Tools for querying the Withings API from R
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
R
man
.Rbuildignore
.gitignore
DESCRIPTION
LICENSE
LICENSE.md
NAMESPACE
README.md
rwithings.Rproj

README.md

rwithings

The goal of rwithings is to enable data extraction from the Withings Health Mate API into R. You will need to follow the instructions on the linked developer's guide to create a Withings account and a Withings developer app.

This package is still at alpha stages, so there could very well be bugs all over the place.

Installation

You can install the development version of rwithings with the following:

devtools::install_github("mathesong/rwithings")

Prerequisites

Before being able to use this package, and the Withings Health Mate API, you will need to create a Withings account and a Withings developer app. Instructions are provided in the developer's guide.

Note: when setting up an app, the Callback URI should be set to http://localhost:1410/. This is not supposed to be allowed, but it seems to work, and I am also not sure how to make it work using another address. Any input on this would be appreciated.

Examples

API Functions

The functions of the package basically correspond to the same commands listed in the developer's guide, with similar input arguments. Below are some examples.

client_id <- 12345
client_secret <- "hunter2"

token <- withings_auth(client_id, client_secret)

getmeas(token, meastype = 1, category=1, "2018-01-01", "2018-10-18")
getsleep(token, "2018-10-16", "2018-10-18")
getsleepsummary(token, "2018-10-16", "2018-10-18")

In the case of entering in start and end dates, I've put in place conversions so that you can enter it as a character vector in YYYY-MM-DD format to make quick queries more easy.

CSV Functions

Unfortunately, the API does not yet appear to have functions for extracting heart rate data, and possibly other things. This data can only be obtained by following these instructions to obtain a csv file. The data arrive in a format which is a bit frustrating to work with, so I've also included some functions for munging the raw data. Most of the raw data is structured according to start, duration and value columns, so for these csv files, the command to read them works as follows:

hr_data <- read_csv_startdurval('raw_tracker_hr.csv')