fetch title from URL

TitleUtil

@Grab(group='nekohtml', module='nekohtml', version='1.9.6')

import org.cyberneko.html.parsers.SAXParser

import java.util.regex.Pattern


// URLからタイトルを取得する.

class TitleUtil {

	String getTitle(URL url)
	throws FileNotFoundException,Exception{
			getTitleFromUrl(url)
	}

	static def getEncode(URL url)
	throws FileNotFoundException,Exception{

		// 文字コード判別
		def encode = null
	
		def r = new BufferedReader(new InputStreamReader(url.openStream()) )
		while( true ){
			def line = r.readLine()
			if( line == null ){
				break
			}
			if( (line =~ /meta/ || line =~ /META/)  && line =~ /charset/ ){
				//println line
				def pat = java.util.regex.Pattern.compile('charset=(.*)"')
				def m   = pat.matcher(line)
				if( m.find() ){
					encode = m.group(1)
				}
				break
			}
		}
		r.close()
	
		encode
	}

	// 取得できなかったときは null 値がかえる
	static def getTitleFromUrl(URL url)
	throws FileNotFoundException,Exception{

		// 1) get encode
		def enc = getEncode(url)
		if( enc==null ){ enc='UTF-8' }
		else {
			// ここにJavaでサポートされているエンコード名を列挙
			//def encode_list = ['UTF-8','SHIFT_JIS','']

			if( enc.toUpperCase().trim() =~ /UTF-8/ ){
				enc = 'UTF-8'
			}
		}

		// 2) get title
		def t = null

		try{ 
			def r = new BufferedReader(new InputStreamReader(url.openStream(),enc) )
			new XmlSlurper(new SAXParser()).parse(r).'**'.findAll{
				if( it.name() == 'TITLE' ){
					//t = it.toString()

					def sb = ''<<''
					new StringReader(it.toString()).each{
						sb << it
					}
					t = sb.toString()
				}
			}
			r.close()
	
			t.trim()
		}
		catch(Exception ex){
		}

		t
	}
}