SwiftMETAR is a Swift library that parses aviation routine weather reports (METARs) and terminal aerodrome forecasts (TAF) into Swift data structures useful for machine interpretation. These products contain weather information of interest to pilots. METARs are textual hourly observations of current conditions within 5 miles of an airport, and TAFs are textual point forecasts of expected conditions within 5 miles of an airport, over a 24- to 28-hour period.
SwiftMETAR does not download METAR or TAF products from the Internet -- you'll have to
do that yourself. Once you have the text product, SwiftMETAR will parse it into a METAR or
TAF object that you can use to programmatically work with the weather information.
The design goal of SwiftMETAR is domain-restricted data as much as possible. Wherever possible, SwiftMETAR avoids representing data as open-ended types such as strings. Instead, enums and other types with small domains are preferred. This obviously has versatility implications -- namely, SwiftMETAR is more likely to generate parsing errors for malformed data, especially data that is keyed in by humans -- but it also results in a library that's more in harmony with the design goals of the Swift language itself (type safety, compile-time checks, etc.).
SwiftMETAR bases is parsing on Advisory Circular 00-45F, which is an American FAA publication. Some affordance has been made beyond AC00-45F to parse non-American variances as well (such as supporting QNH-values, visibility in meters, etc.), but this library has not been extensively tested with non-US products. Similarly, some affordance has been made for parsing TAFs at military airbases, but most of those will fail to parse for now.
Online API and tutorial documentation is available at https://riscfuture.github.io/SwiftMETAR/documentation/swiftmetar/
SwiftMETAR is a Swift Package Manager project. To use SwiftMETAR, simply add this
project to your Package.swift file. Example:
// [...]
dependencies: [
.package(url: "https://github.com/RISCfuture/SwiftMETAR/", .branch("main")),
]
// [...]To parse a METAR in String format, simply call METAR.from. You will get back a struct
that you can query for weather information:
let observation = try await METAR.from(string: myString)
if let winds = observation.winds {
switch winds {
case let .direction(heading, speed, gust):
switch speed {
case let .knots(value):
print("Winds are \(heading) at \(speed) knots")
// [...]
}
// [...]
}
}As you can see, many of the fields in a METAR (or TAF) struct are enums (or even nested
enums as in the case of Wind). This is more cumbersome than working with strings but
results in stronger guarantees about the format and integrity of the data.
For more information on how to use the METAR and TAF struct, see the documentation
comments.
SwiftMETAR prefers DateComponents rather than Date objects generally, to preserve the
original data (day-hour-minute) rather than generating timestamps. Both METAR and TAF
have vars allowing you to retrieve these values as Dates, but the data is stored as
components.
Online API documentation and tutorials are available at https://riscfuture.github.io/SwiftMETAR/documentation/swiftmetar/
DocC documentation is available, including tutorials and API documentation. For Xcode documentation, you can run
swift package generate-documentation --target SwiftMETARto generate a docarchive at
.build/plugins/Swift-DocC/outputs/SwiftMETAR.doccarchive. You can open this
docarchive file in Xcode for browseable API documentation. Or, within Xcode,
open the SwiftMETAR package in Xcode and choose Build Documentation from the
Product menu.
Unit testing is done using Nimble and Quick. Simply test the SwiftMETAR target to run
tests.
A SwiftMETAR_Gauntlet target is also available to do an end-to-end test with live data.
This will download METARs and TAFs from the AWC server and attempt to parse them. Any
failures will be logged with the failing string.