searchenginewatch.com~2

来自「垂直搜索的网络爬虫」· COM~2 代码 · 共 11 行

define pattern digit="[0-9]+"
do (a)h1.$l("title",(a))
do (a)-title-.-.-.-.-.-.-.-.-.(b)-$until[-,$startwith"Search Headlines"](c);(d)-(c).$l("text",(b),(d));(e)-title.$l("_new",(e),(d))
do (a)-title-.-.-.-.-.-.-.-.-.(b)-$until[-,$startwith"Search Engine Forums Spotlight"](c);(d)-(c).$l("text",(b),(d));(e)-title.$l("_new",(e),(d))
do (a)-title-.-.-.-.-.-.-.-.-.(b)-$until[-,$startwith"About The Search Engine Report"](c);(d)-(c).$l("text",(b),(d));(e)-title.$l("_new",(e),(d))
do (a)-title-.-.-.-.-.-.-.-.-.(b)-$until[-,$startwith"SearchDay Archives"](c);(d)-(c).$l("text",(b),(d));(e)-title.$l("_new",(e),(d))
do (a)-title-.-.-.-.-.-.-.-.-.(b)-$until[-,$startwith"Related Articles"](c);(d)-(c).$l("text",(b),(d));(e)-title.$l("_new",(e),(d))
do (a).$to.text.(b).(c)link(d).$l("li",(d))$l("_li",(a),(d))
set article=_new[title,text,link?=_li.li]
output article