ATP parser

parser_atp.R

# Title: Tennis Draw Parser
# Author: Nathan Esau

library(XML)

options(warn = -1) # suppress warnings

# Get all draws for current year

url = "https://www.atpworldtour.com/en/scores/results-archive?year=2017"

links <- xpathSApply(doc, "//a/@href")
links <- links[grepl("draws", links)] # doubles and singles
links <- links[grepl("doubles", links) == FALSE]
drawURLs = sapply(links, function(x){paste("https://www.atpworldtour.com", x, collapse='', sep='')})
names(drawURLs) = NULL

# url: the tournament url
getSeeds <- function(url) {
  
  tables = readHTMLTable(getURL(url))
  draw = tables[[3]]
  
  round1 = sapply(draw[,1], as.character) # get first round (to get initial bracket and seeds)
  rowLengths = sapply(round1, nchar)
  names(rowLengths) = NULL
  round1 = round1[rowLengths > 5]
  
  patterns = c('\\r', '\\t', '\\n', '[0-9]+', '\\(Q\\)', '\\(', '\\)', 'WC', 'PR')
  names = character(0) # parse names from table
  
  for(row in round1) {
    x = row
    for(pattern in patterns) {
      x = gsub(pattern, ' ', x)
    }
    
    words = unlist(strsplit(x, " "))
    wordsIndex = which(nchar(words) > 0)
    dist = diff(wordsIndex)
    splitIndex = which(dist > 5)
    words = words[nchar(words) > 0]
    
    name1 = paste(words[1:splitIndex], collapse = ' ')
    name2 = paste(words[(splitIndex+1):length(words)], collapse = ' ')
    names = c(names, name1, name2)
  }
  
  names
}

out = vector("list", length(drawURLs))

for(i in 1:length(drawURLs)) {
  url = drawURLs[i]
  out[[i]] = getSeeds(url)
}

out[[1]]