Pydarksky!
Pydarksky is a work in progress wrapper for the Dark Sky API.
This is the first python library I have written so any advice and help will be greatly appreciated.
TODO:
- Documentation
- Docstrings
- Code comments
- Readme
- Rewrite unittests
- Example uses
Install
Python versions 2.7, 3.3+
pip install pydarksky
Basic usage
Example uses pendulum for timestamp conversion.
import pendulum
import pydarksky
pendulum.set_formatter("alternative")
weather = pydarksky.weather(api_key=API_KEY, latitude=-34.9285, longitude=138.6005)
# Current weather
if weather.has_now():
print("Now:")
date = pendulum.from_timestamp(weather.now.time, tz=weather.timezone)
try:
temperature = weather.now.temperature
except pydarksky.NoDataError:
temperature = "No Data"
print("Time: {}, Temp: {}\n".format(date.format("DD-MM-YY hh:mm"), temperature))
# Iterating over forecast
if weather.has_daily():
print("Daily:")
for day in weather.daily:
date = pendulum.from_timestamp(day.time, tz=weather.timezone)
try:
temperature = day.temperatureHigh
except pydarksky.NoDataError:
temperature = "No Data"
print("Time: {}, Temp: {}".format(date.format("DD-MM-YY"), temperature))Weather request
# DarkSky instantiation
>>> darksky = pydarksky.DarkSky(API_KEY)
# Pre-define values
>>> darksky.latitude = -34.9285
>>> darksky.longitude = 138.6005
>>> weather = darksky.weather()
# Pass values as params
>>> weather = darksky.weather(latitude=-34.9285, longitude=138.6005)
# Pass values from dict
>>> kwargs = {"longitude": 138.6005, "latitude": -34.9285}
>>> weather = darksky.weather(**kwargs)Historical Request
Unlike the longitude/latitude the date must be passed every time a historical request is wanted. Darksky.net will use the locations timezone by default so specifying it yourself is not required.
>>> import pydarksky
>>> from datetime import datetime
>>> darksky = pydarksky.DarkSky(API_KEY)
>>> darksky.weather(latitude=-34.9285, longitude=138.6005, date=datetime(2017, 1, 1))Setting data blocks to exclude from response
You can opt to exclude specific data blocks from the API response by setting the DarkSky.exclude attribute to one or more of the valid values found in DarkSky.EXCLUDES. Excluding data blocks can be used to reduce server response latency.
DarkSky.exclude_invert() can be used to invert the values in DarkSky.exclude. This is useful if you know which data blocks you need but not the ones you don't.
>>> import pydarksky
>>> darksky = pydarksky.DarkSky(API_KEY)
>>> darksky.EXCLUDES
('currently', 'minutely', 'hourly', 'daily', 'alerts', 'flags')
>>> darksky.exclude = ["alerts", "flags"]
>>> darksky.exclude
['alerts', 'flags']
>>> darksky.exclude = "alerts"
>>> darksky.exclude
['alerts']
>>> darksky.exclude = ["abc"]
ValueError: 'abc' is not a valid exclude value
>>> darksky.exclude_invert()
>>> darksky.exclude
['currently', 'minutely', 'hourly', 'daily']Setting number of hourly data blocks received
You can increase the hour-by-hour data returned from 48, to 168 by setting the DarkSky.extend attribute to True. The default value is False.
Setting response language
>>> import pydarksky
>>> darksky = pydarksky.DarkSky(API_KEY)
>>> darksky.LANGS
['Arabic', 'Azerbaijani', 'Belarusian', 'Bosnian', 'Bulgarian', ...]
>>> darksky.lang
'auto'
>>> darksky.lang = 'English'
>>> darksky.lang
'en'
>>> darksky.lang = 'it' # 'it' == 'Italian'
>>> darksky.lang
'it'
>>> darksky.lang = "abc"
ValueError: 'abc' is not a valid response languageSetting response units
Valid units values:
- auto: automatically select units based on geographic location
- ca: same as si, except that windSpeed is in kilometers per hour
- uk2: same as si, except that nearestStormDistance and visibility are in miles and windSpeed is in miles per hour
- us: Imperial units (the default)
- si: SI units
>>> import pydarksky
>>> darksky = pydarksky.DarkSky(API_KEY)
>>> darksky.UNITS
('auto', 'ca', 'uk2', 'ui', 'si')
>>> darksky.units
'auto'
>>> darksky.units = "si"
>>> darksky.units
'si'
>>> darksky.units = "abc"
ValueError: 'abc' is not a valid unit type
