@alanko0511/open-meteo-js

A type-safe TypeScript/JavaScript library for the Open-Meteo Weather API. This library uses the JSON API format and transforms the data into an easy-to-use structure (for me at least) with full type safety.

If you need maximum performance, you should check out their official library which uses FlatBuffers. This library is slower because it uses JSON over HTTP with Zod validation, but focuses on developer experience and type safety instead.

Features

• Fully Type-Safe: Response types automatically match your requested variables
• Runtime Validation: Zod schemas validate both inputs and outputs
• Transformed Data: Converts Open-Meteo's parallel arrays into arrays of objects
• Timestamps in Milliseconds: Unix timestamps converted to JavaScript milliseconds (ready for new Date())
• Type-Safe Units: Units included for all variables with full type safety

Installation

# Using Bun
bun add @alanko0511/open-meteo-js

# Using npm
npm install @alanko0511/open-meteo-js

# Using pnpm
pnpm add @alanko0511/open-meteo-js

# Using Yarn
yarn add @alanko0511/open-meteo-js

Usage

Basic Example

import { forecastAPI } from "@alanko0511/open-meteo-js";

const forecast = await forecastAPI({
  latitude: 52.52,
  longitude: 13.41,
  hourly: ["temperature_2m", "precipitation", "wind_speed_10m"],
  daily: ["temperature_2m_max", "temperature_2m_min"],
  current: ["temperature_2m", "weather_code"],
});

// TypeScript knows the exact structure
console.log(forecast.hourly[0].temperature_2m); // number
console.log(forecast.hourly[0].precipitation); // number
console.log(forecast.daily[0].temperature_2m_max); // number
console.log(forecast.current.weather_code); // number

// Units are also type-safe
console.log(forecast.hourly_units.temperature_2m); // string (e.g., "°C")
console.log(forecast.daily_units.temperature_2m_max); // string

// These would be TypeScript errors:
// console.log(forecast.hourly[0].cloud_cover); // Error: not requested
// console.log(forecast.hourly_units.cloud_cover); // Error: not requested

Data Format

This library transforms the data from parallel arrays to arrays of objects:

Open-Meteo Format (parallel arrays):

{
  "hourly": {
    "time": [1234567890, 1234571490, ...],
    "temperature_2m": [15.2, 15.5, ...],
    "precipitation": [0, 0.1, ...]
  }
}

This Library (array of objects):

{
  "hourly": [
    {
      "time": 1704067200000, // Unix timestamp in milliseconds
      "temperature_2m": 15.2,
      "precipitation": 0
    },
    {
      "time": 1704070800000, // Unix timestamp in milliseconds
      "temperature_2m": 15.5,
      "precipitation": 0.1
    }
  ],
  "hourly_units": {
    "temperature_2m": "°C",
    "precipitation": "mm"
  }
}

Note: Time values are Unix timestamps in milliseconds (JavaScript standard). You can easily convert them to Date objects:

const date = new Date(forecast.hourly[0].time);

Advanced Options

const forecast = await forecastAPI({
  latitude: 52.52,
  longitude: 13.41,
  hourly: ["temperature_2m"],
  timezone: "Europe/Berlin",
  temperature_unit: "fahrenheit",
  wind_speed_unit: "mph",
  precipitation_unit: "inch",
  forecast_days: 7,
  past_days: 1,
});

Commercial API & Self-Hosted Instances

You can use the commercial Open-Meteo API or self-hosted instances by providing configuration options:

Commercial API with API Key

When you provide an API key, the library automatically uses the commercial API endpoint (https://customer-api.open-meteo.com/v1):

const forecast = await forecastAPI(
  {
    latitude: 52.52,
    longitude: 13.41,
    hourly: ["temperature_2m"],
  },
  {
    apiKey: "your-api-key-here",
  }
);

Self-Hosted Instance

For self-hosted Open-Meteo instances, specify a custom base URL:

const forecast = await forecastAPI(
  {
    latitude: 52.52,
    longitude: 13.41,
    hourly: ["temperature_2m"],
  },
  {
    baseUrl: "https://your-domain.com/v1",
  }
);

Custom Timeout

You can also customize the request timeout (defaults to 30 seconds):

const forecast = await forecastAPI(
  {
    latitude: 52.52,
    longitude: 13.41,
    hourly: ["temperature_2m"],
  },
  {
    timeout: 60000, // 60 seconds
  }
);

Available Variables

Hourly Variables

• temperature_2m - Temperature at 2 meters
• relative_humidity_2m - Relative humidity at 2 meters
• dew_point_2m - Dew point at 2 meters
• apparent_temperature - Apparent temperature
• precipitation_probability - Precipitation probability
• precipitation - Total precipitation (rain + showers + snow)
• rain - Rain
• showers - Showers
• snowfall - Snowfall
• snow_depth - Snow depth
• weather_code - WMO Weather code
• pressure_msl - Sea level pressure
• surface_pressure - Surface pressure
• cloud_cover - Total cloud cover
• cloud_cover_low - Low cloud cover
• cloud_cover_mid - Mid cloud cover
• cloud_cover_high - High cloud cover
• visibility - Visibility
• evapotranspiration - Evapotranspiration
• et0_fao_evapotranspiration - Reference evapotranspiration
• vapour_pressure_deficit - Vapour pressure deficit
• wind_speed_10m - Wind speed at 10m
• wind_speed_80m - Wind speed at 80m
• wind_speed_120m - Wind speed at 120m
• wind_speed_180m - Wind speed at 180m
• wind_direction_10m - Wind direction at 10m
• wind_direction_80m - Wind direction at 80m
• wind_direction_120m - Wind direction at 120m
• wind_direction_180m - Wind direction at 180m
• wind_gusts_10m - Wind gusts at 10m
• temperature_80m - Temperature at 80m
• temperature_120m - Temperature at 120m
• temperature_180m - Temperature at 180m
• soil_temperature_0cm - Soil temperature at 0cm
• soil_temperature_6cm - Soil temperature at 6cm
• soil_temperature_18cm - Soil temperature at 18cm
• soil_temperature_54cm - Soil temperature at 54cm
• soil_moisture_0_1cm - Soil moisture 0-1cm
• soil_moisture_1_3cm - Soil moisture 1-3cm
• soil_moisture_3_9cm - Soil moisture 3-9cm
• soil_moisture_9_27cm - Soil moisture 9-27cm
• soil_moisture_27_81cm - Soil moisture 27-81cm

Daily Variables

• weather_code - WMO Weather code
• temperature_2m_max - Maximum temperature
• temperature_2m_min - Minimum temperature
• apparent_temperature_max - Maximum apparent temperature
• apparent_temperature_min - Minimum apparent temperature
• sunrise - Sunrise time
• sunset - Sunset time
• daylight_duration - Daylight duration
• sunshine_duration - Sunshine duration
• uv_index_max - Maximum UV index
• uv_index_clear_sky_max - Maximum UV index (clear sky)
• rain_sum - Rain sum
• showers_sum - Showers sum
• snowfall_sum - Snowfall sum
• precipitation_sum - Precipitation sum
• precipitation_hours - Precipitation hours
• precipitation_probability_max - Maximum precipitation probability
• wind_speed_10m_max - Maximum wind speed
• wind_gusts_10m_max - Maximum wind gusts
• wind_direction_10m_dominant - Dominant wind direction
• shortwave_radiation_sum - Shortwave radiation sum
• et0_fao_evapotranspiration - Reference evapotranspiration

Current Variables

• temperature_2m - Current temperature
• relative_humidity_2m - Current relative humidity
• apparent_temperature - Current apparent temperature
• is_day - Is it day or night
• precipitation - Current precipitation
• rain - Current rain
• showers - Current showers
• snowfall - Current snowfall
• weather_code - Current weather code
• cloud_cover - Current cloud cover
• pressure_msl - Current sea level pressure
• surface_pressure - Current surface pressure
• wind_speed_10m - Current wind speed
• wind_direction_10m - Current wind direction
• wind_gusts_10m - Current wind gusts

Type Safety in Action

The response type adapts based on what variables you request:

// Request only temperature
const response1 = await forecastAPI({
  latitude: 52.52,
  longitude: 13.41,
  hourly: ["temperature_2m"],
});
// Works:
//   response1.hourly[0].temperature_2m
//   response1.hourly_units.temperature_2m
// Type errors:
//   response1.hourly[0].precipitation
//   response1.hourly_units.precipitation

// Request temperature and precipitation
const response2 = await forecastAPI({
  latitude: 52.52,
  longitude: 13.41,
  hourly: ["temperature_2m", "precipitation"],
});
// Works:
//   response2.hourly[0].temperature_2m
//   response2.hourly[0].precipitation
//   response2.hourly_units.temperature_2m
//   response2.hourly_units.precipitation

// Request only daily variables
const response3 = await forecastAPI({
  latitude: 52.52,
  longitude: 13.41,
  daily: ["temperature_2m_max"],
});
// Works:
//   response3.daily[0].temperature_2m_max
// Type errors:
//   response3.hourly
//   response3.current

Units

Units are automatically included for all requested variables and are also type-safe:

const forecast = await forecastAPI({
  latitude: 52.52,
  longitude: 13.41,
  hourly: ["temperature_2m", "wind_speed_10m"],
  temperature_unit: "fahrenheit",
  wind_speed_unit: "mph",
});

console.log(forecast.hourly_units.temperature_2m); // "°F"
console.log(forecast.hourly_units.wind_speed_10m); // "mph"
// forecast.hourly_units.precipitation // Type error - not requested

Working with Timestamps

All time values are returned as Unix timestamps in milliseconds (JavaScript standard):

const forecast = await forecastAPI({
  latitude: 52.52,
  longitude: 13.41,
  hourly: ["temperature_2m"],
});

// Time is a number (milliseconds since Unix epoch)
const timestamp = forecast.hourly[0].time; // e.g., 1704067200000

// Convert to Date when needed
const date = new Date(timestamp);
console.log(date.toISOString()); // "2024-01-01T00:00:00.000Z"

// Or use directly with Date methods
const hours = new Date(forecast.hourly[0].time).getHours();

Error Handling

The library validates both input parameters and API responses using Zod:

try {
  const forecast = await forecastAPI({
    latitude: 52.52,
    longitude: 13.41,
    hourly: ["temperature_2m"],
  });
} catch (error) {
  if (error instanceof ZodError) {
    // Validation error (invalid params or unexpected API response)
    console.error("Validation error:", error.issues);
  } else {
    // Network or other errors
    console.error("Request failed:", error);
  }
}