# Scrape MiLB game logs for pitchers from Fangraphs, combining 'standard' and 'advanced' tabs

milb_pitcher_game_logs_fg <- function(playerid, year = 2017) {
  
  # url for standard game log table
  url_basic <- paste0("http://www.fangraphs.com/statsd.aspx?playerid=",
                      playerid,
                      "&season=",
                      year,
                      "&position=P","&type=-1")
  
  # url for advanced game log table
  url_adv <- paste0("http://www.fangraphs.com/statsd.aspx?playerid=",
                    playerid,
                    "&season=",
                    year,
                    "&position=P","&type=-2")
  
  # standard table
  payload1 <- xml2::read_html(url_basic) %>%
    rvest::html_nodes("table") %>%
    .[length(.)] %>%
    rvest::html_table() %>%
    as.data.frame()
  
  payload1 <- payload1 %>%
    dplyr::filter(!grepl("Date|Total", Date)) 
  
  # advanced table
  payload2 <- xml2::read_html(url_adv) %>%
    rvest::html_nodes("table") %>%
    .[length(.)] %>%
    rvest::html_table() %>%
    as.data.frame()
  
  payload2 <- payload2 %>%
    dplyr::filter(!grepl("Date|Total", Date)) %>% 
    dplyr::rename(BB_perc = BB., K_perc = K.,
                  K_minus_BB = K.BB., LOB_perc = LOB.)
  
  payload2 <- as.data.frame(sapply(payload2, function(x) (gsub("\\ %", "", x))),
                            stringsAsFactors=F)
  
  payload2$BB_perc <- as.numeric(payload2$BB_perc)/100
  payload2$K_perc <- as.numeric(payload2$K_perc)/100
  payload2$K_minus_BB <- as.numeric(payload2$K_minus_BB)/100
  payload2$LOB_perc <- as.numeric(payload2$LOB_perc)/100
  
  # combine standard & advanced game log tables
  payload <- merge(payload1,payload2)
  
  # separate Team column into Team & MiLB level
  payload <- payload %>% 
    separate(Team, into = c("Team","Level"),sep=" ")
  
  payload
}