TidyTuesday Week 12: Programming Languages

library(tidytuesdayR)
library(tidyverse)
library(skimr)
library(ggthemes)
library(gt)
library(ggrepel)

# Get the Data

# Read in with tidytuesdayR package 
# This loads the readme and all the datasets for the week of interest

# Either ISO-8601 date or year/week works!

#tuesdata <- tidytuesdayR::tt_load('2023-03-21')
tuesdata <- tidytuesdayR::tt_load(2023, week = 12)

languages <- tuesdata$languages

skim(languages)

programming_lang <- languages %>%
  filter(type == 'pl') %>%
  select(-starts_with("github"), -starts_with("wikipedia"),
         -description, -creators, -(website:semantic_scholar)) %>%
  mutate(language_rank = language_rank + 1)

skim(programming_lang)

programming_lang <- programming_lang %>%
  filter(features_has_comments == TRUE)

skim(programming_lang)

programming_lang %>% filter(is.na(line_comment_token) == FALSE) %>%
  group_by(line_comment_token) %>%
  count(line_comment_token) %>%
  ggplot(aes(fct_rev(fct_reorder(line_comment_token, n)), n)) +
  geom_col(fill = "dodgerblue2") +
  ylab("Count") +
  xlab("Comment Token") +
  ggtitle("Popularity of different comment tokens") +
  theme_classic() +
  theme(axis.text.x = element_text(angle = 45,  vjust = 0.25, hjust = 0.25))

# | label: table-tokens
programming_lang2 <- programming_lang %>%
  filter(is.na(line_comment_token) == FALSE) %>%
  count(line_comment_token, sort = TRUE) 

programming_lang2 %>%
gt() %>%
tab_header(title = "Most Common Comment Tokens") %>%
cols_label(line_comment_token = "Token", n = "# of Languages that use token")

programming_lang %>% filter(is.na(line_comment_token) == FALSE) %>%
  group_by(line_comment_token) %>%
  summarize(avg_rank = mean(language_rank)) %>%
  ggplot(aes((fct_reorder(line_comment_token, avg_rank)), avg_rank)) +
  geom_col(fill = "dodgerblue2") +
  ylab("Average Rank of Language") +
  xlab("Comment Token") +
  ggtitle("Average rank of languages using different comment tokens") +
  theme_classic() +
  theme(axis.text.x = element_text(angle = 45,  vjust = 0.25, hjust = 0.25))

programming_lang %>% filter(line_comment_token == "*>") %>%
  select(title, language_rank, line_comment_token)

# A tibble: 1 × 3
  title language_rank line_comment_token
  <chr>         <dbl> <chr>             
1 COBOL            19 *>                

programming_lang %>%
  filter(is.na(line_comment_token) == FALSE) %>%
  ggplot(aes(line_comment_token, language_rank)) +
  geom_boxplot(color = "dodgerblue2") +
  ggtitle("The rank of languages by token.") +
  xlab("Token") +
  ylab ("Language Rank") +
  theme_classic()

programming_lang %>%
  filter(is.na(line_comment_token) == FALSE) %>%
  ggplot(aes(fct_reorder(line_comment_token, language_rank,
                         .fun = median, .desc = FALSE), language_rank)) +
  geom_boxplot(color = "dodgerblue2") +
  ggtitle("The rank of languages by token") +
  xlab("Token") +
  ylab("Language Rank") +
    theme_classic()

programming_lang3 <- programming_lang %>%
  filter(is.na(line_comment_token) == FALSE) %>%
  group_by(line_comment_token) %>%
  summarize(highest_rank = min(language_rank)) 

join_madness <- programming_lang2 %>%
  left_join(programming_lang3, by = "line_comment_token") %>% 
  left_join(programming_lang, 
            by = c("highest_rank" = "language_rank",
                   "line_comment_token" = "line_comment_token")) 

join_madness <- join_madness %>%
  select(line_comment_token, n, highest_rank, title, appeared, number_of_users,
         number_of_jobs)

join_madness %>%
  ggplot(aes(highest_rank, n, size = log(number_of_users), 
             color = log(number_of_users), label = line_comment_token)) +
  scale_y_log10() +
  scale_x_log10() +
  geom_text_repel(show.legend = FALSE) +
  ggtitle("Popularity of tokens by language rank and usage") +
  xlab("Highest Rank of language using Token") +
  ylab("Number of Languages using token") +
  theme_classic()

join_madness %>%
  ggplot(aes(fct_rev(fct_reorder(line_comment_token, highest_rank)), n,
             size = log(number_of_users), color = log(number_of_users),
             label = title)) +
 # geom_point() +
  scale_y_log10() +
  geom_text_repel(show.legend = FALSE) +
   ggtitle("The Most Popular Language for Each Comment Token") +
  xlab("Token") +
  ylab("Number of languages using token") +
  theme_classic()

join_madness %>%
  ggplot(aes(line_comment_token, n, size = log(number_of_users), 
             color = log(number_of_users), label = title)) +
 # geom_point() +
  scale_y_log10() +
  geom_text_repel(show.legend = FALSE) +
   ggtitle("The Most Popular Language for Each Comment Token") +
  xlab("Token") +
  ylab("Number of languages using token") +
  theme_classic()

Citation

@online{e.sinks2023,
  author = {Louise E. Sinks},
  title = {TidyTuesday {Week} 12: {Programming} {Languages}},
  date = {2023-03-21},
  url = {https://lsinks.github.io/posts/2023-03-21-tidytuesday-programming-languages/},
  langid = {en}
}