Plot reverse dependencies and download statistics

downloads.R

library(cranlogs)
library(dplyr)
library(lubridate)
library(ggplot2)
library(hrbrthemes)

pkgs <- c("curl", "RCurl")
df <- cranlogs::cran_downloads(pkgs, from = '2015-01-01', to = '2020-10-31')
monthly <- df %>% group_by(package = package, date=floor_date(date, "month")) %>%
  summarize(count=sum(count))

ggplot(monthly, aes(x=date, y=count, group=package, color=package)) +
  geom_smooth(se=F) +
  geom_point() +
  ggtitle("Popularity of libcurl CRAN packages (downloads)") +
  theme_ipsum() +
  scale_color_ipsum() +
  ylab("Download count")

revdeps.R

library(hrbrthemes)
library(ggplot2)
library(tidyr)

count_revdeps <- function(pkgs, start = '2015-01-01', end = '2020-11-01', recursive = TRUE, n = 25){
  start <- as.Date(start)
  end <- as.Date(end)
  dates <- seq(start, end, length.out = n)
  out <- matrix(nrow = n, ncol = length(pkgs))
  for(i in seq_along(dates)){
    date <- dates[i]
    cat(sprintf("Getting data from %s\n", date))
    url <- paste0('https://cran.microsoft.com/snapshot/', date)
    db <- available.packages(repos = url)
    deps <- tools::package_dependencies(pkgs, db = db, reverse = TRUE, recursive = recursive)
    out[i,] <- vapply(deps, length, integer(1))
  }
  colnames(out) <- pkgs
  cbind(date=dates, as.data.frame(out))
}

plot_revdeps <- function(data){
  df <- tidyr::pivot_longer(data, cols = names(data)[-1], names_to = 'package', values_to = 'count')
  ggplot(df, aes(x=date, y=count, group=package, color=package)) +
    geom_line() +
    geom_point() +
    ggtitle("Popularity of libcurl CRAN packages over time") +
    theme_ipsum() +
    scale_color_ipsum() +
    ylab("Reverse dependencies")
}

pkgs <- c("curl", 'RCurl')
revdeps <- count_revdeps(pkgs)
plot_revdeps(revdeps)