As a quick recap from part 1, scrobbling is a service offered by
Last.fm, where any music you listen to on other platforms (i.e.,
Spotify) is recorded and stored by Last.fm, along with a timestamp and
associated metadata. scrobbler
is intended as an interface
to that database, so you can download and explore your music history
with the full freedom of R. Please note, this 2nd part assumes you have
already set up a Last.fm account and have at least one song already
scrobbled.
As an aside on what scrobbler
is not:
scrobbler
is not intended to be a general wrapper around
the Last.fm scrobble API. The full API has a lot of
methods available, but the scrobbler
package is designed to
do one thing; download your listening history into R. If there is other
functionality of the API you’d like to see in scrobbler
,
please reach out on github.
In addition to having a Last.fm account, downloading your scrobbles requires you to have an API key (this just authenricates you with Last.fm). To get an API key, you can go through the process here (it takes less than 5 mins). Note you should only have to fill out the name and descripton, and you just have to describe that you are downloading your scrobble history for personal analysis. Dont worry about the callback url or homepage. Once you’ve gone through that step, you should be provided with an API key. Save this key somewhere secure!
Now it should be smooth sailing from here. Make sure you start by loading scrobbler
And now call the download_scrobbles
function, using your
username and API key, to start downloading your tracks.
And thats it! You should now have a dataframe called ‘my_songs’ that contains info one row for each song you listened to, along with its timestamp. You can now analyse this in any way you choose. Lets have a look at what it looks like
#> song_mbid song_title
#> 1 192915d7-c2df-44f6-9e08-b7d80745bdd3 So Soon
#> 2 6dd374d1-e707-4de0-89b3-889fbb7d7bad B Team
#> 3 45d25340-5791-4f93-a642-94494b057646 Toy Soldiers
#> 4 0f361896-d6fc-4179-9987-47bf59437c83 Stutter
#> 5 34d419dd-eaf7-48de-b4df-704c61463cd7 Fallout
#> 6 3c8fe6d5-66ac-3b8b-a3f5-36f63fcff693 Porcelain
#> 7 0050b4bf-21d2-47f2-a306-4af3dad79018 Desperate Measures
#> 8 4d834fbe-e72d-3e06-88d9-da4945421182 Truth Or Dare
#> 9 0ddd52c6-45ed-4064-add0-5d0f0f29af34 By Now
#> 10 0651a786-0711-4ab8-8eee-148d277c143a My World
#> artist_mbid artist
#> 1 e358276d-4377-4b9b-88dd-db0d17b0e3c6 Marianas Trench
#> 2 e358276d-4377-4b9b-88dd-db0d17b0e3c6 Marianas Trench
#> 3 e358276d-4377-4b9b-88dd-db0d17b0e3c6 Marianas Trench
#> 4 e358276d-4377-4b9b-88dd-db0d17b0e3c6 Marianas Trench
#> 5 e358276d-4377-4b9b-88dd-db0d17b0e3c6 Marianas Trench
#> 6 e358276d-4377-4b9b-88dd-db0d17b0e3c6 Marianas Trench
#> 7 e358276d-4377-4b9b-88dd-db0d17b0e3c6 Marianas Trench
#> 8 e358276d-4377-4b9b-88dd-db0d17b0e3c6 Marianas Trench
#> 9 e358276d-4377-4b9b-88dd-db0d17b0e3c6 Marianas Trench
#> 10 0103c1cc-4a09-4a5d-a344-56ad99a77193 Avril Lavigne
#> album_mbid album date_unix
#> 1 180bb020-8349-4031-b8a3-bb544a396d84 Ever After NA
#> 2 180bb020-8349-4031-b8a3-bb544a396d84 Ever After 1608151945
#> 3 180bb020-8349-4031-b8a3-bb544a396d84 Ever After 1608151694
#> 4 180bb020-8349-4031-b8a3-bb544a396d84 Ever After 1608151493
#> 5 180bb020-8349-4031-b8a3-bb544a396d84 Ever After 1608151238
#> 6 180bb020-8349-4031-b8a3-bb544a396d84 Ever After 1608150918
#> 7 180bb020-8349-4031-b8a3-bb544a396d84 Ever After 1608150689
#> 8 180bb020-8349-4031-b8a3-bb544a396d84 Ever After 1608150459
#> 9 180bb020-8349-4031-b8a3-bb544a396d84 Ever After 1608150203
#> 10 002a0094-40b7-4403-99ab-f3b0daeffacd Let Go 1608118840
#> date
#> 1 <NA>
#> 2 16 Dec 2020, 20:52
#> 3 16 Dec 2020, 20:48
#> 4 16 Dec 2020, 20:44
#> 5 16 Dec 2020, 20:40
#> 6 16 Dec 2020, 20:35
#> 7 16 Dec 2020, 20:31
#> 8 16 Dec 2020, 20:27
#> 9 16 Dec 2020, 20:23
#> 10 16 Dec 2020, 11:40
The song_, artist_, and album_mbid columns refer to MusicBrainz ID. MusicBrainz is a music encyclopedia that contains lots of data about songs, artists, and albums. You can use the ID to access that entry in MusicBrainz if you want additional metadata (but be warned, these columns may not always be filled). Accessing MusicBrainz will be covered in another post.
As you may have noticed, it takes a while to download all your music.
It would be a pain if you had to go through that process everytime you
wanted to make sure your scrobbles were up to date.
scrobbler
provides a update_scrobbles
function
that makes this easier.
To use update_scrobbles
you pass in an existing
dataframe of songs, and supply a date columm (you should use the
date_unix
column that is included in scrobble dataframes by
default). This will then only download your new scrobbles not included
in the original dataframe, and create a new one with all your
tracks.
updated_songs <- update_scrobbles(data = my_songs,
timestamp_column = "date_unix",
username = "your_username",
api_key = "your_api_key"
)
This is much faster than having to redownload every time. Note that a
current limitation is that if you have added or removed any columns,
this will fail as the new columns will not match the old ones. I plan to
find a neater solution in future versions of scrobbler
.
If you have any questions about scrobbler
, feel free to
reach out on github. Have fun!