address #344 and drive by tm_player_bio fix

DESCRIPTION

@@ -1,7 +1,7 @@
 Type: Package
 Package: worldfootballR
 Title: Extract and Clean World Football (Soccer) Data
-Version: 0.6.4.0013
+Version: 0.6.4.0014
 Authors@R: c(
  person("Jason", "Zivkovic", , "[email protected]", role = c("aut", "cre", "cph")),
  person("Tony", "ElHabr", , "[email protected]", role = "ctb"),

NEWS.md

@@ -8,8 +8,10 @@
 * `fb_league_stats(team_or_player = "player")` returning duplicate player hrefs (0.6.4.0008) [#331](https://github.com/JaseZiv/worldfootballR/issues/331)
 * `fb_league_stats(team_or_player = "player")` returning wrong season's data for Australian league (0.6.4.0009) [#333](https://github.com/JaseZiv/worldfootballR/issues/333)
 * `tm_player_market_values()` returing the player name and valuation on separate rows in the `player_name` column [#338] (https://github.com/JaseZiv/worldfootballR/issues/338) and also returning `NA`s for the `player_age` column (0.6.4.0010) [#336](https://github.com/JaseZiv/worldfootballR/issues/336)
-* `fb_match_results()` returns `NA` goals due to inconsistent `iconv()` behvaior on different systems (0.6.4.0011) [#326](https://github.com/JaseZiv/worldfootballR/issues/326)
+* `fb_match_results()` returns `NA` goals due to inconsistent `iconv()` behavior on different systems (0.6.4.0011) [#326](https://github.com/JaseZiv/worldfootballR/issues/326)
 * `tm_team_player_urls()` fixed after change to server-side loading (0.6.4.0013) [#342](https://github.com/JaseZiv/worldfootballR/issues/342)
+* `fb_teams_urls()` fixed to remove lower division teams being returned as a result of playoff promotion games (0.6.4.0014) [#344](https://github.com/JaseZiv/worldfootballR/issues/344)
+* `tm_player_bio()` column name and data structure change for player date of birth (0.6.4.0014)
 
 
 ### Improvements

R/tm_player_bio.R

@@ -85,8 +85,13 @@ tm_player_bio <- function(player_urls) {
 
  # some of the following columns may not exist, so this series of if statements will handle for this:
  if(any(grepl("date_of_birth", colnames(full_bios)))) {
+ # tm changed the column name from date_of_birth to date_of_birth_age - need to clean up the age in parentheses
+ c_name <- names(full_bios[grepl("date_of_birth", colnames(full_bios))])
  full_bios <- full_bios %>%
- dplyr::mutate(date_of_birth = .tm_fix_dates(.data[["date_of_birth"]]))
+ dplyr::mutate(date_of_birth = .tm_fix_dates(gsub(" \\(.*", "", .data[[c_name]])))
+
+ # to not introduce breaking changes, we will keep the column name consistent with existing expected output and remove the new date_of_birth_age column
+ full_bios[, c_name] <- NULL
  }
 
  if(any(grepl("joined", colnames(full_bios)))) {

R/worldfootballr_helpers.R

@@ -77,6 +77,7 @@ fb_teams_urls <- function(league_url, time_pause=3) {
  main_url <- "https://fbref.com"
 
  urls <- league_season_page %>%
+ rvest::html_elements(".stats_table") %>%
  rvest::html_nodes("img+ a") %>%
  rvest::html_attr("href") %>%
  unique() %>% paste0(main_url, .)
@@ -201,16 +202,16 @@ tm_league_team_urls <- function(country_name, start_year, league_url = NA) {
 #' @export
 #'
 tm_team_player_urls <- function(team_url) {
- 
+
  main_url <- "https://www.transfermarkt.com"
- 
+
  tryCatch({team_page <- xml2::read_html(team_url)}, error = function(e) {team_page <- c()})
- 
+
  player_urls <- team_page %>%
  rvest::html_nodes("#yw1 td.posrela .hauptlink a") %>% rvest::html_attr("href") %>%
  unique() %>%
  paste0(main_url, .)
- 
+
  return(player_urls)
 }