update r4ds TOC

bin/r4ds-toc.R

@@ -1,10 +1,10 @@
-#' extract the table of contents from R for Data Science and Save
+#' extract the table of contents from R for Data Science and save to a file
 library("rvest")
 library("purrr")
 library("stringr")
 library("jsonlite")
 
-URL <- "https://r4ds.had.co.nz/"
+R4DS_INDEX <- "https://r4ds.had.co.nz/"
 
 process_chapter <- function(x) {
   list(number = html_attr(x, "data-level"),
@@ -13,13 +13,44 @@ process_chapter <- function(x) {
                                     trim = TRUE), "^\\d+(\\.\\d+)*\\s+", ""))
 }
 
-r4ds_toc <- function() {
-  index <- read_html(URL)
+r4ds_chapters <- function() {
+  index <- read_html(R4DS_INDEX)
   book_summary <- html_nodes(index, "div.book-summary")
-  chapters <- html_nodes(book_summary, "li.chapter")
-  map(chapters, process_chapter)
+  chapters <- html_nodes(book_summary, "li.chapter") %>%
+    keep(~ str_detect(html_attr(.x, "data-level"), "^\\d+$")) %>%
+    map_dfr(~tibble(path = html_attr(.x, "data-path"),
+                    chapter = html_attr(.x, "data-level")))
+  chapters
 }
 
-toc <- r4ds_toc()
+process_section <- function(section) {
+  header <- html_children(section)[[1]]
+  lvl <- str_split(html_attr(section, "class"), " ")[[1]] %>%
+    str_subset("^level(\\d+)$") %>%
+    str_extract("\\d+")
+  tibble(
+    section_level = lvl,
+    section_id = html_attr(section, "id"),
+    section_number = html_text(html_node(header, "span.header-section-number")),
+    section_name = str_replace(html_text(header), "^[\\d.]+\\s+", "")
+  )
+}
+
+process_path <- function(path) {
+  doc <- read_html(str_c(R4DS_INDEX, path))
+  map_dfr(html_nodes(doc, "div.section"), process_section) %>%
+    mutate(path = path)
+}
+
+create_toc <- function() {
+  chapters <- r4ds_chapters()
+  map_dfr(chapters$path, process_path) %>%
+    mutate(url = str_c(R4DS_INDEX, path, "#", section_id))
+}
+
+main <- function() {
+  toc <- create_toc()
+  write_csv(toc, "r4ds-toc.csv")
+}
 
-write_json(toc, "r4ds-toc.json")
+main()