This respository contains the files necessary to run a webserver that runs with a Enviro Weather. The webserver is designed to run as a service on a Raspberry Pi and provides the HTTP Endpoints to accept the readings from the Enviro Weather and store them in a MySQL database. It also serves responsive webpages to view the recorded data as shown in the following screenshots.
As well as the most recently recorded data the webpages plot historic weather data for both the past 24 hours rainfall and also the last 7 days of temperature, humidity, pressure and daily rainfall.
A link is provided in the top left to an interactive dashboard (runs on port:5001) which enables the data to be analysed for any given time period. A basic summary is provided at the top of the screen and more detailed analysis is possible at the bottom including the ability to select a number of different data fields for more detailed analysis.
In the top right a link is provided to view the data stored in the database. Each page contains a day of data, 96 readings i.e. every 15 minutes.
The webserver is designed to run on a Raspberry Pi and has been developed using Python 3.11.3. This is a home project for fun, it works reliably but may need some modification to work on your system. There are probably many ways to improve it - grateful for any suggestions!
Setup the Enviro Weather as described here adding the IP address of the Raspberry Pi as the custom HTTP endpoint. Taking readings every 15 minutes is recommended.
These are described in the requirements.txt file. Simply run
git clone https://github.com/sdmeers/weatherstation
cd weatherstation
pip install -r requirements.txtThen copy index.php into /var/www/html/.
The weather records are stored in a MySQL database which is recommended to run on a remote Raspberry Pi or similar. Follow the installation instructions here.
Login to MySQL as root user and create a new database called weather and switch to use that database.
sudo mysql -u root -p
CREATE DATABASE weather;
USE weather;
Finally create a new table called data within the database using the following command. This table will store the data sent from the weatherstation.
CREATE TABLE IF NOT EXISTS `data` (
id INT AUTO_INCREMENT PRIMARY KEY,
timestamp DATETIME NOT NULL,
temperature DOUBLE(4,2) NOT NULL,
pressure DOUBLE(6,2) NOT NULL,
humidity DOUBLE(3,1) NOT NULL,
rain DOUBLE(6,2) NOT NULL,
rain_rate DOUBLE(6,5) NOT NULL,
luminance DOUBLE(7,2) NOT NULL,
wind_speed DOUBLE(3,1) NOT NULL,
wind_direction DOUBLE(3,0) NOT NULL,
day int(3) NOT NULL,
week int(2) NOT NULL,
month int(2) NOT NULL,
year int(4) NOT NULL
);
Depending on your user permissions it may be necessary to update privileges using
CREATE USER 'myuser'@'localhost' IDENTIFIED BY 'password';
GRANT ALL PRIVILEGES ON weather.* TO 'myuser'@'localhost';
FLUSH PRIVILEGES;
EXIT;
Add your MySQL <username> and <password> to the two config files using the templates above sql_config-template.py and config-template.php. Rename both files remove the -template suffix and add the <IP_address> of the web server to sql_config.py.
There are three linked services to run the app as follows:
- Server: accepts the post requests from the enviro and writes them into the MySQL database.
- Client: Flask app that serves the summary webpage.
- Dashboard: Plotly-Dash app that runs the interactive dashboard.
The 3 x .service files (<weather_client/server/dashboard.service>) will need to be modified. Follow the process below, repeating for the weather_server and weather_dashboard services.
- Edit weather_client.service with your path to
weather_client.py. The line you will need to edit is as follows.
ExecStart=/usr/bin/python3 <path_to_python_file.py>
e.g.
ExecStart=/usr/bin/python3 /home/pi/weatherstation/weather_client.py
- Save this file to
/etc/systemd/system/weather_client.service - Reload the systemd daemon to recognize your new service
sudo systemctl daemon-reload
- Enable the service to start on boot
sudo systemctl enable weather_client.service
- Start the service immediately without rebooting
sudo systemctl start weather_client.service
This will run the webserver as a background service on boot enabling it to accept data from the Enviro Weather.
The webpages to display the weather data can be accessed via localhost:5001 or <IP_address>:5001. You should something similar to the screenshot below.
Check the status of your service:
sudo systemctl status weather_client.service
Stop, restart & disable your service:
sudo systemctl stop weather_client.service
sudo systemctl restart weather_client.service
sudo systemctl disable weather_client.service
View the most recent 50 log entries via syslog
journalctl -n 50 -u weather_client.service
MySQL command to exporting SQL Data to csv
sudo mysql -u root -p
USE weather;
SELECT *
FROM data
INTO OUTFILE '/tmp/weather_data_export.csv'
FIELDS TERMINATED BY ','
ENCLOSED BY '"'
LINES TERMINATED BY '\n';
A script to convert the weather direction readings from the Enviro Weather (in degrees) to compass cardinal points (N, NE etc...) is included in the weather_helper.py. Use a compass to determine how to modify the values in the helper file depending on how your weatherstation is oriented.
A link is provided to index.php which displays the raw data from the MySQL database. Each page shows 96 records which corresponds to 24 hours of data sampled every 15 minutes.
The Enviro Weather requires a static IP for the web-server (this also makes it easier to access the web-pages). If running on a local network use IP Binding on the router to ensure the web-server is allocated a static IP.
An update_pi script is included to make it easy to copy files from a development laptop across to the remote server (e.g. the Raspberry Pi) using scp. Simply run ./update_pi. You'll need to edit the file to add the IP address of the remote server (also rename to remove the '-template' suffix). You'll also need to have set up ssh to enable this script.
weather_helper.py contains a number of helper Python functions. In particular the get_data(**args) function is included to simplify the extraction of weather records from the MySQL database as a Pandas DataFrame, for example for analysis of the data in a Jupyter Notebook. A summary of the help page for the function is as follows:
get_data(*args):
Fetches data from a database based on the provided time range criteria.
Parameters:
*args : variable length argument list.
- Can be a single string that defines the time range:
- "latest": Returns the most recent record.
- "today": Returns data for the current day.
- "last24h": Returns data for the last 24 hours.
- "yesterday": Returns data for the previous day.
- "day=n": Returns data for the nth day of the current year (n: 1-366).
- "week": Returns data for the current week starting from Monday.
- "week=n": Returns data for the nth week of the year (n: 1-53).
- "last7days": Returns data for the last 7 days.
- "month": Returns data for the current month.
- "month=n": Returns data for the nth month (n: 1-12).
- "year": Returns data for the entire current year.
- "year=n": Returns data for the specified year.
- "all": Returns all the data in the database.
- Or two datetime objects defining the start and end dates for the data fetch.
Returns:
pd.DataFrame: A DataFrame containing data fetched from the database.