static {
IOFile dic = null;
if (Reflect.available(PATTERN)) {
dic = new IOFile(LANG);
if (!dic.exists()) {
dic = new IOFile(Prop.HOME, "etc/" + LANG);
if (!dic.exists()) {
available = false;
}
}
} else {
available = false;
}
if (available) {
Class<?> clz = Reflect.find(PATTERN);
if (clz == null) {
Util.debug("Could not initialize Igo Japanese lexer.");
} else {
/* Igo constructor. */
final Constructor<?> tgr = Reflect.find(clz, String.class);
tagger = Reflect.get(tgr, dic.path());
if (tagger == null) {
available = false;
Util.debug("Could not initialize Igo Japanese lexer.");
} else {
parse = Reflect.method(clz, "parse", CharSequence.class);
if (parse == null) {
Util.debug("Could not initialize Igo lexer method.");
}
clz = Reflect.find("net.reduls.igo.Morpheme");
surface = Reflect.field(clz, "surface");
feature = Reflect.field(clz, "feature");
start = Reflect.field(clz, "start");
}
}
}
}
/**
* Converts an HTML document to XML.
*
* @param io io reference
* @param opts html options
* @return parser
* @throws IOException I/O exception
*/
private static IO toXML(final IO io, final HtmlOptions opts) throws IOException {
// reader could not be initialized; fall back to XML
if (READER == null) return io;
try {
// tries to extract the encoding from the input
final TextInput ti = new TextInput(io);
String enc = ti.encoding();
final byte[] content = ti.content();
// looks for a charset definition
final byte[] encoding = token("charset=");
int cs = indexOf(content, encoding);
if (cs > 0) {
// extracts the encoding string
cs += encoding.length;
int ce = cs;
final int cl = content.length;
while (++ce < cl && content[ce] > 0x28) ;
enc = string(substring(content, cs, ce));
}
// define input
final InputSource is = new InputSource(new ArrayInput(content));
is.setEncoding(supported(enc) ? normEncoding(enc) : UTF8);
// define output
final StringWriter sw = new StringWriter();
final XMLReader reader = (XMLReader) Reflect.get(READER);
final Object writer = Reflect.get(WRITER, sw);
// set TagSoup options
if (opts.get(HtmlOptions.HTML)) {
reader.setFeature("http://xml.org/sax/features/namespaces", false);
opt("method", "html");
opt("omit-xml-declaration", "yes");
}
if (opts.get(HtmlOptions.NONS))
reader.setFeature("http://xml.org/sax/features/namespaces", false);
if (opts.get(HtmlOptions.OMITXML)) opt("omit-xml-declaration", "yes");
if (opts.get(HtmlOptions.NOBOGONS)) reader.setFeature(FEATURES + "ignore-bogons", true);
if (opts.get(HtmlOptions.NODEFAULTS))
reader.setFeature(FEATURES + "default-attributes", false);
if (opts.get(HtmlOptions.NOCOLONS)) reader.setFeature(FEATURES + "translate-colons", true);
if (opts.get(HtmlOptions.NORESTART)) reader.setFeature(FEATURES + "restart-elements", false);
if (opts.get(HtmlOptions.IGNORABLE))
reader.setFeature(FEATURES + "ignorable-whitespace", true);
if (opts.get(HtmlOptions.EMPTYBOGONS)) reader.setFeature(FEATURES + "bogons-empty", true);
if (opts.get(HtmlOptions.ANY)) reader.setFeature(FEATURES + "bogons-empty", false);
if (opts.get(HtmlOptions.NOROOTBOGONS)) reader.setFeature(FEATURES + "root-bogons", false);
if (opts.get(HtmlOptions.NOCDATA)) reader.setFeature(FEATURES + "cdata-elements", false);
if (opts.get(HtmlOptions.LEXICAL))
reader.setProperty("http://xml.org/sax/properties/lexical-handler", writer);
if (opts.contains(HtmlOptions.METHOD)) opt("method", opts.get(HtmlOptions.METHOD));
if (opts.contains(HtmlOptions.DOCTYPESYS))
opt("doctype-system", opts.get(HtmlOptions.DOCTYPESYS));
if (opts.contains(HtmlOptions.DOCTYPEPUB))
opt("doctype-public", opts.get(HtmlOptions.DOCTYPEPUB));
if (opts.contains(HtmlOptions.ENCODING)) is.setEncoding(opts.get(HtmlOptions.ENCODING));
// end TagSoup options
reader.setContentHandler((ContentHandler) writer);
reader.parse(is);
return new IOContent(token(sw.toString()), io.name());
} catch (final SAXException ex) {
Util.errln(ex);
return io;
}
}
