@Override
public String toString() {
final TokenBuilder tb = new TokenBuilder().add(var.toString()).add(' ').add(ASSIGN);
tb.add(' ').add(expr.toString());
if (coll != null) tb.add(' ').add(COLLATION).add(" \"").add(coll.uri()).add('"');
return tb.toString();
}
/**
* Scans an entity value. [9]
*
* @param p pe reference flag
* @return value
* @throws IOException I/O exception
*/
private byte[] entityValue(final boolean p) throws IOException {
final int qu = consume();
if (qu != '\'' && qu != '"') {
prev(1);
return null;
}
TokenBuilder tok = new TokenBuilder();
int ch;
while ((ch = nextChar()) != qu) {
if (ch == '&') tok.add(ref(false));
else if (ch == '%') {
if (!p) error(INVPE);
tok.add(peRef());
} else {
tok.add(ch);
}
}
final XMLInput tmp = input;
input = new XMLInput(new IOContent(tok.finish()));
tok = new TokenBuilder();
while ((ch = consume()) != 0) {
if (ch == '&') tok.add(ref(false));
else tok.add(ch);
}
input = tmp;
return tok.finish();
}
/**
* Joins the path.
*
* @param s segment to start with
* @return joined path
*/
private String join(final int s) {
final TokenBuilder tb = new TokenBuilder();
for (int p = s; p < segments.length; p++) {
if (!tb.isEmpty()) tb.add('/');
tb.add(segments[p]);
}
return tb.toString();
}
@Override
public String toString() {
final TokenBuilder tb = new TokenBuilder(Util.className(this)).add('[');
for (int i = 1; i < size; i++) {
tb.add(Integer.toString(keys[i])).add(": ").addExt(get(keys[i]));
if (i < size - 1) tb.add(",\n\t");
}
return tb.add(']').toString();
}
/**
* Returns the database path (i.e., all path entries except for the first).
*
* @return path depth
*/
public String dbpath() {
final TokenBuilder tb = new TokenBuilder();
final int ps = segments.length;
for (int p = 1; p < ps; p++) {
if (!tb.isEmpty()) tb.add('/');
tb.add(segments[p]);
}
return tb.toString();
}
@Override
public final synchronized String toString() {
final TokenBuilder tb = new TokenBuilder();
for (final Entry<String, Object> e : props.entrySet()) {
if (!tb.isEmpty()) tb.add(',');
tb.add(e.getKey()).add('=').addExt(e.getValue());
}
return tb.toString();
}
@Override
public String toString() {
final int es = expr.length;
final TokenBuilder tb = new TokenBuilder(expr[es - 1].toString()).add('(');
for (int e = 0; e < es - 1; e++) {
tb.add(expr[e].toString());
if (e < es - 2) tb.add(", ");
}
return tb.add(')').toString();
}
/**
* Adds the time zone to the specified token builder.
*
* @param tb token builder
*/
void zone(final TokenBuilder tb) {
if (tz == Short.MAX_VALUE) return;
if (tz == 0) {
tb.add('Z');
} else {
tb.add(tz > 0 ? '+' : '-');
prefix(tb, Math.abs(tz) / 60, 2);
tb.add(':');
prefix(tb, Math.abs(tz) % 60, 2);
}
}
/**
* Scans CDATA.
*
* @throws IOException I/O exception
*/
private void cDATA() throws IOException {
int ch;
while (true) {
while ((ch = nextChar()) != ']') token.add(ch);
if (consume(']')) {
if (consume('>')) return;
prev(1);
}
token.add(ch);
}
}
@Override
public String getMessage() {
final TokenBuilder tb = new TokenBuilder();
if (info != null) tb.add(STOPPED_AT).add(info.toString()).add(COL).add(NL);
final byte[] code = name.local();
if (code.length != 0) tb.add('[').add(name.prefixId(QueryText.ERROR_URI)).add("] ");
tb.add(getLocalizedMessage());
if (!stack.isEmpty()) {
tb.add(NL).add(NL).add(STACK_TRACE).add(COL);
for (final InputInfo ii : stack) tb.add(NL).add(LI).add(ii.toString());
}
return tb.toString();
}
/**
* Scans a reference. [67]
*
* @param f dissolve entities
* @return entity
* @throws IOException I/O exception
*/
private byte[] ref(final boolean f) throws IOException {
// scans numeric entities
if (consume('#')) { // [66]
final TokenBuilder ent = new TokenBuilder();
int b = 10;
int ch = nextChar();
ent.add(ch);
if (ch == 'x') {
b = 16;
ent.add(ch = nextChar());
}
int n = 0;
do {
final boolean m = ch >= '0' && ch <= '9';
final boolean h = b == 16 && (ch >= 'a' && ch <= 'f' || ch >= 'A' && ch <= 'F');
if (!m && !h) {
completeRef(ent);
return QUESTION;
}
n *= b;
n += ch & 15;
if (!m) n += 9;
ent.add(ch = nextChar());
} while (ch != ';');
if (!valid(n)) return QUESTION;
ent.reset();
ent.add(n);
return ent.finish();
}
// scans predefined entities [68]
final byte[] name = name(false);
if (!consume(';')) return QUESTION;
if (!f) return concat(AMPER, name, SEMI);
byte[] en = ents.get(name);
if (en == null) {
// unknown entity: try HTML entities (lazy initialization)
if (HTMLENTS.size() == 0) {
for (int s = 0; s < HTMLENTITIES.length; s += 2) {
HTMLENTS.add(token(HTMLENTITIES[s]), token(HTMLENTITIES[s + 1]));
}
}
en = HTMLENTS.get(name);
}
return en == null ? QUESTION : en;
}
@Override
public byte[] info(final MainOptions options) {
final TokenBuilder tb = new TokenBuilder();
tb.add(LI_STRUCTURE).add(HASH).add(NL);
tb.add(LI_NAMES).add(data.meta.names(type)).add(NL);
final IndexStats stats = new IndexStats(options.get(MainOptions.MAXSTAT));
final int s = values.size();
for (int p = 1; p <= s; p++) {
final int oc = lenList.get(p);
if (oc > 0 && stats.adding(oc)) stats.add(values.key(p), oc);
}
stats.print(tb);
return tb.finish();
}
/**
* Returns an atomized content for any node kind. The atomized value can be an attribute value or
* XML content.
*
* @param pre pre value
* @return atomized value
*/
public final byte[] atom(final int pre) {
switch (kind(pre)) {
case TEXT:
case COMM:
return text(pre, true);
case ATTR:
return text(pre, false);
case PI:
byte[] txt = text(pre, true);
final int i = indexOf(txt, ' ');
return i == -1 ? EMPTY : substring(txt, i + 1);
default:
// create atomized text node
TokenBuilder tb = null;
byte[] t = EMPTY;
int p = pre;
final int s = p + size(p, kind(p));
while (p != s) {
final int k = kind(p);
if (k == TEXT) {
txt = text(p, true);
if (t == EMPTY) {
t = txt;
} else {
if (tb == null) tb = new TokenBuilder(t);
tb.add(txt);
}
}
p += attSize(p, k);
}
return tb == null ? t : tb.finish();
}
}
/**
* Adds some characters to the entity.
*
* @param ent token builder
* @throws IOException I/O exception
*/
private void completeRef(final TokenBuilder ent) throws IOException {
int ch = consume();
while (ent.size() < 10 && ch >= ' ' && ch != ';') {
ent.add(ch);
ch = consume();
}
}
/**
* Scans a processing instruction.
*
* @throws IOException I/O exception
*/
private void pi() throws IOException {
final byte[] tok = name(true);
if (eq(lc(tok), XML)) error(PIRES);
token.add(tok);
int ch = nextChar();
if (ch != '?' && !ws(ch)) error(PITEXT);
do {
while (ch != '?') {
token.add(ch);
ch = nextChar();
}
if ((ch = consume()) == '>') return;
token.add('?');
} while (true);
}
/**
* Consumes an Nmtoken. [7]
*
* @throws IOException I/O exception
*/
private void nmtoken() throws IOException {
final TokenBuilder name = new TokenBuilder();
int c;
while (isChar(c = nextChar())) name.add(c);
prev(1);
if (name.isEmpty()) error(INVNAME);
}
/**
* Scans an attribute value. [10]
*
* @param ch current character
* @throws IOException I/O exception
*/
private void attValue(final int ch) throws IOException {
boolean wrong = false;
int c = ch;
do {
if (c == 0) error(ATTCLOSE, (char) c);
wrong |= c == '\'' || c == '"';
if (c == '<') error(wrong ? ATTCLOSE : ATTCHAR, (char) c);
if (c == 0x0A) c = ' ';
if (c == '&') {
// verify...
final byte[] r = ref(true);
if (r.length == 1) token.add(r);
else if (!input.add(r, false)) error(RECENT);
} else {
token.add(c);
}
} while ((c = consume()) != quote);
}
/**
* Replaces the specified character and returns the result token.
*
* @param token token to be checked
* @param search the character to be replaced
* @param replace the new character
* @return resulting token
*/
public static byte[] replace(final byte[] token, final int search, final int replace) {
final TokenBuilder tb = new TokenBuilder(token.length);
final int tl = token.length;
for (int i = 0; i < tl; i += cl(token, i)) {
final int c = cp(token, i);
tb.add(c == search ? replace : c);
}
return tb.finish();
}
/**
* Scans an external ID.
*
* @param f full flag
* @param r root flag
* @return id
* @throws IOException I/O exception
*/
private byte[] externalID(final boolean f, final boolean r) throws IOException {
byte[] cont = null;
final boolean pub = consume(PUBLIC);
if (pub || consume(SYSTEM)) {
checkS();
if (pub) {
pubidLit();
if (f) checkS();
}
final int qu = consume(); // [11]
if (qu == '\'' || qu == '"') {
int ch;
final TokenBuilder tok = new TokenBuilder();
while ((ch = nextChar()) != qu) tok.add(ch);
if (!f) return null;
final String name = string(tok.finish());
if (!dtd && r) return cont;
final XMLInput tin = input;
try {
final IO file = input.io().merge(name);
cont = file.read();
} catch (final IOException ex) {
Util.debug(ex);
// skip unknown DTDs/entities
cont = new byte[] {'?'};
}
input = new XMLInput(new IOContent(cont, name));
if (consume(XDECL)) {
check(XML);
s();
if (version()) checkS();
s();
if (encoding() == null) error(TEXTENC);
ch = nextChar();
if (s(ch)) ch = nextChar();
if (ch != '?') error(WRONGCHAR, '?', ch);
ch = nextChar();
if (ch != '>') error(WRONGCHAR, '>', ch);
cont = Arrays.copyOfRange(cont, input.pos(), cont.length);
}
s();
if (r) {
extSubsetDecl();
if (!consume((char) 0)) error(INVEND);
}
input = tin;
} else {
if (f) error(SCANQUOTE, (char) qu);
prev(1);
}
}
return cont;
}
/**
* Scans a comment.
*
* @throws IOException I/O exception
*/
private void comment() throws IOException {
do {
final int ch = nextChar();
if (ch == '-' && consume('-')) {
check('>');
return;
}
token.add(ch);
} while (true);
}
/**
* Scans XML text.
*
* @param ch current character
* @throws IOException I/O exception
*/
private void content(final int ch) throws IOException {
type = Type.TEXT;
boolean f = true;
int c = ch;
while (c != 0) {
if (c != '<') {
if (c == '&') {
// scan entity
final byte[] r = ref(true);
if (r.length == 1) token.add(r);
else if (!input.add(r, false)) error(RECENT);
} else {
if (c == ']') {
// ']]>' not allowed in content
if (consume() == ']') {
if (consume() == '>') error(CONTCDATA);
prev(1);
}
prev(1);
}
// add character to cached content
token.add(c);
}
} else {
if (!f && !isCDATA()) {
text = false;
prev(1);
if (chop) token.trim();
return;
}
cDATA();
}
c = consume();
f = false;
}
// end of file
if (!fragment) {
if (!ws(token.finish())) error(AFTERROOT);
type = Type.EOF;
}
}
/**
* Deletes the specified character from the token.
*
* @param token token
* @param ch character to be removed
* @return resulting token
*/
public static byte[] delete(final byte[] token, final int ch) {
// skip step if specified ascii character is not contained
if (ch < 0x80 && !contains(token, ch)) return token;
// remove character
final TokenBuilder tb = new TokenBuilder(token.length);
final int tl = token.length;
for (int i = 0; i < tl; i += cl(token, i)) {
final int c = cp(token, i);
if (c != ch) tb.add(c);
}
return tb.finish();
}
/**
* Consumes an XML name. [5]
*
* @param f force parsing
* @return name
* @throws IOException I/O exception
*/
private byte[] name(final boolean f) throws IOException {
final TokenBuilder name = new TokenBuilder();
int c = consume();
if (!isStartChar(c)) {
if (f) error(INVNAME);
prev(1);
return null;
}
do name.add(c);
while (isChar(c = nextChar()));
prev(1);
return name.finish();
}
/**
* Converts a string to a UTF8 byte array.
*
* @param string string to be converted
* @return byte array
*/
private static byte[] utf8(final String string) {
final char[] arr = string.toCharArray();
final int al = arr.length;
final TokenBuilder tb = new TokenBuilder(al << 1);
for (int c = 0; c < al; ++c) {
final char ch = arr[c];
tb.add(
Character.isHighSurrogate(ch) && c < al - 1 && Character.isLowSurrogate(arr[c + 1])
? Character.toCodePoint(ch, arr[++c])
: ch);
}
return tb.finish();
}
/**
* Scans an XML tag.
*
* @param ch current character
* @throws IOException I/O exception
*/
private void scanTAG(final int ch) throws IOException {
int c = ch;
// scan tag end...
if (c == '>') {
type = Type.R_BR;
state = State.CONTENT;
} else if (c == '=') {
// scan equal sign...
type = Type.EQ;
} else if (c == '\'' || c == '"') {
// scan quote...
type = Type.QUOTE;
state = State.QUOTE;
quote = c;
} else if (c == '/') {
// scan empty tag end...
type = Type.CLOSE_R_BR;
if ((c = nextChar()) == '>') {
state = State.CONTENT;
} else {
token.add(c);
error(CLOSING);
}
} else if (s(c)) {
// scan whitespace...
type = Type.WS;
} else if (isStartChar(c)) {
// scan tag name...
type = state == State.ATT ? Type.ATTNAME : Type.ELEMNAME;
do token.add(c);
while (isChar(c = nextChar()));
prev(1);
state = State.ATT;
} else {
// undefined character...
error(CHARACTER, (char) c);
}
}
/**
* Returns an information string for an unexpected exception.
*
* @param ex exception
* @return dummy object
*/
public static String bug(final Throwable ex) {
final TokenBuilder tb = new TokenBuilder(BUGINFO);
tb.add(NL).add("Contact: ").add(MAIL);
tb.add(NL).add("Version: ").add(TITLE);
tb.add(NL).add("Java: ").add(System.getProperty("java.vendor"));
tb.add(", ").add(System.getProperty("java.version"));
tb.add(NL).add("OS: ").add(System.getProperty("os.name"));
tb.add(", ").add(System.getProperty("os.arch"));
tb.add(NL).add("Stack Trace: ");
for (final String e : toArray(ex)) tb.add(NL).add(e);
return tb.toString();
}
/**
* Converts the path to a string array, containing the single segments.
*
* @param path path, or {@code null}
* @return path depth
*/
public static String[] toSegments(final String path) {
final StringList sl = new StringList();
if (path != null) {
final TokenBuilder tb = new TokenBuilder();
for (int s = 0; s < path.length(); s++) {
final char ch = path.charAt(s);
if (ch == '/') {
if (tb.isEmpty()) continue;
sl.add(tb.toString());
tb.reset();
} else {
tb.add(ch);
}
}
if (!tb.isEmpty()) sl.add(tb.toString());
}
return sl.toArray();
}
/**
* Splits a token around matches of the given separator.
*
* @param token token to be split
* @param sep separation character
* @return array
*/
public static byte[][] split(final byte[] token, final int sep) {
final int l = token.length;
final byte[][] split = new byte[l][];
int s = 0;
final TokenBuilder tb = new TokenBuilder();
for (int i = 0; i < l; i += cl(token, i)) {
final int c = cp(token, i);
if (c == sep) {
if (!tb.isEmpty()) {
split[s++] = tb.finish();
tb.reset();
}
} else {
tb.add(c);
}
}
if (!tb.isEmpty()) split[s++] = tb.finish();
return Array.copyOf(split, s);
}
@Override
protected byte[] stem(final byte[] word) {
int ln = 0;
final int wl = word.length;
final char[] s = new char[wl];
for (int i = 0; i < wl; i += Token.cl(word, i)) {
s[ln++] = (char) Token.cp(word, i);
}
if (ln < 4) return word;
final int olen = ln;
// "short rules": if it hits one of these, it skips the "long list"
int l = rule0(s, ln);
l = rule1(s, l);
l = rule2(s, l);
l = rule3(s, l);
l = rule4(s, l);
l = rule5(s, l);
l = rule6(s, l);
l = rule7(s, l);
l = rule8(s, l);
l = rule9(s, l);
l = rule10(s, l);
l = rule11(s, l);
l = rule12(s, l);
l = rule13(s, l);
l = rule14(s, l);
l = rule15(s, l);
l = rule16(s, l);
l = rule17(s, l);
l = rule18(s, l);
l = rule19(s, l);
l = rule20(s, l);
if (l == olen) l = rule21(s, l);
// "long list"
l = rule22(s, l);
final TokenBuilder tb = new TokenBuilder(l << 1);
for (int i = 0; i < l; i++) tb.add(s[i]);
return tb.finish();
}
/**
* Returns a string representation of the index structure.
*
* @param all include database contents in the representation. During updates, database lookups
* must be avoided, as the data structures will be inconsistent.
* @return string
*/
public String toString(final boolean all) {
final TokenBuilder tb = new TokenBuilder();
tb.addExt(type).add(" INDEX, '").add(data.meta.name).add("':\n");
final int s = lenList.size();
for (int m = 1; m < s; m++) {
final int len = lenList.get(m);
if (len == 0) continue;
final int[] ids = idsList.get(m);
tb.add(" ").addInt(m);
if (all)
tb.add(", key: \"").add(data.text(data.pre(ids[0]), type == IndexType.TEXT)).add('"');
tb.add(", ids");
if (all) tb.add("/pres");
tb.add(": ");
for (int n = 0; n < len; n++) {
if (n != 0) tb.add(",");
tb.addInt(ids[n]);
if (all) tb.add('/').addInt(data.pre(ids[n]));
}
tb.add("\n");
}
return tb.toString();
}
