udn_crawler

UDN 新聞爬蟲

抓取新聞列表

library(rvest)

## Loading required package: xml2

domain <- 'http://udn.com'
res <-  read_html('http://udn.com/news/breaknews/1')
urls <-res %>% html_nodes('#breaknews_body a') %>% html_attr('href') %>% paste0(domain, x=.)

內容頁剖析器

parseUDN = function(ele){
  title = read_html(ele) %>% html_nodes('#story_art_title') %>% html_text() %>% iconv(from='UTF-8', to='UTF-8')
  meta = read_html(ele) %>% html_nodes('#story_bady_info') %>% html_text() %>% iconv(from='UTF-8', to='UTF-8')
  article = read_html(ele) %>% html_nodes('#story_body_content') %>% html_text() %>% iconv(from='UTF-8', to='UTF-8')
  data.frame(title = title, meta=meta, article = article)
}

抓取內容資訊

udndf <- lapply(urls, parseUDN)

合併抓取新聞資料

udn <- do.call("rbind", udndf)