kleinlennart

add_tweet_count <- function(dat) {
  return(
    dat %>%
      group_by(user_id) %>%
      mutate(
        tweet_count = n()
      ) %>%
      ungroup() # to remove grouped_df class
  )
}

kleinlennart

# clean Hashtag extraction method
## evtl. noch Umlaute entfernen!

add_hashtags <- function(dat) {
  return(
    dat %>%
      mutate(
        hashtags = text %>% tolower() %>% str_extract_all("#[[:alnum:]_]+")
      )
  )

kleinlennart

dat <- dat %>% 
  rowwise() %>% 
  mutate(bundesland = unlist(bundesland),
         community = unlist(community),
         fach = unlist(fach)) %>% 
  ungroup() # important to remove rowwise_df class! (slows down all other operations)

kleinlennart

# Mutate numeric (0 / 1) coded columns to logical
dat <- s1 %>% 
  mutate(across(where(is.numeric), as.logical))

kleinlennart

dat %>% select_if(is.list)

kleinlennart

# Length of sublist
dat$media_url[sapply(dat$media_url, function(x) length(x) == 2)] 
                     
# not empty sublists                     
dat$media_url[sapply(dat$media_url, function(x) !is.na(x))]                                    
dat$media_url[sapply(dat$media_url, function(x) all(!is.na(x)))] # multiple NA values per sublist     
                     
na.omit.list <- function(list) {
  return(list[sapply(list, function(x) all(!is.na(x)))])
}

kleinlennart

drake_plan(
  # dplyr::distinct instead?
  user_data = tidy_data %>% unique(user_id) %>% select(user_related_cols), # remove "tweet level" vars for faster runtime
  
  geo_data = do_user_data_stuff(),
  
  
  # left_join: Join matching rows from b to a.
  joined_data = target(command = left_join(user_data, tidy_data, by = "user_id"),
                       format = "fst" # useful ???

kleinlennart

# needs to be outside of the drake plan for `rmd_files = !!pages_paths` to work
pages_paths <- dir("reports", pattern = "*.Rmd", full.names = TRUE)
# TODO: exclude _file.Rmd with regex in dir

plan <- drake_plan(
  render_pages = target(
    # TODO: Add a trigger dependency on _site.yml
    command = rmarkdown::render(
      knitr_in(rmd_files),
      output_dir = file_out("docs/"),

kleinlennart

codebook = file_in(URL)

kleinlennart

---
date: "`r format(Sys.time(), '%d. %B, %Y')`"
---