multiweatherapi

Multi Weather API

Consistent Python bindings for select commercial weather station APIs, such as Zentra Cloud REST API v3, created for MSU EnviroWeather Project.

Background

This is inspired by python bindings for v1 of the Zentra API from the Montana Climate office : MSU Agricultural Weather Office

The EnviroWeather system from the MSU Agricultural Weather Office will interface with weather stations (select list of vendors) deployed and owned privately. Since each vendor has a differrent REST/Web API to access station's weather data, this package aims to provide a consistent interface to all of those stations in order to make it easy for incoporating them into a data pipeline.

Installation

The Multi-Weather API is currently available on PyPI. Install using pip:

pip install multiweatherapi

Usage

>>> from multiweatherapi import multiweatherapi
>>> param = {'sn': 'STATION_ID',
...          'apikey': 'API_KEY',
...          'apisec': 'API_SECRET'}
>>> resp = multiweatherapi.get_reading('davis', **param)
>>> resp.resp_raw # raw JSON response of the reading
>>> resp.resp_parsed # parsed JSON response into list of dict for EnviroWeather project

Refer to the respective link below for parameter and authentication requirements of weather station vendor APIs

• Zentra
• Spectrum
• Onset
• Davis
• Rainwise
• Campbell

Supported Python Versions

Python 3.6 and higher are supported.

Supported Station Time Zones

Currently following time zones are supported:

Name	Description
HT	US/Hawaii
AT	US/Alaska
PT	US/Pacific
MT	US/Mountain
CT	US/Central
ET	US/Eastern

Requirements

• Requests
• pytz
• pytest
• python-dotenv

Outputs

• API Response (JSON)

  The (raw) API response is formatted in JSON and 0th element illustrates the metadata of the API call and 1st (and onward - Davis stations) element(s) depicts weather station measurements retrieved from the API call.

  [
    {
      "vendor": "spectrum",
      "station_id": "50400123",
      "start_datetime": "2022-05-26 12:08:07",
      "end_datetime": "2022-05-26 13:08:07",
      "timezone": "ET",
      "request_time": "2022-05-27 12:08:07",
      "python_binding_version": "0.0.17"
    },
    {
      "EquipmentRecords": [
        {
          "SerialNumber": "50400123",
          "TimeStamp": "2022-05-26T16:10:00",
          "SensorData": [
            {
              "ChannelNumber": 0,
              "PortNumber": "A",
              "SensorNumber": 0,
              "SensorType": "Rainfall",
              "TimeStamp": "2022-05-26T16:10:00",
              "FormattedTimeStamp": "2022-05-26 16:10",
              "Value": "0.00",
              "AccumlatedValue": "0.00",
              "DecimalValue": 0.0,
              "AccumlatedDecimalValue": 0.0,
              "Units": "inches",
              "ValueType": "Average"
            },
            ...
    }
  ]
  

• Transformed Response (List of Dict)

  The transformed response is formatted in list (array) of Python dict object. Each dict object consists of that measurements that stakeholders are interested in and the metadata that would be used to facilitate data loading into the backend.

  [
    {
      "station_id": "50400123",
      "request_datetime": "2022-05-27 12:08:07",
      "pcpn": 0.0,
      "data_datetime": "2022-05-26 16:10",
      "atemp": 21.0,
      "relh": "92.3"
    },
    ...
  ]
  

License

Released under the MIT License

Testing the package

If you want to test without installing via PIP, there is a script get_sample_data.py in the main directory which requires a configuration file named '.env' in the root folder.

$ pip install -r requirements.txt
$ python get_sample_data.py rainwise /Users/jhp/Desktop/

Please refer to docs/get_sample_data.md for detail documentation of get_sample_data.py

see also docs/Test Suite Doco.py for more formal tests and how to provide station configuration file