import grails.test.*

import groovyx.net.http.HTTPBuilder
import static groovyx.net.http.Method.GET
import static groovyx.net.http.Method.POST
import static groovyx.net.http.ContentType.TEXT
import static groovyx.net.http.ContentType.HTML

class ArticleTests extends GrailsUnitTestCase {
    protected void setUp() {
        super.setUp()
    }

    protected void tearDown() {
        super.tearDown()
    }

    // ?? ?? ?? 
    void testParsingArticle() {
    	String articleUrl = 'http://news.nate.com/view/20091203n08304'
		def url = new URL(articleUrl)
    	def articleReader = new HTTPBuilder(url.getProtocol() + '://' + url.getHost()).request(GET,TEXT) { 
    		uri.path = url.getPath()
		}
        assert articleReader instanceof Reader // response data is buffered in-memory
        
		def slurper = new XmlSlurper(new org.ccil.cowan.tagsoup.Parser())
        
    	def html = slurper.parse(articleReader)
		def article = html.body.'**'.find{ it.name() == 'div' && it.@id == 'articleView' }
    	
		def info = article.p.find{ it.@class == 'articleInfo' }
    	def dateCreated
    	def dateModified
		try {
			dateCreated = Date.parse('yyyy-MM-dd hh:mm', info.span.find{it.@class == 'firstDate'}.em.text())
		} catch(java.text.ParseException e) {}
		try {
			dateModified = Date.parse('yyyy-MM-dd hh:mm', info.span.find{it.@class == 'lastDate'}.em.text())
		} catch(java.text.ParseException e) {}
		
		def contentNode = article.div.find{ it.@id == 'articleContetns' }.table.tr.td
		contentNode.'**'.findAll{ it.name() == 'br' }.each{ br -> br.replaceBody('\r\n') }
		contentNode.'**'.findAll{ it.name() == 'img' }.each{ img -> img.replaceBody("<img src='${img.@src.text()}'>") }
		
		contentNode.'**'.findAll{ it.name() == 'script' }.each { script -> script.replaceBody('') }
		
		String title = article.h3
		String imageLink = contentNode.div.span.img.it.@src.text()
		String thumbnailLink = imageLink.replace('orgImg', 'thumb90')
		String content = contentNode.text().trim()
		String articleId = url.getPath().replace('/view/', '')
		String media = info.span.a[0]
		
		assert title
		assert content
		assert articleId
		assert info
		assert media
		assert dateCreated
		
		println "$title($articleId, $dateCreated, $dateModified, $media)"
		println "$imageLink, $thumbnailLink"
		println "$content"
    }
    
    void testArticleClass() {
    	// ?? ??
		String articleUrl = 'http://news.nate.com/view/20091203n08304'
		def article = Article.parseArticle(articleUrl)
    	
		println "$article.title($article.articleId, $article.url, $article.dateCreated, $article.dateModified, $article.media)"
		println "$article.imageLink, $article.thumbnailLink"
		println "$article.content"
    }
}