FAQ

A page for frequently asked questions.

Q: Why do I need special libraries to run PyWeather?

A: Those special libraries make PyWeather work. PyWeather could in theory run in built-in Python stuff, but libraries make my job easier.

Q: Why won't you make PyWeather for Python 2?

A: I didn't want to code for Python 2, as it's now out of date, and will reach end of life eventually.

Q: Why won't PyWeather work on my platform?

A: PyWeather can't run on every platform, because unexpected stuff happens. On some platforms, some stuff works great, but on others, some stuff doesn't. On most normal platforms, PyWeather will run flawlessly. On other specialized, or unusual platforms, PyWeather might not run at all, or somewhat work.

Q: Will you release a fix for X platform?

A: No, unless the issue you have exists in most common OSes. I don't want to end up with releases looking like this:

pyweather-0.7beta
pyweather-0.7beta-termux
pyweather-0.7beta-pitopos
pyweather-0.7beta-freenas9
pyweather-0.7beta-incompatibleos1
pyweather-0.7beta-incompatibleos2

Having OS-specific releases will increase the time between releases, as I have to "port" over code to the OS-specific code.

Q: Why do I need to get an API key from Wunderground?

A: Q: Why are you so lazy?

Q: When will the next version of PyWeather release?

A: You can get a "rolling" idea of when a PyWeather release will be by looking at issues and seeing what's getting done and what's not.

Nowadays, I'm adding more features to a release than older releases, I'm doing more extensive QA testing for each release, and I don't work on PyWeather as much as I use to. As such, the time between releases is now increased.

Q: How can I contribute to PyWeather?

A: It's easy to help me with PyWeather. Even if you can't code Python, you can report bugs which tremendously help PyWeather. If you can code PyWeather, you can code in new features, or make minor changes that you feel is necessary.

More info is in the CONTRIBUTING.md file in the .github folder.

Q: Should I use PyWeather during a severe storm?

A: No, but let me explain.

Using PyWeather during normal weather is completely safe, so if you're checking the weather a few days out, it's fine.

During severe thunderstorms, I'd consider PyWeather safe. However, if you're using PyWeather during a local tornado outbreak, local tsunamis, and a tropical system hitting your location, it's best advised to not use PyWeather.

Even though PyWeather has an alert and radar system, I can't guarantee that you'll get this data in a timely fashion compared to emergency alerts pushed to your phone, or listening to local authorities that have the latest radar information.

Moreover, PyWeather only has alerts data for the following countries:

North America

• The lower 48 US states, Alaska & neighboring islands, Hawaii, and Puerto Rico & neighboring US-claimed islands
• Canada

Europe

• Austria
• Croatia
• Denmark
• Estonia
• Finland
• Germany
• Iceland
• Ireland
• Italy
• Latvia
• Malta
• Montenegro
• Netherlands
• Norway
• Poland
• Portugal
• Serbia
• Sweden

In addition, PyWeather only has radar coverage for NWS/FAA WSR-88D sites. The NEXRAD WSR-88D system covers these areas with NWS/FAA-owned sites:

• ~80% of the mainland US (east coast/mid US has good coverage, west coast has subpar coverage)
• The southern border of Canada
• The northern border of Mexico
• ~40% of Alaska (southern alaska good coverage, northern alaska poor coverage)
• Puerto Rico
• Hawaii

Basically this is to say that it's not a good idea to use PyWeather during severe storms that could threaten your life.

Q: How did PyWeather start? Is it going to be another one of those odd starting stories?

A: Yes! PyWeather was created in one of the most unusual ways possible. On January 9, it all begun, and I made a small script that showed the weather to be a show-off to people coming from a company to a small class in my school.

PyWeather was launched that evening at 9:59pm EST, and 30-line version that only outputted the current conditions was shortly published to GitHub an hour later.

I can hardly remember the start of PyWeather, but this was at a time where PyTerm was still the project I worked on the most, and I was still heavily working on it through January. After ditching the darksky.io library, I began to work with raw .json files, and the first edition of PyWeather that utilized this was published on January 22.

PyWeather then again stalled for another few weeks, up to early February. However, it was around this time that I made the call to prioritize PyWeather over PyTerm, and PyTerm was effectively dead.

I began to ramp up the development process of PyWeather starting on February 11. On February 22, PyWeather 0.2 beta was released, and I changed the readme of PyTerm's indev branch to say that I was prioritizing PyWeather over PyTerm. Official EoL was March 4.

And that's how PyWeather got started. A project initially used to be a show-off hacked together in a night that I stuck with.

Q: When will PyWeather be done?

A: See the timeline page.

Q: Will PyWeather support Python 4 when it comes out?

A: It honestly depends. It's fun to look at Python 2 and 3, and how people who learned to use Python 2 still do, and people who learned to use Python 3 use Python 3. Since I'll want to keep working with Python for a long time, PyWeather will most likely get ported to Python 4.