Influxr

The goal of influxr is to provide a simple interface for retrieving data from InfluxDB v2 buckets. It was written for data workflows in the IEQ Lab but can be generalised for any user. It replaces the archived influxdbclient CRAN package with direct HTTP calls via httr2, and adds features for data science workflows including timezone handling, monthly query chunking, compressed file caching, and incremental dataset updates.

Installation

You can install influxr from GitHub with:

# install.packages("pak")
pak::pak("IEQLab/influxr")

Or from a local clone:

devtools::install_local("/path/to/influxr")

Configuration

The easiest way to store your InfluxDB credentials is with influx_set_env():

influx_set_env(
  url   = "http://your-host:8086",
  token = "your-api-token",
  org   = "your-org"
)

This writes the values to your ~/.Renviron file and loads them in the current session. Call influx_set_env() with no arguments for interactive prompts.

You can also edit ~/.Renviron manually (use usethis::edit_r_environ() to open it):

INFLUXDB_URL=http://your-host:8086
INFLUXDB_TOKEN=your-api-token
INFLUXDB_ORG=your-org

Or pass credentials explicitly:

cfg <- influx_config(
  url   = "http://your-host:8086",
  token = "your-api-token",
  org   = "your-org"
)

Usage

library(influxr)

# Download a specific time range
data <- influx_get_range(
  measurements = c("tvoc", "temperature"),
  start = "2024-06-01",
  end = "2024-06-30"
)

# Incrementally update — picks up where the last download left off
new_data <- influx_get_update(
  measurements = c("tvoc", "temperature", "humidity"),
  save_files = TRUE
)

# Read back cached .csv.gz files
cached <- influx_read_cached("tvoc", start = "2024-06-01")

Key functions

Function	Purpose
influx_set_env()	Store InfluxDB credentials in ~/.Renviron
influx_config()	Create a connection configuration
influx_get_range()	Download data for a time range, with optional file saving
influx_get_update()	Incrementally download new data since the last fetch
influx_read_cached()	Read previously saved .csv.gz files
influx_query()	Execute a raw Flux query
influx_build_query()	Construct a Flux query string
influx_parse_time()	Parse dates/times with timezone handling
influx_chunk_range()	Split a time range into monthly/weekly/daily chunks

Time handling

All functions default to tz = "Australia/Sydney". Times are converted to UTC for Flux queries and back to local time in the returned data. The influx_parse_time() helper accepts Date objects, POSIXct, or character strings in common formats.

# All of these work
influx_parse_time("2024-06-15")
influx_parse_time("2024-06-15 14:30:00")
influx_parse_time(as.Date("2024-06-15"))
influx_parse_time(Sys.time())

Dependencies

influxr uses httr2 for HTTP requests and readr for CSV parsing. Other dependencies: dplyr, lubridate, stringr, glue, purrr, tibble.