XMLTokener.java
package org.json;
/*
Public Domain.
*/
import java.io.Reader;
/**
* The XMLTokener extends the JSONTokener to provide additional methods
* for the parsing of XML texts.
* @author JSON.org
* @version 2015-12-09
*/
public class XMLTokener extends JSONTokener {
/** The table of entity values. It initially contains Character values for
* amp, apos, gt, lt, quot.
*/
public static final java.util.HashMap<String, Character> entity;
private XMLParserConfiguration configuration = XMLParserConfiguration.ORIGINAL;
static {
entity = new java.util.HashMap<String, Character>(8);
entity.put("amp", XML.AMP);
entity.put("apos", XML.APOS);
entity.put("gt", XML.GT);
entity.put("lt", XML.LT);
entity.put("quot", XML.QUOT);
}
/**
* Construct an XMLTokener from a Reader.
* @param r A source reader.
*/
public XMLTokener(Reader r) {
super(r);
}
/**
* Construct an XMLTokener from a string.
* @param s A source string.
*/
public XMLTokener(String s) {
super(s);
}
/**
* Construct an XMLTokener from a Reader and an XMLParserConfiguration.
* @param r A source reader.
* @param configuration the configuration that can be used to set certain flags
*/
public XMLTokener(Reader r, XMLParserConfiguration configuration) {
super(r);
this.configuration = configuration;
}
/**
* Get the text in the CDATA block.
* @return The string up to the <code>]]></code>.
* @throws JSONException If the <code>]]></code> is not found.
*/
public String nextCDATA() throws JSONException {
char c;
int i;
StringBuilder sb = new StringBuilder();
while (more()) {
c = next();
sb.append(c);
i = sb.length() - 3;
if (i >= 0 && sb.charAt(i) == ']' &&
sb.charAt(i + 1) == ']' && sb.charAt(i + 2) == '>') {
sb.setLength(i);
return sb.toString();
}
}
throw syntaxError("Unclosed CDATA");
}
/**
* Get the next XML outer token, trimming whitespace. There are two kinds
* of tokens: the <pre>{@code '<' }</pre> character which begins a markup
* tag, and the content
* text between markup tags.
*
* @return A string, or a <pre>{@code '<' }</pre> Character, or null if
* there is no more source text.
* @throws JSONException if a called function has an error
*/
public Object nextContent() throws JSONException {
char c;
StringBuilder sb;
do {
c = next();
} while (Character.isWhitespace(c) && configuration.shouldTrimWhiteSpace());
if (c == 0) {
return null;
}
if (c == '<') {
return XML.LT;
}
sb = new StringBuilder();
for (;;) {
if (c == 0) {
return sb.toString().trim();
}
if (c == '<') {
back();
if (configuration.shouldTrimWhiteSpace()) {
return sb.toString().trim();
} else return sb.toString();
}
if (c == '&') {
sb.append(nextEntity(c));
} else {
sb.append(c);
}
c = next();
}
}
/**
* <pre>{@code
* Return the next entity. These entities are translated to Characters:
* & ' > < ".
* }</pre>
* @param ampersand An ampersand character.
* @return A Character or an entity String if the entity is not recognized.
* @throws JSONException If missing ';' in XML entity.
*/
public Object nextEntity(@SuppressWarnings("unused") char ampersand) throws JSONException {
StringBuilder sb = new StringBuilder();
for (;;) {
char c = next();
if (Character.isLetterOrDigit(c) || c == '#') {
sb.append(Character.toLowerCase(c));
} else if (c == ';') {
break;
} else {
throw syntaxError("Missing ';' in XML entity: &" + sb);
}
}
String string = sb.toString();
return unescapeEntity(string);
}
/**
* Unescape an XML entity encoding;
* @param e entity (only the actual entity value, not the preceding & or ending ;
* @return the unescaped entity string
* @throws JSONException if the entity is malformed
*/
static String unescapeEntity(String e) throws JSONException {
// validate
if (e == null || e.isEmpty()) {
return "";
}
// if our entity is an encoded unicode point, parse it.
if (e.charAt(0) == '#') {
if (e.length() < 2) {
throw new JSONException("Invalid numeric character reference: &#;");
}
int cp = (e.charAt(1) == 'x' || e.charAt(1) == 'X')
? parseHexEntity(e)
: parseDecimalEntity(e);
return new String(new int[] {cp}, 0, 1);
}
Character knownEntity = entity.get(e);
if (knownEntity == null) {
// we don't know the entity so keep it encoded
return '&' + e + ';';
}
return knownEntity.toString();
}
/**
* Parse a hexadecimal numeric character reference (e.g., "઼").
* @param e entity string starting with '#' (e.g., "#x1F4A9")
* @return the Unicode code point
* @throws JSONException if the format is invalid
*/
private static int parseHexEntity(String e) throws JSONException {
// hex encoded unicode - need at least one hex digit after #x
if (e.length() < 3) {
throw new JSONException("Invalid hex character reference: missing hex digits in &#" + e.substring(1) + ";");
}
String hex = e.substring(2);
if (!isValidHex(hex)) {
throw new JSONException("Invalid hex character reference: &#" + e.substring(1) + ";");
}
try {
return Integer.parseInt(hex, 16);
} catch (NumberFormatException nfe) {
throw new JSONException("Invalid hex character reference: &#" + e.substring(1) + ";", nfe);
}
}
/**
* Parse a decimal numeric character reference (e.g., "{").
* @param e entity string starting with '#' (e.g., "#123")
* @return the Unicode code point
* @throws JSONException if the format is invalid
*/
private static int parseDecimalEntity(String e) throws JSONException {
String decimal = e.substring(1);
if (!isValidDecimal(decimal)) {
throw new JSONException("Invalid decimal character reference: &#" + decimal + ";");
}
try {
return Integer.parseInt(decimal);
} catch (NumberFormatException nfe) {
throw new JSONException("Invalid decimal character reference: &#" + decimal + ";", nfe);
}
}
/**
* Check if a string contains only valid hexadecimal digits.
* @param s the string to check
* @return true if s is non-empty and contains only hex digits (0-9, a-f, A-F)
*/
private static boolean isValidHex(String s) {
if (s == null || s.isEmpty()) {
return false;
}
for (int i = 0; i < s.length(); i++) {
char c = s.charAt(i);
if (!((c >= '0' && c <= '9') || (c >= 'a' && c <= 'f') || (c >= 'A' && c <= 'F'))) {
return false;
}
}
return true;
}
/**
* Check if a string contains only valid decimal digits.
* @param s the string to check
* @return true if s is non-empty and contains only digits (0-9)
*/
private static boolean isValidDecimal(String s) {
if (s == null || s.isEmpty()) {
return false;
}
for (int i = 0; i < s.length(); i++) {
char c = s.charAt(i);
if (c < '0' || c > '9') {
return false;
}
}
return true;
}
/**
* <pre>{@code
* Returns the next XML meta token. This is used for skipping over <!...>
* and <?...?> structures.
* }</pre>
* @return <pre>{@code Syntax characters (< > / = ! ?) are returned as
* Character, and strings and names are returned as Boolean. We don't care
* what the values actually are.
* }</pre>
* @throws JSONException If a string is not properly closed or if the XML
* is badly structured.
*/
public Object nextMeta() throws JSONException {
char c;
char q;
do {
c = next();
} while (Character.isWhitespace(c));
switch (c) {
case 0:
throw syntaxError("Misshaped meta tag");
case '<':
return XML.LT;
case '>':
return XML.GT;
case '/':
return XML.SLASH;
case '=':
return XML.EQ;
case '!':
return XML.BANG;
case '?':
return XML.QUEST;
case '"':
case '\'':
q = c;
for (;;) {
c = next();
if (c == 0) {
throw syntaxError("Unterminated string");
}
if (c == q) {
return Boolean.TRUE;
}
}
default:
for (;;) {
c = next();
if (Character.isWhitespace(c)) {
return Boolean.TRUE;
}
switch (c) {
case 0:
throw syntaxError("Unterminated string");
case '<':
case '>':
case '/':
case '=':
case '!':
case '?':
case '"':
case '\'':
back();
return Boolean.TRUE;
}
}
}
}
/**
* <pre>{@code
* Get the next XML Token. These tokens are found inside of angle
* brackets. It may be one of these characters: / > = ! ? or it
* may be a string wrapped in single quotes or double quotes, or it may be a
* name.
* }</pre>
* @return a String or a Character.
* @throws JSONException If the XML is not well formed.
*/
public Object nextToken() throws JSONException {
char c;
char q;
StringBuilder sb;
do {
c = next();
} while (Character.isWhitespace(c));
switch (c) {
case 0:
throw syntaxError("Misshaped element");
case '<':
throw syntaxError("Misplaced '<'");
case '>':
return XML.GT;
case '/':
return XML.SLASH;
case '=':
return XML.EQ;
case '!':
return XML.BANG;
case '?':
return XML.QUEST;
// Quoted string
case '"':
case '\'':
q = c;
sb = new StringBuilder();
for (;;) {
c = next();
if (c == 0) {
throw syntaxError("Unterminated string");
}
if (c == q) {
return sb.toString();
}
if (c == '&') {
sb.append(nextEntity(c));
} else {
sb.append(c);
}
}
default:
// Name
sb = new StringBuilder();
for (;;) {
sb.append(c);
c = next();
if (Character.isWhitespace(c)) {
return sb.toString();
}
switch (c) {
case 0:
return sb.toString();
case '>':
case '/':
case '=':
case '!':
case '?':
case '[':
case ']':
back();
return sb.toString();
case '<':
case '"':
case '\'':
throw syntaxError("Bad character in a name");
}
}
}
}
/**
* Skip characters until past the requested string.
* If it is not found, we are left at the end of the source with a result of false.
* @param to A string to skip past.
*/
// The Android implementation of JSONTokener has a public method of public void skipPast(String to)
// even though ours does not have that method, to have API compatibility, our method in the subclass
// should match.
public void skipPast(String to) {
boolean b;
char c;
int i;
int j;
int offset = 0;
int length = to.length();
char[] circle = new char[length];
/*
* First fill the circle buffer with as many characters as are in the
* to string. If we reach an early end, bail.
*/
for (i = 0; i < length; i += 1) {
c = next();
if (c == 0) {
return;
}
circle[i] = c;
}
/* We will loop, possibly for all of the remaining characters. */
for (;;) {
j = offset;
b = true;
/* Compare the circle buffer with the to string. */
for (i = 0; i < length; i += 1) {
if (circle[j] != to.charAt(i)) {
b = false;
break;
}
j += 1;
if (j >= length) {
j -= length;
}
}
/* If we exit the loop with b intact, then victory is ours. */
if (b) {
return;
}
/* Get the next character. If there isn't one, then defeat is ours. */
c = next();
if (c == 0) {
return;
}
/*
* Shove the character in the circle buffer and advance the
* circle offset. The offset is mod n.
*/
circle[offset] = c;
offset += 1;
if (offset >= length) {
offset -= length;
}
}
}
}