Skip to content

Latest commit

 

History

History
94 lines (81 loc) · 2.96 KB

write_operation_notes.md

File metadata and controls

94 lines (81 loc) · 2.96 KB
Raw

Write operation

Important

Influx use timestamp in nanosecond precision. Don't forget to transform your time.

val wrongTs = System.currentTimeMillis() // not right
val rightTs = wrongTs * 1000000

// or just not specify it, and InfluxServer will generate it himself.

Write syntax

First of all create database connection. In this situation we choose non type safe connection

val db = influx.database("db")

There are many opportunities to store data. They separated on 2 groups. First one is non typesafe. You can used from db instance. Like below First one save point in pure Line Protocol Format

// single
db.writeNative("cpu_load_short,host=server02,region=us-west value=0.55 1422568543702900257")
res0: Future[Result]

// bulk
db.bulkWriteNative(Seq("cpu_load_short,host=server02,region=us-west value=0.55 1422568543702900257", "cpu_load_short,host=server03,region=us-west value=0.56 1422568539002900257"))
res1: Future[Result]

writePoint add possibility to save Point into InfluxDB:

// single
val p1 = Point("testMeasurement")
            .addTag("name", "Jimbo")
            .addTag("surname", "Bimbo")
            .addField("age", 56)

db.writePoint(p1)
res0: Future[Result]

// bulk
db.bulkWritePoints(Seq(p1, ...))
res0: Future[Result]

Another option is to write from file. File format example:

test1,host=server02 value=0.67
test1,host=server02,region=us-west value=0.55 1422568543702900257
test1,direction=in,host=server01,region=us-west value=2.0 1422568543702900257

every point must be on separate line in Line Protocol format. And then:

db.writeFromFile("path/to/your/file")
res0: Future[Result]

The same methods able for InfluxUdpClient. Like:

updInflux.writeNative("cpu_load_short,host=server02,region=us-west value=0.55 1422568543702900257")
res0: Future[Unit]

main difference in return type. All udp methods return type is result of UDP Protocol nature

The second group are typesafe operation. To use it create type safe connection:

val meas = influx.measurement[FakeEntity]("db", "meas")

That one can take any type that have implicit InfluxWriterobject in the scope, that parse your object to Line Protocol String. For example:

import com.github.fsanaulla.annotation._

case class FakeEntity(@tag firstName: String,
                      @tag lastName: String,
                      @field age: Int)
                      
implicit val wr = new InfluxWriter[FakeEntity] {
    def write(fe: FakeEntity): String = {
        // parsing code
    }
}

Then you can simply:

val fe = FakeEntity("Name", "Surname", 54)

// single
meas.write(fe)
res0: Future[Result]

// bulk
meas.bulkWrite(Seq(fe, ...))
res0: Future[Result]

To avoid boilerplate code, and headache. Just use Macros.