Bresler7

getBREFTeamStatTable <- function(season_end = 2015, table_name = 'team', date = T){
    c('rvest','dplyr','pipeR') -> packages
    lapply(packages, library, character.only = T)
    'http://www.basketball-reference.com/leagues/' -> base
    'NBA' -> league
    '#' %>>% paste0(table_name) -> css_page
    css_page %>>% paste0(" , ", css_page,' a') -> css_id
    table_name %>>% tolower -> table_name
    table_name %>>% paste('stats', sep = "_") -> table
    base %>>% paste0(league,'_',season_end,".html") -> url
    url %>>% ## get table
        html %>>%
        html_nodes(css_page) %>>%
        html_table(header = F) %>>% data.frame() %>>% tbl_df() -> df

    if(df$X.1[1] == 'Rk'){
        df %>>%
            filter(X.1 == "Rk") %>>% as.character -> names
        'Rk' %>>% grep(x = df$X.1) -> row_of_header #находим ранг
        (row_of_header + 1) %>>% (df[.:nrow(df),]) -> df #пропускаем этот ряд и идем до конца
        names %>>% tolower-> names(df)} else{
            df %>>%
                filter(X.1 == "Rk") %>>% as.character -> names
            'Rk' %>>% grep(x = df$X.1) -> row_of_header #находим ранг
            (row_of_header + 1) %>>% (df[.:nrow(df),]) -> df #пропускаем этот ряд и идем до конца
            names %>>% tolower-> names(df)
        }
    names(df) %>>% (gsub('\\%|/','\\.',.)) -> names(df)
    NULL -> df$rk
    c('team','arena') -> table_name_character
    df[,!(df %>>% names) %in% table_name_character] %>>%
        apply(2, function(x) gsub('\\,','',x) %>>% as.numeric(x))  ->
        df[,!(df %>>% names) %in% table_name_character] #убираем запятые, переводим в числовой формат
    df$team %>>% grepl(pattern = '\\*') -> df$playoff_team
    df$team %>>% (gsub('\\*','',.)) -> df$team
    df %>>% nrow() -1  -> rows
    df[1:rows,] -> df
    (season_end-1) %>>% paste0("-",season_end) -> season
    ##Собираем ID команд
    url %>>% ## get table
        html %>>%
        html_nodes(css_id) %>>%
        html_attrs() %>>% unlist %>>% as.character -> stems
    stems[3:length(stems)] -> stems #пропускаем первые два ряда, потому что там заголовки
    stems %>>% (gsub('\\/|.html|teams','',.)) %>>% #убираем ненужный текст
        (gsub(season_end,'',.)) -> bref_team_id #убираем год, получаем id команды
    data.frame(season,table_name = table, bref_team_id, df) -> df  #объединяем в 1 df
    if(date == T){
        Sys.time() -> df$scrape_time #add scrape time if you want it
    }
    return(df)
}

#This paste is a code from tutorial by Alex Bresler
# http://asbcllc.com/blog/2014/november/creating_bref_scraper/
# Comments translated into Russian for
# http://www.datadrivenjournalism.ru/2015/03/webscrape-in-r/