parse from comment

JaseZiv · Jan 17, 2024 · 912875a · 912875a
1 parent dbd52b2
commit 912875a
Show file tree

Hide file tree

Showing 2 changed files with 38 additions and 32 deletions.
diff --git a/R/chromote-fbref.R b/R/chromote-fbref.R
@@ -25,10 +25,10 @@ WorldfootballRDynamicPage <- R6::R6Class("WorldfootballRDynamicPage", public = l
  unlist(self$session$DOM$querySelectorAll(self$root_id, css)$nodeIds)
  },
 
- call_node_method = function(node_id) {
- js_fun <- paste0("function() { return this.outerHTML}")
+ call_node_method = function(node_id, method, ...) {
+ js_fun <- paste0("function() { return this", method, "}")
  obj_id <- self$object_id(node_id)
- self$session$Runtime$callFunctionOn(js_fun, objectId = obj_id)
+ self$session$Runtime$callFunctionOn(js_fun, objectId = obj_id, ...)
  },
 
  object_id = function(node_id) {
@@ -41,14 +41,23 @@ WorldfootballRDynamicPage <- R6::R6Class("WorldfootballRDynamicPage", public = l
 #' @importFrom purrr map_chr
 #' @importFrom xml2 xml_children read_html
 #' @noRd
-worldfootballr_html_page <- function(x) {
- stopifnot(identical(class(x), c("WorldfootballRDynamicPage", "R6")))
- nodes <- x$find_nodes("table")
-
- elements <- purrr::map_chr(nodes, function(node_id) {
- json <- x$call_node_method(node_id)
- json$result$value
- })
+worldfootballr_html_player_table <- function(session) {
+ stopifnot(identical(class(session), c("WorldfootballRDynamicPage", "R6")))
+
+ ## find element "above" commented out table
+ node_id1 <- session$find_nodes("#stats_shooting_sh")
+ ## find element "below" commented out table
+ node_id2 <- session$find_nodes("#stats_shooting_control")
+ ## find commented out element in-between
+ node_id <- round((node_id1 + node_id2) / 2)
+
+ elements <- session$call_node_method(node_id, ".textContent")[['result']][['value']]
+ n_elements <- length(elements)
+ if (n_elements != 1) {
+ warning(sprintf("Did not find the expected number of tables on the page (3). Found %s.", n_elements))
+ return(NULL)
+ }
+
  html <- paste0("<html>", paste0(elements, collapse = "\n"), "</html>")
  xml2::read_html(html)
 }
diff --git a/R/fb_league_stats.R b/R/fb_league_stats.R
@@ -37,33 +37,30 @@
  .frequency_id = "fb_league_stats-player"
  )
  session <- worldfootballr_chromote_session(url)
- page <- worldfootballr_html_page(session)
+ player_table <- worldfootballr_html_player_table(session)
  session$session$close(wait_ = FALSE)
- elements <- xml2::xml_children(xml2::xml_children(page))
- tables <- rvest::html_table(elements)
-
- n_tables <- length(tables)
- if (n_tables != 3) {
- warning(sprintf("Did not find the expected number of tables on the page (3). Found %s.", n_tables))
+ if (is.null(player_table)) {
  return(tibble::tibble())
  }
- renamed_table <- .rename_fb_cols(tables[[3]])
- renamed_table <- renamed_table[renamed_table$Rk != "Rk", ]
- renamed_table <- .add_player_href(
- renamed_table,
- parent_element = elements[[3]],
+
+ player_table_elements <- xml2::xml_children(xml2::xml_children(player_table))
+ parsed_player_table <- rvest::html_table(player_table_elements)
+ renamed_player_table <- worldfootballR:::.rename_fb_cols(parsed_player_table[[1]])
+ renamed_table <- renamed_player_table[renamed_player_table$Rk != "Rk", ]
+ renamed_player_table <- worldfootballR:::.add_player_href(
+ renamed_player_table,
+ parent_element = player_table_elements,
  player_xpath = ".//tbody/tr/td[@data-stat='player']/a"
  )
- }
-
- suppressMessages(
- readr::type_convert(
- clean_table,
- guess_integer = TRUE,
- na = "",
- trim_ws = TRUE
+ suppressMessages(
+ readr::type_convert(
+ renamed_player_table,
+ guess_integer = TRUE,
+ na = "",
+ trim_ws = TRUE
+ )
  )
- )
+ }
 }