R Code for Reading HTML Tables with XML

test.R

require(XML)

pg1 = 'http://www.wxtj.gov.cn/tjxx/tjsj/ydzyzb/index.shtml'
pg2 = 'http://www.wxtj.gov.cn/tjxx/tjsj/ydzyzb/index_2.shtml'
pg3 = 'http://www.wxtj.gov.cn/tjxx/tjsj/ydzyzb/index_3.shtml'
pg4 = 'http://www.wxtj.gov.cn/tjxx/tjsj/ydzyzb/index_4.shtml'

url1 = htmlTreeParse(pg1, useInternal = TRUE)
url2 = htmlTreeParse(pg2, useInternal = TRUE)
url3 = htmlTreeParse(pg3, useInternal = TRUE)
url4 = htmlTreeParse(pg4, useInternal = TRUE)

urls1 = unlist(xpathApply(url1, path = "//tr//td[@class='newstitle']//a", xmlGetAttr, "href"))
urls2 = unlist(xpathApply(url2, path = "//tr//td[@class='newstitle']//a", xmlGetAttr, "href"))
urls3 = unlist(xpathApply(url3, path = "//tr//td[@class='newstitle']//a", xmlGetAttr, "href"))
urls4 = unlist(xpathApply(url4, path = "//tr//td[@class='newstitle']//a", xmlGetAttr, "href"))

title1 = unlist(xpathApply(url1, path = "//tr//td[@class='newstitle']//a", xmlGetAttr, "title"))
title2 = unlist(xpathApply(url2, path = "//tr//td[@class='newstitle']//a", xmlGetAttr, "title"))
title3 = unlist(xpathApply(url3, path = "//tr//td[@class='newstitle']//a", xmlGetAttr, "title"))
title4 = unlist(xpathApply(url4, path = "//tr//td[@class='newstitle']//a", xmlGetAttr, "title"))

urls = paste('http://www.wxtj.gov.cn', c(urls1, urls2, urls3, urls4), sep = '')
title = c(title1, title2, title3, title4)

year = substr(title, 1, 4)
tmp = sub('^.*年([^ ]+)月.*$', '\\1', title)
tmp = sub('^.*1－([^ ]+).*$', '\\1', tmp)
tmp = sub('^.*1―([^ ]+).*$', '\\1', tmp)
tmp = sub('^.*1-([^ ]+).*$', '\\1', tmp)
month = replace(tmp, which(nchar(tmp) == 1), 
                paste('0', tmp[which(nchar(tmp) == 1)], sep = ''))
name = paste('data', year, month, sep = '_')

for (i in 1:length(urls)) eval(parse(text = paste(name[i], "= readHTMLTable('", urls[i], "')[[1]]", sep = "")))

还真是只能在*nix下用 windows下汉字是乱码