influxdb-client-swift

This repository contains the reference Swift client for the InfluxDB 2.x.

• Features
• Supported Platforms
• Installation
  • Swift Package Manager
• Usage
  • Creating a client
  • Writing data
  • Querying data
  • Delete data
  • Management API
• Advanced Usage
  • Default Tags
  • Proxy and redirects
• Contributing
• License

Documentation

This section contains links to the client library documentation.

• Product documentation, Getting Started
• Examples
• API Reference
• Changelog

Features

InfluxDB 2.x client consists of two packages

• InfluxDBSwift
  • Querying data using the Flux language
  • Writing data
    • batched in chunks on background
    • automatic retries on write failures
• InfluxDBSwiftApis
  • provides all other InfluxDB 2.x APIs for managing
    • health check
    • sources, buckets
    • tasks
    • authorizations
    • …
  • built on top of InfluxDBSwift

Supported Platforms

This package requires Swift 5 and Xcode 12+.

• iOS 14.0+
• macOS 11.0+
• tvOS 14.0+
• watchOS 7.0+
• Linux

Installation

Swift Package Manager

Add this line to your Package.swift :

// swift-tools-version:5.3
import PackageDescription
let package = Package(
    name: "MyPackage",
    dependencies: [
        .package(name: "influxdb-client-swift", url: "https://github.com/influxdata/influxdb-client-swift", from: "1.6.0"),
    ],
    targets: [
        .target(name: "MyModule", dependencies: [
          .product(name: "InfluxDBSwift", package: "influxdb-client-swift"),
          // or InfluxDBSwiftApis for management API
          .product(name: "InfluxDBSwiftApis", package: "influxdb-client-swift")
        ])
    ]
)

Usage

Important: You should call close() at the end of your application to release allocated resources.

Creating a client

Specify url and token via parameters:

let client = InfluxDBClient(url: "http://localhost:8086", token: "my-token")
...
client.close()

Client Options

Configure default Bucket, Organization and Precision

let options: InfluxDBClient.InfluxDBOptions = InfluxDBClient.InfluxDBOptions(
        bucket: "my-bucket",
        org: "my-org",
        precision: .ns)
let client = InfluxDBClient(url: "http://localhost:8086", token: "my-token", options: options)
...
client.close()

InfluxDB 1.8 API compatibility

client = InfluxDBClient(
        url: "http://localhost:8086", 
        username: "user", 
        password: "pass",
        database: "my-db", 
        retentionPolicy: "autogen")
...
client.close()

Writes

The WriteApi supports asynchronous writes into InfluxDB 2.x. The results of writes could be handled by (response, error), Swift.Result or Combine.

The data could be written as:

1. String that is formatted as a InfluxDB’s Line Protocol
2. Data Point structure
3. Tuple style mapping with keys: measurement, tags, fields and time
4. Array of above items

The following example demonstrates how to write data with Data Point structure. For further information see docs and examples.

import ArgumentParser
import Foundation
import InfluxDBSwift
import InfluxDBSwiftApis
@main
struct WriteData: AsyncParsableCommand {
  @Option(name: .shortAndLong, help: "The name or id of the bucket destination.")
  private var bucket: String
  @Option(name: .shortAndLong, help: "The name or id of the organization destination.")
  private var org: String
  @Option(name: .shortAndLong, help: "Authentication token.")
  private var token: String
  @Option(name: .shortAndLong, help: "HTTP address of InfluxDB.")
  private var url: String
}
extension WriteData {
  mutating func run() async throws {
    //
    // Initialize Client with default Bucket and Organization
    //
    let client = InfluxDBClient(
            url: url,
            token: token,
            options: InfluxDBClient.InfluxDBOptions(bucket: bucket, org: org))
    //
    // Record defined as Data Point
    //
    let recordPoint = InfluxDBClient
            .Point("demo")
            .addTag(key: "type", value: "point")
            .addField(key: "value", value: .int(2))
    //
    // Record defined as Data Point with Timestamp
    //
    let recordPointDate = InfluxDBClient
            .Point("demo")
            .addTag(key: "type", value: "point-timestamp")
            .addField(key: "value", value: .int(2))
            .time(time: .date(Date()))
    try await client.makeWriteAPI().write(points: [recordPoint, recordPointDate])
    print("Written data:\n\n\([recordPoint, recordPointDate].map { "\t- \($0)" }.joined(separator: "\n"))")
    print("\nSuccess!")
    client.close()
  }
}

• sources - WriteData/WriteData.swift

Queries

The result retrieved by QueryApi could be formatted as a:

1. Lazy sequence of FluxRecord
2. Raw query response as a Data.

Query to FluxRecord

import ArgumentParser
import Foundation
import InfluxDBSwift
import InfluxDBSwiftApis
@main
struct QueryCpu: AsyncParsableCommand {
  @Option(name: .shortAndLong, help: "The name or id of the bucket destination.")
  private var bucket: String
  @Option(name: .shortAndLong, help: "The name or id of the organization destination.")
  private var org: String
  @Option(name: .shortAndLong, help: "Authentication token.")
  private var token: String
  @Option(name: .shortAndLong, help: "HTTP address of InfluxDB.")
  private var url: String
}
extension QueryCpu {
  mutating func run() async throws {
    //
    // Initialize Client with default Bucket and Organization
    //
    let client = InfluxDBClient(
            url: url,
            token: token,
            options: InfluxDBClient.InfluxDBOptions(bucket: bucket, org: org))
    // Flux query
    let query = """
                from(bucket: "\(self.bucket)")
                    |> range(start: -10m)
                    |> filter(fn: (r) => r["_measurement"] == "cpu")
                    |> filter(fn: (r) => r["cpu"] == "cpu-total")
                    |> filter(fn: (r) => r["_field"] == "usage_user" or r["_field"] == "usage_system")
                    |> last()
                """
    print("\nQuery to execute:\n\(query)\n")
    let records = try await client.queryAPI.query(query: query)
    print("Query results:")
    try records.forEach { print(" > \($0.values["_field"]!): \($0.values["_value"]!)") }
    client.close()
  }
}

• sources - QueryCpu/QueryCpu.swift

Query to Data

@main
struct QueryCpuData: AsyncParsableCommand {
  @Option(name: .shortAndLong, help: "The name or id of the bucket destination.")
  private var bucket: String
  @Option(name: .shortAndLong, help: "The name or id of the organization destination.")
  private var org: String
  @Option(name: .shortAndLong, help: "Authentication token.")
  private var token: String
  @Option(name: .shortAndLong, help: "HTTP address of InfluxDB.")
  private var url: String
}
extension QueryCpuData {
  mutating func run() async throws {
    //
    // Initialize Client with default Bucket and Organization
    //
    let client = InfluxDBClient(
            url: url,
            token: token,
            options: InfluxDBClient.InfluxDBOptions(bucket: bucket, org: org))
    // Flux query
    let query = """
                from(bucket: "\(self.bucket)")
                    |> range(start: -10m)
                    |> filter(fn: (r) => r["_measurement"] == "cpu")
                    |> filter(fn: (r) => r["cpu"] == "cpu-total")
                    |> filter(fn: (r) => r["_field"] == "usage_user" or r["_field"] == "usage_system")
                    |> last()
                """
    print("\nQuery to execute:\n\(query)\n")
    let response = try await client.queryAPI.queryRaw(query: query)
    let csv = String(decoding: response, as: UTF8.self)
    print("InfluxDB response: \(csv)")
    client.close()
  }
}

• sources - QueryCpuData/QueryCpuData.swift

Parameterized queries

InfluxDB Cloud supports Parameterized Queries that let you dynamically change values in a query using the InfluxDB API. Parameterized queries make Flux queries more reusable and can also be used to help prevent injection attacks.

InfluxDB Cloud inserts the params object into the Flux query as a Flux record named params. Use dot or bracket notation to access parameters in the params record in your Flux query. Parameterized Flux queries support only int , float, and string data types. To convert the supported data types into other Flux basic data types, use Flux type conversion functions.

Parameterized query example:

⚠️ Parameterized Queries are supported only in InfluxDB Cloud, currently there is no support in InfluxDB OSS.

import ArgumentParser
import Foundation
import InfluxDBSwift
import InfluxDBSwiftApis
@main
struct ParameterizedQuery: AsyncParsableCommand {
  @Option(name: .shortAndLong, help: "The bucket to query. The name or id of the bucket destination.")
  private var bucket: String
  @Option(name: .shortAndLong,
          help: "The organization executing the query. Takes either the `ID` or `Name` interchangeably.")
  private var org: String
  @Option(name: .shortAndLong, help: "Authentication token.")
  private var token: String
  @Option(name: .shortAndLong, help: "HTTP address of InfluxDB.")
  private var url: String
}
extension ParameterizedQuery {
  mutating func run() async throws {
    // Initialize Client with default Organization
    let client = InfluxDBClient(
            url: url,
            token: token,
            options: InfluxDBClient.InfluxDBOptions(bucket: bucket, org: org))
    for index in 1...3 {
      let point = InfluxDBClient
              .Point("demo")
              .addTag(key: "type", value: "point")
              .addField(key: "value", value: .int(index))
      try await client.makeWriteAPI().write(point: point)
    }
    // Flux query
    let query = """
                from(bucket: params.bucketParam)
                    |> range(start: -10m)
                    |> filter(fn: (r) => r["_measurement"] == params.measurement)
                """
    // Query parameters [String:String]
    let queryParams = ["bucketParam": "\(bucket)", "measurement": "demo"]
    print("\nQuery to execute:\n\n\(query)\n\n\(queryParams)")
    let records = try await client.queryAPI.query(query: query, params: queryParams)
    print("\nSuccess response...\n")
    try records.forEach { print(" > \($0.values["_field"]!): \($0.values["_value"]!)") }
    client.close()
  }
}

• sources - ParameterizedQuery/ParameterizedQuery.swift

Delete data

The DeleteAPI supports deletes points from an InfluxDB bucket. Use the DeletePredicateRequest identifies which points to delete.

import ArgumentParser
import Foundation
import InfluxDBSwift
import InfluxDBSwiftApis
@main
struct DeleteData: AsyncParsableCommand {
  @Option(name: .shortAndLong, help: "Specifies the bucket name to delete data from.")
  private var bucket: String
  @Option(name: .shortAndLong,
          help: "Specifies the organization name to delete data from.")
  private var org: String
  @Option(name: .shortAndLong, help: "Authentication token.")
  private var token: String
  @Option(name: .shortAndLong, help: "HTTP address of InfluxDB.")
  private var url: String
  @Option(name: .shortAndLong, help: "InfluxQL-like delete predicate statement.")
  private var predicate: String
}
extension DeleteData {
  mutating func run() async throws {
    // Initialize Client with default Organization
    let client = InfluxDBClient(
            url: url,
            token: token,
            options: InfluxDBClient.InfluxDBOptions(org: self.org))
    // Create DeletePredicateRequest
    let predicateRequest = DeletePredicateRequest(
            start: Date(timeIntervalSince1970: 0),
            stop: Date(),
            predicate: predicate)
    try await client.deleteAPI.delete(predicate: predicateRequest, bucket: bucket, org: org)
    print("\nDeleted data by predicate:\n\n\t\(predicateRequest)")
    // Print date after Delete
    try await queryData(client: client)
    client.close()
  }
  private func queryData(client: InfluxDBClient) async throws {
    let query = """
                from(bucket: "\(bucket)")
                    |> range(start: 0)
                    |> filter(fn: (r) => r["_measurement"] == "server")
                    |> pivot(rowKey:["_time"], columnKey: ["_field"], valueColumn: "_value")
                """
    let response = try await client.queryAPI.query(query: query)
    print("\nRemaining data after delete:\n")
    try response.forEach { record in
      let provider = record.values["provider"]!
      let production = record.values["production"]
      let app = record.values["app"]
      return print("\t\(provider),production=\(production!),app=\(app!)")
    }
  }
}

• sources - DeleteData/DeleteData.swift

Management API

The client supports following management API:

The following example demonstrates how to use a InfluxDB 2.0 Management API to create new bucket. For further information see docs and examples.

import ArgumentParser
import Foundation
import InfluxDBSwift
import InfluxDBSwiftApis
@main
struct CreateNewBucket: AsyncParsableCommand {
  @Option(name: .shortAndLong, help: "New bucket name.")
  private var name: String
  @Option(name: .shortAndLong, help: "Duration bucket will retain data.")
  private var retention: Int64 = 3600
  @Option(name: .shortAndLong, help: "Specifies the organization name.")
  private var org: String
  @Option(name: .shortAndLong, help: "Authentication token.")
  private var token: String
  @Option(name: .shortAndLong, help: "HTTP address of InfluxDB.")
  private var url: String
}
extension CreateNewBucket {
  mutating func run() async throws {
    // Initialize Client and API
    let client = InfluxDBClient(url: url, token: token)
    let api = InfluxDB2API(client: client)
    let orgId = (try await api.organizationsAPI.getOrgs(org: org)!).orgs?.first?.id
    // Bucket configuration
    let request = PostBucketRequest(
            orgID: orgId!,
            name: name,
            retentionRules: [RetentionRule(type: RetentionRule.ModelType.expire, everySeconds: retention)])
    // Create Bucket
    let bucket = try await api.bucketsAPI.postBuckets(postBucketRequest: request)!
    // Create Authorization with permission to read/write created bucket
    let bucketResource = Resource(
            type: Resource.ModelType.buckets,
            id: bucket.id,
            orgID: orgId
    )
    // Authorization configuration
    let authorizationRequest = AuthorizationPostRequest(
            description: "Authorization to read/write bucket: \(name)",
            orgID: orgId!,
            permissions: [
              Permission(action: Permission.Action.read, resource: bucketResource),
              Permission(action: Permission.Action.write, resource: bucketResource)
            ])
    // Create Authorization
    let authorization = try await api.authorizationsAPI.postAuthorizations(authorizationPostRequest: authorizationRequest)!
    print("The bucket: '\(bucket.name)' is successfully created.")
    print("The following token could be use to read/write:")
    print("\t\(authorization.token!)")
    client.close()
  }
}

• sources - CreateNewBucket/CreateNewBucket.swift

Advanced Usage

Default Tags

Sometimes is useful to store same information in every measurement e.g. hostname, location, customer. The client is able to use static value or env variable as a tag value.

The expressions:

• California Miner - static value
• ${env.HOST_NAME} - environment property

Example

client = InfluxDBClient(
        url: "http://localhost:8086",
        token: "my-token",
        options: InfluxDBClient.InfluxDBOptions(bucket: "my-bucket", org: "my-org"))
let tuple: InfluxDBClient.Point.Tuple
        = (measurement: "mem", tags: ["tag": "a"], fields: ["value": .int(3)], time: nil)
let records: [Any] = [
        InfluxDBClient.Point("mining")
                .addTag(key: "sensor_state", value: "normal")
                .addField(key: "depth", value: .int(2)),
        tuple
]
let defaultTags = InfluxDBClient.PointSettings()
        .addDefaultTag(key: "customer", value: "California Miner")
        .addDefaultTag(key: "sensor_id", value: "${env.SENSOR_ID}")
try await client.makeWriteAPI(pointSettings: defaultTags).writeRecords(records: records)
print("Successfully written default tags")

Example Output

mining,customer=California\ Miner,sensor_id=123-456-789,sensor_state=normal depth=2i
mining,customer=California\ Miner,sensor_id=123-456-789,sensor_state=normal pressure=3i

Proxy and redirects

⚠️ The connectionProxyDictionary cannot be defined on Linux. You have to set HTTPS_PROXY or HTTP_PROXY system environment.

You can configure the client to tunnel requests through an HTTP proxy by connectionProxyDictionary option:

var connectionProxyDictionary = [AnyHashable: Any]()
connectionProxyDictionary[kCFNetworkProxiesHTTPEnable as String] = 1
connectionProxyDictionary[kCFNetworkProxiesHTTPProxy as String] = "localhost"
connectionProxyDictionary[kCFNetworkProxiesHTTPPort as String] = 3128
let options: InfluxDBClient.InfluxDBOptions = InfluxDBClient.InfluxDBOptions(
        bucket: "my-bucket",
        org: "my-org",
        precision: .ns,
        connectionProxyDictionary: connectionProxyDictionary)
client = InfluxDBClient(url: "http://localhost:8086", token: "my-token", options: options)

For more info see - URLSessionConfiguration.connectionProxyDictionary, Global Proxy Settings Constants.

Redirects

Client automatically follows HTTP redirects. You can disable redirects by an urlSessionDelegate configuration:

class DisableRedirect: NSObject, URLSessionTaskDelegate {
    func urlSession(_ session: URLSession,
                    task: URLSessionTask,
                    willPerformHTTPRedirection response: HTTPURLResponse,
                    newRequest request: URLRequest,
                    completionHandler: @escaping (URLRequest?) -> Void) {
        completionHandler(nil)
    }
}
let options = InfluxDBClient.InfluxDBOptions(
        bucket: "my-bucket",
        org: "my-org",
        urlSessionDelegate: DisableRedirect())
client = InfluxDBClient(url: "http://localhost:8086", token: "my-token", options: options)

For more info see - URLSessionDelegate.

Contributing

If you would like to contribute code you can do through GitHub by forking the repository and sending a pull request into the master branch.

Build Requirements:

• swift 5.3 or higher

Build source and test targets:

swift build --build-tests

Run tests:

./Scripts/influxdb-restart.sh
swift test

Check code coverage:

./Scripts/influxdb-restart.sh
swift test --enable-code-coverage

You could also use a docker-cli without installing the Swift SDK:

make docker-cli
swift build

License

The client is available as open source under the terms of the MIT License.