Useful linksUseful links

Quick startQuick start

To make your first request to Meteum API, you need to get an access key.
You can get a trial key on the developer portal.

To authorize access, pass the key in the X-Meteum-API-Key header of the HTTP request.

{
  weatherByPoint(request: { lat: 52.37125, lon: 4.89388 }) {
    now {
      temperature
    }
  }
}

https://api.meteum.ai/v1/fact?lat=52.37125&lon=4.89388

Example in PythonExample in Python

Here's a minimal working example in Python. To run it, just paste your API access key into the value of the access_key variable and execute this code using an interpreter.

import requests

# insert your real key here!
access_key = "your_key"

headers = {
    "X-Meteum-API-Key": access_key
}

query = """{
weatherByPoint(request: { lat: 52.37125, lon: 4.89388 }) {
    now {
    temperature
    }
}
}"""

response = requests.post('https://api.meteum.ai/graphql/query', headers=headers, json={'query': query})

print(response.content)

import requests

# insert your real key here!
access_key = "your_key"

headers = {
    "X-Meteum-API-Key": access_key
}

response = requests.get('https://api.meteum.ai/v1/forecast?lat=52.37125&lon=4.89388', headers=headers)

print(response.content)

How to create a request to GraphQL Meteum APIHow to create a request to GraphQL Meteum API

The GraphQL query language is used to create requests to Meteum API. GraphQL is used to capture a schema describing the data in the API in a way that is understandable to users. This schema unambiguously defines how a query to the API should be generated and what the response format will be. The main advantage of the GraphQL schema (compared to REST API, for example) is the ability to get any subset of data in a single request. Thanks to this, you can write much shorter and clearer code since you don't have to build the logic for merging response data when requesting different data types. Moreover, most weather API providers charge for a particular number of API requests per unit of time. This means that getting all the data you need in one request (instead of multiple requests) is more cost-effective.

A request is simple text that's similar to a JSON object. Each request is enclosed in curly braces

{
  # request parameters
}

Earlier, we saw an example of a request for "current temperature at a location". The main point of this request is to call the weatherByPoint method with the request parameter, for which a specific value is set

{
  lat: 52.37125,
  lon: 4.89388
}

By doing so, you're basically telling the system "give me weather info at this location" and specifying which point we want to get data for. Next, you need to specify exactly what data you need. To do this, add the necessary structures to the request:

{
  weatherByPoint(request: { lat: 52.37125, lon: 4.89388 }) {
    now {          # weather for now
      temperature  # temperature value
    }
  }
}

By executing this request, here's what we'll get in the response:

{
  "data": {
    "weatherByPoint": {
      "now": {
        "temperature": 20
      }
    }
  }
}

How to use the playground and learn more about the GraphQL schemaHow to use the playground and learn more about the GraphQL schema

Meteum API is equipped with an interactive playground interface where you can try to create and execute requests. The playground lets you create requests using autocomplete features and interactive documentation, which describes the GraphQL API schema in detail. Once the request is created, you can immediately execute it and make sure it's correct. You can also generate working code for Python and Node.JS in the playground and execute it via the command line using the curl utility.