/**
* Initializes the scanner.
*
* @param f input file
* @param pr database properties
* @param frag allow parsing of document fragment
* @throws IOException I/O exception
*/
XMLScanner(final IO f, final Prop pr, final boolean frag) throws IOException {
input = new XMLInput(f);
fragment = frag;
try {
for (int e = 0; e < ENTITIES.length; e += 2) {
ents.add(token(ENTITIES[e]), token(ENTITIES[e + 1]));
}
dtd = pr.is(Prop.DTD);
chop = pr.is(Prop.CHOP);
String enc = null;
// process document declaration...
if (consume(DOCDECL)) {
if (s()) {
if (!version()) error(DECLSTART);
boolean s = s();
enc = encoding();
if (enc != null) {
if (!s) error(WSERROR);
s = s();
}
if (sddecl() != null && !s) error(WSERROR);
s();
final int ch = nextChar();
if (ch != '?' || nextChar() != '>') error(DECLWRONG);
} else {
prev(5);
}
}
encoding = enc == null ? UTF8 : enc;
if (!fragment) {
final int n = consume();
if (!s(n)) {
if (n != '<') error(BEFOREROOT);
prev(1);
}
}
} catch (final IOException ex) {
input.close();
throw ex;
}
}
/**
* Constructor.
*
* @param d data reference
* @throws IOException IOException
*/
FTBuilder(final Data d) throws IOException {
super(d);
final Prop prop = d.meta.prop;
final FTOpt fto = new FTOpt();
fto.set(FTFlag.DC, prop.is(Prop.DIACRITICS));
fto.set(FTFlag.CS, prop.is(Prop.CASESENS));
fto.set(FTFlag.ST, prop.is(Prop.STEMMING));
fto.sw = new StopWords(d, prop.get(Prop.STOPWORDS));
fto.ln = Language.get(prop);
if (!Tokenizer.supportFor(fto.ln)) throw new BaseXException(NO_TOKENIZER_X, fto.ln);
if (prop.is(Prop.STEMMING) && !Stemmer.supportFor(fto.ln))
throw new BaseXException(NO_STEMMER_X, fto.ln);
scm = d.meta.scoring;
max = -1;
min = Integer.MAX_VALUE;
lex = new FTLexer(fto);
}
