API to Melbourne Pedestrian Data

Provides API to Melbourne pedestrian data in tidy data form.


Travis-CI BuildStatus CRAN_Status_Badge Downloads

The goal of rwalkr is to provide APIs to the pedestrian data from the City of Melbourne in tidy data form.

Installation

You could install the stable version from CRAN:

install.packages("rwalkr")

You could install the development version from Github using:

# install.packages("devtools")
devtools::install_github("earowang/rwalkr")

Usage

APIs

There are two APIs available to access Melbourne pedestrian data: compedapi and Socrata. The former drives the melb_walk() function, where counts are uploaded on a daily basis; the latter powers the melb_walk_fast() function, where counts are uploaded on a monthly basis. Given the function names, the function melb_walk_fast() pulls the data at a much faster speed than melb_walk().

The function melb_walk() specifies the starting and ending dates to be pulled, whereas melb_walk_fast() requires years to define the time frame. If a selection of sensors are of interest, melb_walk_fast() provides the flexibility for sensor choices.

library(rwalkr)
start_date <- as.Date("2017-07-01")
ped_walk <- melb_walk(from = start_date, to = start_date + 6L)
ped_walk
#> # A tibble: 7,224 x 5
#>   Sensor                     Date_Time           Date        Time Count
#>   <chr>                      <dttm>              <date>     <int> <int>
#> 1 Bourke Street Mall (North) 2017-07-01 00:00:00 2017-07-01     0   280
#> 2 Bourke Street Mall (South) 2017-07-01 00:00:00 2017-07-01     0   177
#> 3 Melbourne Central          2017-07-01 00:00:00 2017-07-01     0   826
#> 4 Town Hall (West)           2017-07-01 00:00:00 2017-07-01     0   682
#> 5 Princes Bridge             2017-07-01 00:00:00 2017-07-01     0     0
#> # … with 7,219 more rows
ped_run <- melb_walk_fast(year = 2016:2017, sensor = NULL) # NULL means all sensors
ped_run
#> # A tibble: 912,288 x 5
#>   Sensor                      Date_Time           Date        Time Count
#>   <chr>                       <dttm>              <date>     <int> <int>
#> 1 Alfred Place                2016-01-01 00:00:00 2016-01-01     0    NA
#> 2 Birrarung Marr              2016-01-01 00:00:00 2016-01-01     0  1405
#> 3 Bourke St-Russell St (West) 2016-01-01 00:00:00 2016-01-01     0  1900
#> 4 Bourke Street Mall (North)  2016-01-01 00:00:00 2016-01-01     0   461
#> 5 Bourke Street Mall (South)  2016-01-01 00:00:00 2016-01-01     0   883
#> # … with 9.123e+05 more rows

There are missing values (i.e. NA) in the dataset. By setting na.rm = TRUE in both functions, missing values will be removed.

Here’s an example to use ggplot2 for visualisation:

library(ggplot2)
ggplot(data = subset(ped_walk, Sensor == "Melbourne Central")) +
  geom_line(aes(x = Date_Time, y = Count))

The dictionary for checking sensor names between two functions is available through lookup_sensor().

It’s recommended to include an application token in melb_walk_fast(app_token = "YOUR-APP-TOKEN"), which you can sign up here.

Shiny app

The function melb_shine() launches a shiny app to give a glimpse of the data. It provides two basic plots: one is an overlaying time series plot, and the other is a dot plot indicating missing values. Below is a screen-shot of the shiny app.

News

rwalkr 0.4.0

  • Fixed status "400" for melb_walk_fast() (previously run_melb()) due to Socrata API changes.
  • Deprecated walk_melb(), run_melb() and shine_melb() in favour of suffixed function names.

rwalkr 0.3.4

  • Fixed date-time formatting for Socrata data update

rwalkr 0.3.3

API changes

  • Dropped the argument of tweak in walk_melb(), as the sensor names from the data source match with run_melb().

Minor changes

  • Dropped the match column in the data frame called from lookup_sensor().

rwalkr 0.3.2

Major changes

  • Returned a tibble (tbl_ts) instead of data.frame.

Minor changes

  • Specified the requirement version of shiny to the DESCRIPTION file.

rwalkr 0.3.1

Bug fixes

  • Fixed "Count" to be returned as integers instead of characters in run_melb(na.rm = FALSE).
  • Fixed duplicated data entries when walk_melb(tweak = TRUE).
  • Fixed one non-matching sensor in lookup_sensor().

Updates

  • A new sensor "Pelham St (S)" added to run_melb(), pull_sensor(), and lookup_sensor() using Socrata.
  • Changed the shiny app using shine_melb() to use walk_melb(tweak = TRUE).

rwalkr 0.3.0

New functions

  • run_melb() pulls Melbourne pedestrian data using Socrata, which is faster than walk_melb().
  • pull_sensor() pulls Melbourne pedestrian sensor locations using Socrata.
  • lookup_sensor() provides a dictionary for sensor names used in walk_melb() and run_melb().

Minor changes

  • Added new arguments na.rm = FALSE and tweak = FALSE to the function walk_melb(). If na.rm = TRUE, it removes NAs from the data. If tweak = TRUE, it ensures the consistency of sensor names to run_melb().

rwalkr 0.2.0

  • Added the function shine_melb() to launch a shiny app. It provides two basic plots to take a glimpse at the data: one is an overlaying time series plot and the other showing a dot plot of missing values.

rwalkr 0.1.0

  • Added a NEWS.md file to track changes to the package.
  • Added the function walk_melb() to scrape Melbourne pedestrian data.

Reference manual

It appears you don't have a PDF plugin for this browser. You can click here to download the reference manual.

install.packages("rwalkr")

0.4.0 by Earo Wang, 5 months ago


http://pkg.earo.me/rwalkr


Report a bug at https://github.com/earowang/rwalkr/issues


Browse source code at https://github.com/cran/rwalkr


Authors: Earo Wang [aut, cre]


Documentation:   PDF Manual  


MIT + file LICENSE license


Imports tidyr, dplyr, httr

Suggests shiny, plotly


See at CRAN