podr is a very simple package containing two functions for reading and cleaning data from a podcast feed.

Installation

devtools::install_github("erramirez/podr")

Key Functions

podr is primarily composed of two functions:

1. The first function, readpod, will extract relevant information from a podcast feed according to the information available within the item node. Additionally, it will read and append the podcast title according to what is present in channel/title.
2. The second function, cleapod, will read a data frame created with readpod and reduce to a tibble with four variables:
   • podtitle: the title of the podcast
   • title: the episdoe title
   • date: the date the episode was published
   • showlength: the duration of the podcast in seconds

cleanpod is primarily used to clean up the date and duration information in a podcast feed in order to make it computable.

Usage

Let't say you were really interested in learning how much time you spent listening to your favorite podcast, Jordan, Jesse Go!, in 2017.

First we would use readpod to read the podcast feed.

library(podr)

jjgo <- readpod("http://thornmorris.libsyn.com/rss")

jjgo

## # A tibble: 529 x 14
##    title  pubDate guid  link  image description encoded enclosure duration
##    <chr>  <chr>   <chr> <chr> <chr> <chr>       <chr>   <chr>     <chr>   
##  1 Ep. 5… Tue, 3… 6158… http… ""    "<p>Jordan… "<p>Jo… ""        01:13:18
##  2 Ep. 5… Tue, 2… 4907… http… ""    "<p>Jordan… "<p>Jo… ""        01:13:35
##  3 Ep. 5… Tue, 1… 593e… http… ""    "<p>Jordan… "<p>Jo… ""        01:22:55
##  4 Ep. 5… Tue, 0… 7f52… http… ""    <p>Eliza S… <p>Eli… ""        01:24:50
##  5 Ep. 5… Tue, 0… 8b79… http… ""    <p>Live fr… <p>Liv… ""        01:42:59
##  6 Ep. 5… Tue, 1… e9bf… http… ""    "<p class=… "<p cl… ""        01:00:31
##  7 Ep. 5… Tue, 1… 868a… http… ""    <p>Comedia… <p>Com… ""        01:26:16
##  8 Ep. 5… Tue, 0… a0a8… http… ""    "<p>Comedi… <p>Com… ""        01:17:34
##  9 Ep. 5… Tue, 2… e349… http… ""    <p>Fan fav… <p>Fan… ""        01:08:55
## 10 Ep. 5… Tue, 2… 3322… http… ""    <p>Interna… <p>Int… ""        01:04:24
## # ... with 519 more rows, and 5 more variables: explicit <chr>,
## #   keywords <chr>, subtitle <chr>, episodeType <chr>, podtitle <fct>

Not sure how to get the rss feed url for your podcast and useing iTunes? Check out this tip.

We can then use cleanpod to clean up the podcast data to something usable to answer our question.

jjgoclean <- cleanpod(jjgo)

jjgoclean

## # A tibble: 529 x 4
##    podtitle          title                           date       showlength
##    <fct>             <chr>                           <date>          <dbl>
##  1 Jordan, Jesse GO! Ep. 516: The Tropical Itch wit… 2018-01-30       4398
##  2 Jordan, Jesse GO! Ep. 515: Catillion with Haley … 2018-01-23       4415
##  3 Jordan, Jesse GO! Ep. 514: Tree Bees with Kevin … 2018-01-16       4975
##  4 Jordan, Jesse GO! Ep. 513: Martinelli Toots with… 2018-01-09       5090
##  5 Jordan, Jesse GO! Ep. 512: Live in London with N… 2018-01-02       6179
##  6 Jordan, Jesse GO! Ep. 511: Bunny Shy with Elizab… 2017-12-19       3631
##  7 Jordan, Jesse GO! Ep. 510: Hammerstice with Josh… 2017-12-12       5176
##  8 Jordan, Jesse GO! Ep. 509: Hand to Hand to Mouth… 2017-12-05       4654
##  9 Jordan, Jesse GO! Ep. 508: Wide Dumper with Chri… 2017-11-28       4135
## 10 Jordan, Jesse GO! Ep. 507: Gooble Gooble with Jo… 2017-11-21       3864
## # ... with 519 more rows

Now that we have a cleaned data set that contains all the relevant information we can use a few simple tidy tools.

library(tidyverse)
library(lubridate)

jjgoclean2k17 <- jjgoclean %>% 
  filter(date > "2016-12-31" & date < "2018-01-01")

#find the total duration of Jordan, Jesse GO! in seconds
sum(jjgoclean2k17$showlength)

## [1] 240727

#find the total as a period 
seconds_to_period(sum(jjgoclean2k17$showlength))

## [1] "2d 18H 52M 7S"

So there you have it. You can use podr to find out Jordan, Jesse GO! put out 2d 18H 52M 7S of hilarious shows in 2017.

Updates

v0.2.0: 2018-01-30. Updated to accept rss feeds that store duration in seconds in addition to v0.1.0 support for hh:mm:ss or mm:ss.

Note

Please feel free to share issues, notes, and ideas. This is my first public R package and I'd love to learn from you.

made with RStudio and ❤️ in ☀️ Los Angeles