In jrminter/fitbitHelpR: Convenience functions for processing messy csv files exported from FitBit website

knitr::opts_chunk$set(echo = TRUE, warning=FALSE, comment=NA)
# suppressWarnings(suppressMessages(suppressPackageStartupMessages(library(dplyr))))

Introduction

The interface for downloading data from Fitbit trackers has changed over the years. Currently all of the R packages that use the Fitbit API are having problems with authorization. I developed this package to scrape data from very messy .csv files that can be downoaded from the website.

Get the csv file

I added three csv files I downloaded as external data in the package. We will use one of these.

inF <- system.file("extdata", "2019-01.csv", package = "fitbitHelpR")
lines <- readLines(inF) 
length(lines)

Let's look at lines 1 to 5:

for (i in seq(from=1, to=5)){
  print(lines[i])
}

Let's look at lines 34 to 40:

for (i in seq(from=34, to=50)){
  print(lines[i])
}

and finally lines 67 to 72

for (i in seq(from=67, to=72)){
  print(lines[i])
}

This is quite messy data...

Process this messy data

I wrote the function extract_fitbit_activity_data(csvFileName)to do this. It is not pretty but it works. Until they change the interface again...

library(fitbitHelpR)
tbl <- extract_fitbit_activity_data(inF)
tbl

jrminter/fitbitHelpR documentation built on May 24, 2019, 1:33 a.m.