Hi, all I'm trying to make Nutch support Chinese and got a funny issue: the crawler printed out following log infomation: Indexing [http://sc.yfsz.com/cat.asp?catid=120] with analyzer [EMAIL PROTECTED] (it) Indexing [http://sc.yfsz.com/cat.asp?catid=126] with analyzer [EMAIL PROTECTED] (nl) Indexing [http://sc.yfsz.com/cat.asp?catid=130] with analyzer [EMAIL PROTECTED] (fi) Indexing [http://sc.yfsz.com/cat.asp?catid=128] with analyzer [EMAIL PROTECTED] (en) Indexing [http://sc.yfsz.com/help.asp?action=fukuan] with analyzer [EMAIL PROTECTED] (zh) ... Actually, all these weg page are in Chinese, so doc.get("lang") should always return "zh" in Indexer.java, line 91. But you can see it somtimes returns "it", "fi", "en" and so on, then the build-in analyzer is called. Is it a bug of language-identifier or NekoHtml? songjue 2007-04-16
