meteomatics

3.0.0 verified Mon Apr 27 auth: no python

Python client for the Meteomatics Weather API. Provides access to historical, current, and forecast weather data. Latest version: 3.0.0, actively maintained.

pip install meteomatics

Common errors

error ImportError: cannot import name 'MeteomaticsConnector' from 'meteomatics' ↓

cause v3.0.0 removed the MeteomaticsConnector class.

fix

Use module-level functions like from meteomatics import query_time_series.

error TypeError: query_time_series() got an unexpected keyword argument 'on_invalid' ↓

cause `on_invalid` parameter is only available for grid queries, not time series.

fix

Remove on_invalid argument from time series calls, or use it with query_grid_unpivoted.

error ValueError: startdate and enddate must be timezone-aware ↓

cause SQL-like parameters require timezone-aware datetime objects.

fix

Convert naive datetimes using pytz.UTC.localize() or datetime.replace(tzinfo=zoneinfo.ZoneInfo('UTC')).

Warnings

breaking v3.0.0 removed deprecated functions like `query_dem` and `query_grid` without the `_unpivoted` suffix. ↓

fix Update calls to current API functions: e.g., use `query_grid_unpivoted` instead of `query_grid`.

gotcha The module-level functions (e.g., query_time_series) require username and password passed directly; the old MeteomaticsConnector class is gone. ↓

fix Pass credentials to each function call, not through a connector object.

gotcha Time parameters must be timezone-aware for SQL-like queries; naive datetimes may raise errors. ↓

fix Use pytz or zoneinfo to make datetimes aware before passing.

Imports

MeteomaticsConnector
wrong
```
from meteomatics import MeteomaticsConnector
```
correct
```
import meteomatics
```
In v3.0.0 the main class has been removed; use the module-level functions.
query_time_series
```
from meteomatics import query_time_series
```
Commonly used function, ensure correct import path.

Quickstart

Query hourly temperature for a single point over the last 24 hours.

import datetime as dt
from meteomatics import query_time_series

username = os.environ.get('METEOMATICS_USER', '')
password = os.environ.get('METEOMATICS_PASSWORD', '')

now = dt.datetime.utcnow()
startdate = now - dt.timedelta(days=1)
enddate = now
interval = dt.timedelta(hours=1)
parameters = ['t_2m:C']
locations = [(50.0, 10.0)]

df = query_time_series(startdate, enddate, interval, parameters, locations, username, password)
print(df.head())