--- title: "EDC 2024" author: "Joe" date: "`r Sys.Date()`" editor: source format: html: theme: quartz #fontcolor: black toc: true toc-location: left code-fold: true code-tools: true df-print: kable fig-width: 10 fig-height: 7.5 embed-resources: true grid: body-width: 2000px execute: warning: false --- ```{r} <- Sys.time ()suppressPackageStartupMessages (library (ggpath))suppressPackageStartupMessages (library (plotly))suppressPackageStartupMessages (library (tidyverse))options (scipen = 10L)<- theme_minimal () + theme (text= element_text (size= 14 ),axis.text= element_text (size= 12 ),axis.title.y = element_text (margin = margin (t = 0 , r = 8 , b = 0 , l = 0 )),axis.title.x = element_text (margin = margin (t = 0 , r = 0 , b = 8 , l = 0 )))<- theme_minimal () + theme (axis.title.y = element_text (margin = margin (t = 0 , r = 8 , b = 0 , l = 0 )),axis.title.x = element_text (margin = margin (t = 0 , r = 0 , b = 8 , l = 0 )),text= element_text (size= 20 , face= "bold" , color= "white" ),axis.text.x= element_text (size= 10 , color= "white" ),axis.text.y= element_text (size= 13 , color= "white" ),plot.title= element_text (face= "bold" , color= "white" ),plot.background = element_rect (fill = "#5E61AF" ),plot.margin = margin (1 ,1 ,1.5 ,1.2 , "cm" ))# Create color pallete based on lineup image @ https://coolors.co/image-picker <- c ("#52BFEC" ,"#AA1880" ,"#EC0059" ,"#08BCDF" ,"#4C1064" , "#FF00BC" , "#2249CD" ,"#53007D" , "#FF6B02" ,"#B319B2" ,"#EAE100" , "#BF068F" )``` # Summary # EDC 2024 Lineup  # Convert lineup image to text [ imagetotext.io ](https://www.imagetotext.io/) ```{r} #https://www.imagetotext.io/ to get artist names from edc artist lineup PNG <- read_tsv ("C:/Users/joseph.mcgirr/Personal/R_fun/spotify/edc_2024_artists_text_from_png.txt" ) |> unique () |> arrange (artist)#edc_artists ``` # Collect Artist Data ## Spotify [ spotifyr ](https://www.rdocumentation.org/packages/spotifyr/versions/2.2.4) [ here ](https://developer.spotify.com/dashboard/create) . It is very quick and easy.```{r} #| include: false Sys.setenv (SPOTIFY_CLIENT_ID = '15ef5105d34c4cdaa761f8287847f50a' )Sys.setenv (SPOTIFY_CLIENT_SECRET = 'c1c04d84b2094e5994759aa3275b9683' )``` ```{r} #| eval: false #install.packages("spotifyr") library (spotifyr)# set up a dev account to get valid API IDs # Sys.setenv(SPOTIFY_CLIENT_ID = '########################') # Sys.setenv(SPOTIFY_CLIENT_SECRET = '#########################') <- get_spotify_access_token ()# Use search_spotify() to find spotify artist ids from artist names # No ID for Domina, Hint of Lavender, Marlie, VUIIIGUR <- vector ("character" , length (edc_artists$ artist))for (i in seq_along (edc_artists$ artist)){<- search_spotify (edc_artists$ artist[i])if (edc_artists$ artist[i] %in% toupper (i_search_spotify$ artists$ items$ name)){<- which (toupper (i_search_spotify$ artists$ items$ name) == edc_artists$ artist[i])[1 ]<- i_search_spotify$ artists$ items$ id[exact_artist_name_match]else {<- NA cat (paste0 (" \n No exact match for artist name: " ,edc_artists$ artist[i], " \n\n " ))# "\n", "Using: ", i_search_spotify$artists$items$name[1], # "\n", i_search_spotify$artists$items$external_urls.spotify[1], "\n\n")) # spotify_artist_id[[i]] <- i_search_spotify$artists$items$id[1] # cat(paste0("\nNo exact match for artist name: ",edc_artists$artist[i], # "\n", "Using: ", i_search_spotify$artists$items$name[1], # "\n", i_search_spotify$artists$items$external_urls.spotify[1], "\n\n")) $ spotify_artist_id <- spotify_artist_id# Use get_artist() to get genres, followers, and popularity <- vector ("character" , length (edc_artists$ artist))<- vector ("character" , length (edc_artists$ artist))<- vector ("character" , length (edc_artists$ artist))<- vector ("character" , length (edc_artists$ artist))for (i in seq_along (edc_artists$ spotify_artist_id)){if (! is.na (edc_artists$ spotify_artist_id[i])){<- get_artist (edc_artists$ spotify_artist_id[i])<- paste0 (i_artist_info$ genres, collapse = "," )<- i_artist_info$ followers$ total<- i_artist_info$ popularity<- ifelse (! is.null (i_artist_info$ images$ url[1 ]), i_artist_info$ images$ url[1 ], NA )else {<- followers[[i]] <- popularity[[i]] <- image_url[[i]] <- NA if (! is.na (genres[[i]]) & genres[[i]] == "" ){genres[[i]] <- NA }$ genres <- genres $ followers <- as.numeric (followers)$ popularity <- as.numeric (popularity)$ image_url <- image_url``` ## last.fm [ lastfmR ](https://github.com/ppatrzyk/lastfmR) ```{r} #| eval: false #devtools::install_github("ppatrzyk/lastfmR") library (lastfmR)# masks get_tracks() <- get_artist_info (artist_vector = edc_artists$ artist) |> tibble ()<- full_join (edc_artists, lastfm_artist_info)# write.table(edc_artists, "C:/Users/joseph.mcgirr/Personal/R_fun/spotify/edc_2024_artist_data.txt", row.names = F, quote = F, sep = "\t") ``` # Spotify Followers ```{r} #| fig-width: 15 #| fig-height: 9 <- read_tsv ("C:/Users/joseph.mcgirr/Personal/R_fun/spotify/edc_2024_artist_data.txt" )<- function (column_name, top, plot_title, include_images = "false" ){<- arrange (edc_artists, desc (!! sym (column_name))) |> head (top) <- ggplot (i_plot, aes (reorder (artist, !! sym (column_name)), !! sym (column_name))) + geom_segment (aes (x = reorder (artist, !! sym (column_name)), xend = reorder (artist, !! sym (column_name)), y= 0 , yend = !! sym (column_name), color = artist),linewidth = 3 ) + #geom_from_path(aes(path = image_url), width = 0.052) + coord_flip (clip = "off" ) + scale_color_manual (values = rep (my_cols,3 )) + + theme (axis.title.x= element_blank (),axis.title.y= element_blank (),legend.position = "none" ) + ggtitle (plot_title)if (include_images == "true" ){<- p1 + geom_from_path (aes (path = image_url), width = 0.052 ) return (p1)plot.top.artists ("followers" , 10 , "Top 10 EDC artists with the most followers on Spotify" , include_images = "true" )plot.top.artists ("followers" , 30 , "Top 30 EDC artists with the most followers on Spotify" , include_images = "false" )``` # Spotify "Popularity" ```{r} #| fig-width: 15 #| fig-height: 9 plot.top.artists ("popularity" , 10 , "Top 10 most popular EDC artists according to Spotify" , include_images = "true" )plot.top.artists ("popularity" , 30 , "Top 30 most popular EDC artists according to Spotify" , include_images = "false" )``` # last.fm Global Listeners ```{r} #| fig-width: 15 #| fig-height: 9 plot.top.artists ("global_listeners" , 10 , "Top 10 artists with the most listeners on last.fm" , include_images = "true" )plot.top.artists ("global_listeners" , 30 , "Top 30 artists with the most listeners on last.fm" , include_images = "false" )``` # Genres ```{r} #| fig-width: 15 #| fig-height: 9 # filter(edc_artists, !is.na(genres)) |> nrow() # n_distinct(edc_artists$genres) <- unlist (str_split (edc_artists$ genres, "," )) |> na.omit () |> as.character () |> str_trim () |> toupper () |> tibble ()names (all_genres) <- "genre" <- function (all_genres, column_name, plot_title){<- group_by (all_genres, genre) |> summarise (n_genres = dplyr:: n ()) |> arrange (desc (n_genres)) |> head (30 ) <- ggplot (i_plot, aes (reorder (genre, !! sym (column_name)), !! sym (column_name))) + geom_segment (aes (x = reorder (genre, !! sym (column_name)), xend = reorder (genre, !! sym (column_name)), y= 0 , yend = !! sym (column_name), color = genre),linewidth = 3 ) + coord_flip (clip = "off" ) + scale_color_manual (values = rep (my_cols,100 )) + + theme (axis.title.x= element_blank (),axis.title.y= element_blank (),legend.position = "none" ) + ggtitle (plot_title)return (p1)plot.top.genres (all_genres, "n_genres" , "Top 30 most represented Spotify genres" )# png(paste0("C:/Users/joseph.mcgirr/Personal/R_fun/spotify/genres/top_30_genres.png"), height = 9, width = 15,units = 'in', res = 1000) # plot.top.genres(all_genres, "n_genres", "Top 30 most represented Spotify genres") # dev.off() # # filter(edc_artists, !is.na(artist_tags)) |> nrow() # # n_distinct(edc_artists$artist_tags) # all_genres <- unlist(str_split(edc_artists$artist_tags, ";")) |> # na.omit() |> # as.character() |> # str_trim() |> # toupper() |> # tibble() # names(all_genres) <- "genre" # # # plot.top.genres(all_genres, "n_genres", "Top 30 most represented last.fm aritst tags") ``` ## Popular artists in each genre ```{r} #| fig-width: 15 #| fig-height: 9 <- group_by (all_genres, genre) |> summarise (n_genres = dplyr:: n ()) |> arrange (desc (n_genres)) |> filter (n_genres >= 10 ) |> pull (genre)for (i in seq_along (top_genres)){<- "popularity" <- filter (edc_artists, str_detect (toupper (genres), top_genres[i])) |> arrange (desc (popularity)) |> head (10 )<- ggplot (i_plot, aes (reorder (artist, !! sym (column_name)), !! sym (column_name))) + geom_segment (aes (x = reorder (artist, !! sym (column_name)), xend = reorder (artist, !! sym (column_name)), y= 0 , yend = !! sym (column_name), color = artist),linewidth = 3 ) + geom_from_path (aes (path = image_url), width = 0.052 ) + coord_flip (clip = "off" ) + scale_color_manual (values = rep (my_cols,3 )) + + theme (axis.title.x= element_blank (),axis.title.y= element_blank (),legend.position = "none" ) + ggtitle (paste0 ("Top 10 " ,top_genres[i]," artists by popularity" ))print (p1)# png(paste0("C:/Users/joseph.mcgirr/Personal/R_fun/spotify/genres/", i,".png"), height = 9, width = 15,units = 'in', res = 700) # print(p1) # dev.off() ``` # All artists in each genre [ See here ](https://docs.google.com/spreadsheets/d/1DFnTFjlDW9zS2Bkv8YediH602jCJOaxf42j6qbsDBA0/edit?usp=sharing) for a table of all artists labeled by each genre.```{r} #| results: asis <- unlist (str_split (edc_artists$ genres, "," )) |> na.omit () |> as.character () |> str_trim () |> tibble ()names (all_genres) <- "genre" # for(i in sort(unique(all_genres$genre))){ # # dat <- tibble() # dat <- filter(edc_artists, genres == i) |> arrange(desc(popularity)) |> select(artist, popularity) |> rename(!!i := "artist") # dat <- bind_rows(dat, # filter(edc_artists, str_detect(genres, paste0(i,",|,", i, ",|,", i))) |> arrange(desc(popularity)) |> rename(!!i := "artist")) |> # select(!!i) # # print(knitr::kable(dat)) # # cat("\n") # } <- vector ("character" , length (unique (all_genres$ genre)))for (i in seq_along (unique (all_genres$ genre))){<- sort (unique (all_genres$ genre))[i]<- filter (edc_artists, genres == i_genre) |> arrange (desc (popularity)) |> select (artist, popularity)<- bind_rows (i_dat,filter (edc_artists, str_detect (genres, paste0 (i_genre,",|," , i_genre, ",|," , i_genre)))) |> arrange (desc (popularity)) |> pull (artist)<- paste0 (i_dat, collapse = "," )<- tibble (genre = sort (unique (all_genres$ genre)), artists = artists_tagged_by_genre)# write.table(artists_tagged_by_genre, "C:/Users/joseph.mcgirr/Personal/R_fun/spotify/artists_in_each_genre.txt", row.names = F, quote = F, sep = "\t") ``` # Notes ## Run time ```{r} Sys.time () - start_time``` ## Session ```{r} sessionInfo ()``` 
