PenerarXML.groovy

REGEX = "^[\\w\\s/<>?\".=-]*[áéíóúçàãõâêô:\\s]*\$"

nome = "    jonatas#!@#%¨&* emidio"
texto = '''
<?xml¨&*()¨&*( version="1.0" encoding="ISO-8859-1"?>
<SAIDA>
    <HEADER>
        <ERRO>
            <CODIGO-ERRO>1454014207309</CODIGO-ERRO>%¨&*()
            <ORIGEM-ERRO>The data "ˆ" is not legal for a JDOM character content 0x0 is not a legal XML character.</ORIGEM-ERRO>
        </ERRO>
    </HEADER>
</SAIDA>
'''

xmlvalido0 = texto.collect { linha -> linha.collect{ if(it.matches(REGEX)) it else '' }.join() }.join()

println xmlvalido0

xmlvalido1 = texto.collect { linha -> 
    linha.collect{ 
        if(it.matches(REGEX)) 
            it 
        else 
            '' 
     }.join() 
}.join()

println xmlvalido1

StringBuilder xmlvalido2 = new StringBuilder();
for (String linha in texto.split("\n")){
    String[] caracteres = linha.split("");
    StringBuilder linhaValida = new StringBuilder();
    for (int i = 0; i < caracteres.length; i++){
        if(caracteres[i].matches(REGEX)){
            linhaValida.append(caracteres[i]);
        }
    }
    xmlvalido2.append(linhaValida + "\n");
}

println xmlvalido2





//def parseline(line){
//    sb = new StringBuffer()
//    line.collect{ if(it.matches(REGEX)) { sb << it } }
//    return sb.toString()
//}

//sb = new StringBuffer()
//nome.collect{ if(it.matches(REGEX)) { sb << it } }
//println sb.toString()

//println nome.collect{ if(it.matches(REGEX)) it else '' }.join()