1d74bd9188a55",
                "MacAddress": "02:42:ac:12:00:02",
                "IPv4Address": "172.18.0.2/16",
                "IPv6Address": ""
            }

Then you can connect load the data from a dump to the database:

psql -h 172.18.0.2 -U postgres -f database.sql

Performing SQL queries

To just connect to the database and perform SQL queries, use the instructions above to get to the database container IP address and then use psql:

psql -h 172.18.0.2 -U postgres

Running web dashboard

Restart the containers (for the first time database is empty) and go to http://localhost:8050/

Windows 10 Installation

1. Modify docker-compose-win10.yml

Replace {source file} with a your database.sql file path

  db:
    image: postgres
    restart: always
    environment:
      POSTGRES_PASSWORD: ${POSTGRES_PASSWORD}
    volumes:
        - {source file}:/docker-entrypoint-initdb.d/database.sql

example:

C:/Users/mshukun/Desktop/database.sql:/docker-entrypoint-initdb.d/database.sql

Note: You can set folder to volume instead of file such as:

C:/Users/mshukun/Desktop/data:/docker-entrypoint-initdb.d/

In this case, all file contents in the data folder will be copied in docker-entrypoint-initdb.d folder, but sql file will not be loaded into database. You will need to load the data after successfuly build Docker containers.

docker exec -it pathways-visualization-tool_db_1 bash

psql -U postgres -f /docker-entrypoint-initdb.d/pathways.sql

2. Modify .env file

The .env file in the same directory as .yml files sets environment variables. The content of curent .env file is example. Note that you only need to replace seacretpassword for SQLALCHEMY_DATABASE_URI. It should be the same as POSTGRES_PASSWORD.

POSTGRES_PASSWORD=secretpassword
PGADMIN_DEFAULT_EMAIL=example@example.com
PGADMIN_DEFAULT_PASSWORD=example
SQLALCHEMY_DATABASE_URI=postgresql://postgres:secretpassword@db/postgres

3. Run docker-compose-win10.yml file

Open CMD or Powershell

cd <path to >/pathwyas-visualization-tool

docker-compose -f docker-compose-win10.yml up --build

You will see something like below after successful run.

Successfully built 04e25c98d625
Successfully tagged pathways-visualization-tool_dash:latest
Creating pathways-visualization-tool_pgadmin_1 ... done
Creating pathways-visualization-tool_adminer_1 ... done
Creating pathways-visualization-tool_db_1      ... done
Creating pathways-visualization-tool_dash_1    ... done

4. Open Dash application

Go to http://localhost:8050/

Little More...

Checking network

docker network ls

You will see something like:

1aa0d4fa2c31        pathways-visualization-tool_default   bridge              local

Then inspect the network:

docker network inspect pathways-visualization-tool_default

You see four containers in "Containers" JSON variable with the name listed below.

pathways-visualization-tool_adminer_1
pathways-visualization-tool_pgadmin_1
pathways-visualization-tool_db_1
pathways-visualization-tool_dash_1

Conecting PostgreSQL database in pgAdmin

Open the browser and go to http://localhost:8000/
It will open pgAdmin4 login page. Enter email address and password you used in .env file.

Now, you will add a new server in pgAdmin which connect to pathways-visualization-tool_db_1 that we just created.

 * Click Add New Server and name the server.  You can name whatever you want here.
 * Click Connection tab
         - Host name/address: pathways-visualization-tool_db_1
         - Port: 5432
         - Maintenance database: postgres
         - Username: postgres
         - Password: <Use POSTGRES_PASSWORD in your .env file>
 * Click Save. You will see a new server just you created in the left pane.

In the left pane, click Database > Schemas > Tables. You should see tables created from database.sql file.

Query without pgAdmin

You will need to get into container

docker exec -it pathways-visualization-tool_db_1 bash

psql -U postgres

To quit PostgreSQL

\q

To quit bash

exit

Run Dash application on your machine (not Docker container)

It seems Windows doesn't know the IvP address of Postgresql DB. There might have other solutions, but I was successful by using steps below.

Add 'ports: 5432:5432' in docker-compose-win10.yml.
Create .env file in the same directory as run.py.
Add a line below and change your password for database and save.

SQLALCHEMY_DATABASE_URI="postgresql://postgres:<your_password>@localhost:5432/postgres"

Run docker-compose:

docker-compose -f docker-compose-win10.yml up --build

Stop container that runs Dash application.

docker stop pathways-visualization-tool_dash_1

Activate your local python virtual environment and run:

python run.py

Go to http://localhost:8050/

Getting data from Postgresql running in Docker container

This is a little code just for you to get started.

import psycopg2

# Connect to the database
conn = psycopg2.connect(host='localhost',
      database='postgres', user='postgres', 
      password='secretpassword')
cursor = conn.cursor()

# Show all tables
cursor.execute('SELECT relname FROM pg_class WHERE relkind='r' AND relname !~ '^(pg_|sql_)';')
print(cursor.fetchall(), '\n\n')

cursor.execute('SELECT * FROM disp_code')
print(cursor.fetchall())

ncsu-landscape-dynamics / pathways-visualization-tool