A easy-use client to influxdb
This is an InfluxDB driver for Rust.
This project has been able to run properly, PR is welcome.
[dependencies]
influx_db_client = "^0.7.0"
tokio = { version = "1", features = ["rt-multi-thread"] }
use influx_db_client::{Client, Point, Points, Precision, point, points}; fn main() { // default with "http://127.0.0.1:8086", db with "test" let client = Client::default().set_authentication("root", "root"); let point = point!("test1") .add_field("foo", "bar") .add_field("integer", 11) .add_field("float", 22.3) .add_field("'boolean'", false); let point1 = Point::new("test1") .add_tag("tags", "\\\"fda") .add_tag("number", 12) .add_tag("float", 12.6) .add_field("fd", "'3'") .add_field("quto", "\\\"fda") .add_field("quto1", "\"fda"); let points = points!(point1, point); tokio::runtime::Runtime::new().unwrap().block_on(async move { // if Precision is None, the default is nanosecond // Multiple write client.write_points(points, Some(Precision::Seconds), None).await.unwrap(); // query, it's type is Option<Vec<Node>> let res = client.query("select * from test1", None).await.unwrap(); println!("{:?}", res.unwrap()[0].series) }); }
Client defaults to reqwest::Client when the default reqwest feature is enabled,
but it is generic over the HTTP implementation.
If you need a custom transport, implement the transport traits for the APIs you want to support,
then create it with Client::new_with_client(...).
Borrowing APIs such as ping_borrow, get_version_borrow, query_borrow,
query_chunked_borrow, write_point_borrow, write_points_borrow, and the corresponding
query-backed *_borrow management APIs require BorrowHttpClient and BorrowHttpResponse.
Borrowed chunked queries also require BorrowChunkedHttpResponse. Spawn-safe APIs such as ping,
get_version, query, query_chunked, write_point, write_points, and the query-backed
management commands require HttpClient and HttpResponse. Spawn-safe chunked queries also
require ChunkedHttpResponse. You can implement borrowed-only, spawn-safe-only, or both modes on the same
transport type.
Borrowing APIs use borrowed HttpRequest data, while spawnable query/write APIs receive owned
HttpRequest<'static> values. Chunked responses now expose an async byte stream rather than a
blocking reader.
query_chunked is an incompatible API change in this release: it now returns an async stream
instead of a synchronous iterator. Add futures = "0.3" if you want to consume it with
StreamExt::next:
use futures::StreamExt; use influx_db_client::{Client, Query}; # #[cfg(feature = "reqwest")] { # tokio::runtime::Runtime::new().unwrap().block_on(async move { let client = Client::default(); let mut stream = client.query_chunked("select * from test1", None).await.unwrap(); while let Some(result) = stream.next().await { let query: Query = result.unwrap(); println!("{:?}", query.results); } # }); # }
To avoid compiling reqwest, disable default features and provide your own HTTP client:
[dependencies] influx_db_client = { version = "^0.7.0", default-features = false }
The crate's default reqwest/default-tls path currently resolves to a rustls-based backend.
This is an incompatible feature-name update: the previous rustls-tls* feature names were removed.
To build the default reqwest transport with the native-tls backend, disable default features
and enable one of the native-tls* features explicitly. On Linux that typically means OpenSSL;
on macOS and Windows it uses the platform TLS stack:
[dependencies] influx_db_client = { version = "^0.7.0", default-features = false, features = ["native-tls"] }
use influx_db_client::{Point, UdpClient, point}; fn main() { let mut udp = UdpClient::new("127.0.0.1:8089".parse().unwrap()); udp.add_host("127.0.0.1:8090".parse().unwrap()); let point = point!("test").add_field("foo", "bar"); udp.write_point(point).unwrap(); }
This is the API Document, it may apply to version 1.0 or higher.
I have tested it in version 1.0.2/1.3.5/1.5.
Because influent seems to have no longer updated, and only support to the 0.9 version. I read influent.rs and influxdb-python source, and then try to write a library for 1.0+ version for support for my own use.