wetter

This repository provides an Python application for checking the current weather conditions at your favourite location. Additionally, it allows querying historical data to run analysis on past measurements.

Usage

The tool provides the wetter command line tool with several subcommands:

Entering nothing but the wetter command will return the latest measurement of the location similar to wetter latest.

Getting started

There are several ways to install wetter. The easiest method would be to use pypi, download releases from GitHub or clone the repository and use e.g. poetry to install it. If you don't want to install library immediately you could first try wetter by spinning up its docker image.

• Install using pypi

  pip install wetter  # the more courageous can add a --pre flag
  

• Install by (1) downloading the .whl from our releases and (2) install using pip

  python3 -m pip install wetter-0.3.0-py3-none-any.whl
  

• Install from source by cloning the repository and using poetry

  git clone https://github.com/ucyo/wetter-py.git  # clone the repo
  cd wetter-py/wetter  # change to library
  poetry install
  

• Get started using docker

  docker run -it ucyo/wetter:latest bash  # ucyo/wetter:testing for pre-releases
  

Checking if everything is working appropriately can be done using wetter latest. It should return something like the following:

> wetter latest
Currently it is 🌡️ 12.7°C and wind speed 🌬️ 13.0 km/h.
Latest measurement on 📅 2023-01-01 @ 01:00AM.

Now that we know everything is working as expected. Go ahead and update the database by executing wetter update.

Your mileage might vary on getting the exact same output. The timestamp of the above command adjusts to the local time zone and might be different on yours.

Configuration

I don't know why, but you might be interested in measurements from a different location. You can do this by adjusting the wetter.toml file. The location of the wetter.toml depends on your operating system. Additionally, you can find the locations of the measurement data itself i.e. wetter.json and logs i.e. wetter.log.

If you have problems finding the proper location there is a gimmick that got you covered. The path to the configuration file is returned by wetter configure --config. The logging level can be set by the WETTER_LOG environmental variable.

Sample configuration

[location]
lat = 49
lon = 8.41

The configuration file is very basic. Simply type in the lat/lon position :earth_africa: of your favourite location. You might use an online service to look up the coordinates of a certain city like a LatLongFinder.

Sample database

{
    "lat": 49,
    "lon": 8.41,
    "version": 1,
    "data": {
        "index": ["temperature", "wind"],
        "columns": ["2023-01-01T00:00:00.000000+0000", "2023-01-01T01:00:00.000000+0000"],
        "data": [[7.7, 12.7], [6.8, 13.0]],
    },
}

The json database stores only the data for a single position. Those are included in the database with the lat and lon tags. This information aligned on system start with the configuration file. Should they not match within 1 degree (in the above case for latitude 48 <= lat <= 50), then a reset will be triggered.

Note: The update to the new location happens without user interaction. The tool assumes the user changed the settings knowingly and will update the the database on the next execution.

Setup background daemon

There are several ways to enable a background process on Unix systems (incl. macOS). The easiest and most supported is to setup using cron. Another scheduling daemon is systemd. In the following are the instructions for both systems.

Spoiler alert! Use cron.

Crontab

Execute the following command:

(crontab -l ; echo "5 * * * * wetter update") 2> /dev/null | sort -u | crontab -

Check crontab guru for details on the scheduling syntax.

Systemd

Go back :point_up: Just use cron. All that glistens is not gold :eyes:

tl;dr

In user space:

wetter configure --systemd > wetter.service  # generate service file
wetter configure --systemdtimer > wetter.timer # generate schedule file

As privileged user:

ln -s $(which wetter) /usr/bin/wetter  # get binary out of home
mv wetter.service /etc/systemd/system/wetter.service  # move service file
mv wetter.timer /etc/systemd/system/wetter.timer  # move schedule service
systemctl enable wetter.service  # enable service
systemctl start wetter.service  # start service

Detailed

First you need to create the necessary files for a systemd service. This involves two files: First, the service file which defines the background process in wetter.service. Afterwards, the scheduler file which defines when the background process needs to be run i.e. wetter.timer.

wetter configure --systemd > wetter.service  # generate service file
wetter configure --systemdtimer > wetter.timer # generate schedule file

Note: Timers are not mandatory to run services on a certain schedule. One could use the the systemd-run command to schedule calls to services without a timer configuration. See systemd-run manpage on ArchLinux for details.

Now that have the necessary files set up, need to make them available for systemd. The most common location of systemd files is /etc/systemd/system/. Therefore you need to move the files to that location.

mv wetter.service /etc/systemd/system/wetter.service  # move service file
mv wetter.timer /etc/systemd/system/wetter.timer  # schedule service

Note: Fedora/RedHat/CentOS users might need to adjust the above mentioned location.

Before you can enable and start the service you must address a caveat of systemd. The systemd daemon is not allowed to access binaries in home directories of users. That's why you need to create a symbolic link in /usr/bin to allow the service to execute updates on the database. That can be done using the following command:

ln -s $(which wetter) /usr/bin/wetter

This concludes the setup process. Now the service can be enabled and started using the systemctl command.

• Enabling the systemd service systemctl enable wetter.service
• Starting the systemd service systemctl start wetter.service

Resources

The measurement data is being gathered using the archive and forecast API from Open-Meteo.com :heart: