01_scrapeData.Rmd

---
title: "01_scrape athletes olympics"
author: "Duc-Quang Nguyen"
date: "2 Aug 2016"
output: html_document
---


```{r setup, include=FALSE}
library(readr)
library(tidyr)
library(dplyr)
library(magrittr)
library(countrycode)
### Getting data in packages
library(rvest)


# hack for some IOC country abbrevations 

ioc_country <- structure(
  c("ROT", "KOS", "IVB", "SSD", "IOA"), 
  names = c("Refugee Olympic Team", "Kosovo", "British Virgin Islands", "South Sudan", "Individual Olympic Athletes")
)

```


```{r scrape}
base.url <- "https://www.rio2016.com/en/"

# Get all the sports and their URLs
sports <- read_html("https://www.rio2016.com/en/sports") %>% 
  html_nodes(".olympic-pictograms__item") %>% html_nodes("a")

sports.suffix <- gsub('<a href="/en/(.*)" class.*', "\\1", sports)


athletes.list <- lapply(sports.suffix, function(sport) {
  cat("\n", sport)
  sport.url <- paste0(base.url, sport)
  
  name <- read_html(sport.url) %>%
    html_nodes(".athletes-teams-graphic__full-list-name") %>% html_text()
  gender <- read_html(sport.url) %>%
    html_nodes(".athletes-teams-graphic__full-list-gender") %>% html_text()
  iso3c <- read_html(sport.url) %>%
    html_nodes(".athletes-teams-graphic__full-list-country")  %>% html_text()
  stopifnot(length(name) == length(iso3c), length(gender) == length(name))
  
  if(length(name) == 0) {
    warning("\n\n", "No althletes list for ", sport, "!!!", "\n")
    NULL
  } else {
    data.frame(athletes = name, gender = gender, iso3 = iso3c, sport =  sport, stringsAsFactors = F)     
  }
})

athletes <- do.call(rbind, athletes.list)
athletes$country <- countrycode(athletes$iso3, "ioc", "country.name")

# Unmatched IOC nation abbrevations
idx <- match(athletes[which(is.na(athletes$country)),'iso3'], ioc_country)
if(any(is.na(idx))) {
  warning("\nSome country IOC abbrevations have no English name!\n")
}
athletes[which(is.na(athletes$country)),'country'] <- names(ioc_country)[idx]


write.csv(athletes, "input/athletes_rio2016.csv", row.names = F)
```