jodd.json.JsonParser Maven / Gradle / Ivy
// Copyright (c) 2003-present, Jodd Team (http://jodd.org)
// All rights reserved.
//
// Redistribution and use in source and binary forms, with or without
// modification, are permitted provided that the following conditions are met:
//
// 1. Redistributions of source code must retain the above copyright notice,
// this list of conditions and the following disclaimer.
//
// 2. Redistributions in binary form must reproduce the above copyright
// notice, this list of conditions and the following disclaimer in the
// documentation and/or other materials provided with the distribution.
//
// THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
// AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
// IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
// ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE
// LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
// CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
// SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
// INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
// CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
// ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
// POSSIBILITY OF SUCH DAMAGE.
package jodd.json;
import jodd.introspector.ClassDescriptor;
import jodd.introspector.ClassIntrospector;
import jodd.introspector.PropertyDescriptor;
import jodd.json.meta.JsonAnnotationManager;
import jodd.json.meta.TypeData;
import jodd.util.CharArraySequence;
import jodd.util.CharUtil;
import jodd.util.StringPool;
import jodd.util.UnsafeUtil;
import java.math.BigInteger;
import java.util.ArrayList;
import java.util.Collection;
import java.util.HashMap;
import java.util.List;
import java.util.Map;
import java.util.function.Supplier;
/**
* Simple, developer-friendly JSON parser. It focuses on easy usage
* and type mappings. Uses Jodd's type converters, so it is natural
* companion for Jodd projects.
*
* This JSON parser also works in {@link #lazy(boolean)} mode. This
* mode is for top performance usage: parsing is done very, very lazy.
* While you can use all the mappings and other tools, for best performance
* the lazy mode should be used only with maps and lists (no special mappings).
* Also, the performance has it's price: more memory consumption, because the
* original input is hold until the result is in use.
*
* See: http://www.ietf.org/rfc/rfc4627.txt
*/
public class JsonParser extends JsonParserBase {
public static class Defaults {
public static final String DEFAULT_CLASS_METADATA_NAME = "__class";
/**
* Flag for enabling the lazy mode.
*/
public static boolean lazy = false;
/**
* Defines if parser will use extended paths information
* and path matching.
*/
public static boolean useAltPathsByParser = false;
/**
* Default value for loose mode.
*/
public static boolean loose = false;
/**
* Specifies if 'class' metadata is used and its value. When set, class metadata
* is used by {@link jodd.json.JsonSerializer} and all objects
* will have additional field with the class type in the resulting JSON.
* {@link jodd.json.JsonParser} will also consider this flag to build
* correct object type. If null, class information is not used.
*/
public static String classMetadataName = null;
public static boolean strictTypes = true;
}
/**
* Static ctor.
*/
public static JsonParser create() {
return new JsonParser();
}
/**
* Creates a lazy implementation of the JSON parser.
*/
public static JsonParser createLazyOne() {
return new JsonParser().lazy(true);
}
private static final char[] T_RUE = new char[] {'r', 'u', 'e'};
private static final char[] F_ALSE = new char[] {'a', 'l', 's', 'e'};
private static final char[] N_ULL = new char[] {'u', 'l', 'l'};
/**
* Map keys.
*/
public static final String KEYS = "keys";
/**
* Array or map values.
*/
public static final String VALUES = "values";
protected int ndx = 0;
protected char[] input;
protected int total;
protected Path path;
protected boolean useAltPaths = Defaults.useAltPathsByParser;
protected boolean lazy = Defaults.lazy;
protected boolean looseMode = Defaults.loose;
protected Class rootType;
protected MapToBean mapToBean;
private boolean notFirstObject;
private final JsonAnnotationManager jsonAnnotationManager;
public JsonParser() {
super(Defaults.strictTypes);
this.text = new char[512];
this.jsonAnnotationManager = JsonAnnotationManager.get();
}
/**
* Resets JSON parser, so it can be reused.
*/
protected void reset() {
this.ndx = 0;
this.textLen = 0;
this.path = new Path();
this.notFirstObject = false;
if (useAltPaths) {
path.altPath = new Path();
}
if (classMetadataName != null) {
mapToBean = createMapToBean(classMetadataName);
}
}
/**
* Enables usage of additional paths.
*/
public JsonParser useAltPaths() {
this.useAltPaths = true;
return this;
}
/**
* Enables 'loose' mode for parsing. When 'loose' mode is enabled,
* JSON parsers swallows also invalid JSONs:
*
* - invalid escape character sequence is simply added to the output
* - strings can be quoted with single-quotes
* - strings can be unquoted, but may not contain escapes
*
*/
public JsonParser looseMode(final boolean looseMode) {
this.looseMode = looseMode;
return this;
}
/**
* Defines if type conversion is strict. If not, all exceptions will be
* caught and replaced with {@code null}.
*/
public JsonParser strictTypes(final boolean strictTypes) {
this.strictTypes = strictTypes;
return this;
}
/**
* Defines how JSON parser works. In non-lazy mode, the whole JSON is parsed as it is.
* In the lazy mode, not everything is parsed, but some things are left lazy.
* This way we gain performance, especially on partial usage of the whole JSON.
* However, be aware that parser holds the input memory until the returned
* objects are disposed.
*/
public JsonParser lazy(final boolean lazy) {
this.lazy = lazy;
this.mapSupplier = lazy ? LAZYMAP_SUPPLIER : HASHMAP_SUPPLIER;
this.listSupplier = lazy ? LAZYLIST_SUPPLIER : ARRAYLIST_SUPPLIER;
return this;
}
// ---------------------------------------------------------------- mappings
protected Map mappings;
/**
* Maps a class to JSONs root.
*/
public JsonParser map(final Class target) {
rootType = target;
return this;
}
/**
* Maps a class to given path. For arrays, append values
* to the path to specify component type (if not specified by
* generics).
*/
public JsonParser map(final String path, final Class target) {
if (path == null) {
rootType = target;
return this;
}
if (mappings == null) {
mappings = new HashMap<>();
}
mappings.put(Path.parse(path), target);
return this;
}
/**
* Replaces type with mapped type for current path.
*/
protected Class replaceWithMappedTypeForPath(final Class target) {
if (mappings == null) {
return target;
}
Class newType;
// first try alt paths
Path altPath = path.getAltPath();
if (altPath != null) {
if (!altPath.equals(path)) {
newType = mappings.get(altPath);
if (newType != null) {
return newType;
}
}
}
// now check regular paths
newType = mappings.get(path);
if (newType != null) {
return newType;
}
return target;
}
// ---------------------------------------------------------------- converters
protected Map convs;
/**
* Defines {@link jodd.json.ValueConverter} to use on given path.
*/
public JsonParser withValueConverter(final String path, final ValueConverter valueConverter) {
if (convs == null) {
convs = new HashMap<>();
}
convs.put(Path.parse(path), valueConverter);
return this;
}
/**
* Lookups for value converter for current path.
*/
protected ValueConverter lookupValueConverter() {
if (convs == null) {
return null;
}
return convs.get(path);
}
// ---------------------------------------------------------------- class meta data name
protected String classMetadataName = Defaults.classMetadataName;
/**
* Sets local class meta-data name.
*
* Note that by using the class meta-data name you may expose a security hole in case untrusted source
* manages to specify a class that is accessible through class loader and exposes set of methods and/or fields,
* access of which opens an actual security hole. Such classes are known as “deserialization gadget”s.
*
* Because of this, use of "default typing" is not encouraged in general, and in particular is recommended against
* if the source of content is not trusted. Conversely, default typing may be used for processing content in
* cases where both ends (sender and receiver) are controlled by same entity.
*/
public JsonParser setClassMetadataName(final String name) {
classMetadataName = name;
return this;
}
/**
* Sets usage of default class meta-data name.
* Using it may introduce a security hole, see {@link #setClassMetadataName(String)} for more details.
* @see #setClassMetadataName(String)
*/
public JsonParser withClassMetadata(final boolean useMetadata) {
if (useMetadata) {
classMetadataName = Defaults.DEFAULT_CLASS_METADATA_NAME;
}
else {
classMetadataName = null;
}
return this;
}
/**
* Adds a {@link jodd.util.Wildcard wildcard} pattern for white-listing classes.
* @see #setClassMetadataName(String)
*/
public JsonParser allowClass(final String classPattern) {
if (super.classnameWhitelist == null) {
super.classnameWhitelist = new ArrayList<>();
}
classnameWhitelist.add(classPattern);
return this;
}
/**
* Removes the whitelist of allowed classes.
* @see #setClassMetadataName(String)
*/
public JsonParser allowAllClasses() {
classnameWhitelist = null;
return this;
}
// ---------------------------------------------------------------- parse
/**
* Parses input JSON as given type.
*/
@SuppressWarnings("unchecked")
public T parse(final String input, final Class targetType) {
rootType = targetType;
return _parse(UnsafeUtil.getChars(input));
}
/**
* Parses input JSON to {@link JsonObject}, special case of {@link #parse(String, Class)}.
*/
public JsonObject parseAsJsonObject(final String input) {
return new JsonObject(parse(input));
}
/**
* Parses input JSON to {@link JsonArray}, special case of parsing.
*/
public JsonArray parseAsJsonArray(final String input) {
return new JsonArray(parse(input));
}
/**
* Parses input JSON to a list with specified component type.
*/
public List parseAsList(final String string, final Class componentType) {
return new JsonParser()
.map(JsonParser.VALUES, componentType)
.parse(string);
}
/**
* Parses input JSON to a list with specified key and value types.
*/
public Map parseAsMap(
final String string, final Class keyType, final Class valueType) {
return new JsonParser()
.map(JsonParser.KEYS, keyType)
.map(JsonParser.VALUES, valueType)
.parse(string);
}
/**
* Parses input JSON string.
*/
public T parse(final String input) {
return _parse(UnsafeUtil.getChars(input));
}
/**
* Parses input JSON as given type.
*/
@SuppressWarnings("unchecked")
public T parse(final char[] input, final Class targetType) {
rootType = targetType;
return _parse(input);
}
/**
* Parses input JSON char array.
*/
public T parse(final char[] input) {
return _parse(input);
}
private T _parse(final char[] input) {
this.input = input;
this.total = input.length;
reset();
skipWhiteSpaces();
Object value;
try {
value = parseValue(rootType, null, null);
}
catch (IndexOutOfBoundsException iofbex) {
syntaxError("End of JSON");
return null;
}
skipWhiteSpaces();
if (ndx != total) {
syntaxError("Trailing chars");
return null;
}
if (lazy) {
// lets resolve root lazy values
value = resolveLazyValue(value);
}
// convert map to target type
if (classMetadataName != null && rootType == null) {
if (value instanceof Map) {
Map map = (Map) value;
value = mapToBean.map2bean(map, null);
}
}
return (T) value;
}
// ---------------------------------------------------------------- parser
/**
* Parses a JSON value.
* @param targetType target type to convert, may be null
* @param componentType component type for maps and arrays, may be null
*/
protected Object parseValue(final Class targetType, final Class keyType, final Class componentType) {
final ValueConverter valueConverter;
final char c = input[ndx];
switch (c) {
case '\'':
if (!looseMode) {
break;
}
case '"':
ndx++;
Object string = parseStringContent(c);
valueConverter = lookupValueConverter();
if (valueConverter != null) {
return valueConverter.convert(string);
}
if (targetType != null && targetType != String.class) {
string = convertType(string, targetType);
}
return string;
case '{':
ndx++;
if (lazy) {
if (notFirstObject) {
final Object value = new ObjectParser(this, targetType, keyType, componentType);
skipObject();
return value;
}
else {
notFirstObject = true;
}
}
return parseObjectContent(targetType, keyType, componentType);
case '[':
ndx++;
return parseArrayContent(targetType, componentType);
case '0':
case '1':
case '2':
case '3':
case '4':
case '5':
case '6':
case '7':
case '8':
case '9':
case '-':
Object number = parseNumber();
valueConverter = lookupValueConverter();
if (valueConverter != null) {
return valueConverter.convert(number);
}
if (targetType != null) {
number = convertType(number, targetType);
}
return number;
case 'n':
ndx++;
if (match(N_ULL)) {
valueConverter = lookupValueConverter();
if (valueConverter != null) {
return valueConverter.convert(null);
}
return null;
}
break;
case 't':
ndx++;
if (match(T_RUE)) {
Object value = Boolean.TRUE;
valueConverter = lookupValueConverter();
if (valueConverter != null) {
return valueConverter.convert(value);
}
if (targetType != null) {
value = convertType(value, targetType);
}
return value;
}
break;
case 'f':
ndx++;
if (match(F_ALSE)) {
Object value = Boolean.FALSE;
valueConverter = lookupValueConverter();
if (valueConverter != null) {
return valueConverter.convert(value);
}
if (targetType != null) {
value = convertType(value, targetType);
}
return value;
}
break;
}
if (looseMode) {
// try to parse unquoted string
Object string = parseUnquotedStringContent();
valueConverter = lookupValueConverter();
if (valueConverter != null) {
return valueConverter.convert(string);
}
if (targetType != null && targetType != String.class) {
string = convertType(string, targetType);
}
return string;
}
syntaxError("Invalid char: " + input[ndx]);
return null;
}
// ---------------------------------------------------------------- lazy
/**
* Resolves lazy value during the parsing runtime.
*/
private Object resolveLazyValue(Object value) {
if (value instanceof Supplier) {
value = ((Supplier)value).get();
}
return value;
}
/**
* Skips over complete object. It is not parsed, just skipped. It will be
* parsed later, but only if required.
*/
private void skipObject() {
int bracketCount = 1;
boolean insideString = false;
while (ndx < total) {
final char c = input[ndx];
if (insideString) {
if (c == '\"' && notPrecededByEvenNumberOfBackslashes()) {
insideString = false;
}
} else if (c == '\"') {
insideString = true;
} else if (c == '{') {
bracketCount++;
} else if (c == '}') {
bracketCount--;
if (bracketCount == 0) {
ndx++;
return;
}
}
ndx++;
}
}
private boolean notPrecededByEvenNumberOfBackslashes() {
int pos = ndx;
int count = 0;
while (pos > 0 && input[pos - 1] == '\\') {
count++;
pos--;
}
return count % 2 == 0;
}
// ---------------------------------------------------------------- string
protected char[] text;
protected int textLen;
/**
* Parses a string.
*/
protected String parseString() {
char quote = '\"';
if (looseMode) {
quote = consumeOneOf('\"', '\'');
if (quote == 0) {
return parseUnquotedStringContent();
}
} else {
consume(quote);
}
return parseStringContent(quote);
}
/**
* Parses string content, once when starting quote has been consumed.
*/
protected String parseStringContent(final char quote) {
final int startNdx = ndx;
// roll-out until the end of the string or the escape char
while (true) {
final char c = input[ndx];
if (c == quote) {
// no escapes found, just use existing string
ndx++;
return new String(input, startNdx, ndx - 1 - startNdx);
}
if (c == '\\') {
break;
}
ndx++;
}
// escapes found, proceed differently
textLen = ndx - startNdx;
growEmpty();
// for (int i = startNdx, j = 0; j < textLen; i++, j++) {
// text[j] = input[i];
// }
System.arraycopy(input, startNdx, text, 0, textLen);
// escape char, process everything until the end
while (true) {
char c = input[ndx];
if (c == quote) {
// done
ndx++;
final String str = new String(text, 0, textLen);
textLen = 0;
return str;
}
if (c == '\\') {
// escape char found
ndx++;
c = input[ndx];
switch (c) {
case '\"' : c = '\"'; break;
case '\\' : c = '\\'; break;
case '/' : c = '/'; break;
case 'b' : c = '\b'; break;
case 'f' : c = '\f'; break;
case 'n' : c = '\n'; break;
case 'r' : c = '\r'; break;
case 't' : c = '\t'; break;
case 'u' :
ndx++;
c = parseUnicode();
break;
default:
if (looseMode) {
if (c != '\'') {
c = '\\';
ndx--;
}
}
else {
syntaxError("Invalid escape char: " + c);
}
}
}
text[textLen] = c;
textLen++;
growAndCopy();
ndx++;
}
}
/**
* Grows empty text array.
*/
protected void growEmpty() {
if (textLen >= text.length) {
int newSize = textLen << 1;
text = new char[newSize];
}
}
/**
* Grows text array when {@code text.length == textLen}.
*/
protected void growAndCopy() {
if (textLen == text.length) {
int newSize = text.length << 1;
char[] newText = new char[newSize];
if (textLen > 0) {
System.arraycopy(text, 0, newText, 0, textLen);
}
text = newText;
}
}
/**
* Parses 4 characters and returns unicode character.
*/
protected char parseUnicode() {
int i0 = CharUtil.hex2int(input[ndx++]);
int i1 = CharUtil.hex2int(input[ndx++]);
int i2 = CharUtil.hex2int(input[ndx++]);
int i3 = CharUtil.hex2int(input[ndx]);
return (char) ((i0 << 12) + (i1 << 8) + (i2 << 4) + i3);
}
// ---------------------------------------------------------------- un-quoted
private final static char[] UNQUOTED_DELIMETERS = ",:[]{}\\\"'".toCharArray();
/**
* Parses un-quoted string content.
*/
protected String parseUnquotedStringContent() {
final int startNdx = ndx;
while (true) {
final char c = input[ndx];
if (c <= ' ' || CharUtil.equalsOne(c, UNQUOTED_DELIMETERS)) {
final int currentNdx = ndx;
// done
skipWhiteSpaces();
return new String(input, startNdx, currentNdx - startNdx);
}
ndx++;
}
}
// ---------------------------------------------------------------- number
/**
* Parses JSON numbers.
*/
protected Number parseNumber() {
final int startIndex = ndx;
char c = input[ndx];
boolean isDouble = false;
boolean isExp = false;
if (c == '-') {
ndx++;
}
while (true) {
if (isEOF()) {
break;
}
c = input[ndx];
if (c >= '0' && c <= '9') {
ndx++;
continue;
}
if (c <= 32) { // white space
break;
}
if (c == ',' || c == '}' || c == ']') { // delimiter
break;
}
if (c == '.') {
isDouble = true;
}
else if (c == 'e' || c == 'E') {
isExp = true;
}
ndx++;
}
final String value = new String(input, startIndex, ndx - startIndex);
if (isDouble) {
return Double.valueOf(value);
}
long longNumber;
if (isExp) {
longNumber = Double.valueOf(value).longValue();
}
else {
if (value.length() >= 19) {
// if string is 19 chars and longer, it can be over the limit
BigInteger bigInteger = new BigInteger(value);
if (isGreaterThanLong(bigInteger)) {
return bigInteger;
}
longNumber = bigInteger.longValue();
}
else {
longNumber = Long.parseLong(value);
}
}
if ((longNumber >= Integer.MIN_VALUE) && (longNumber <= Integer.MAX_VALUE)) {
return (int) longNumber;
}
return longNumber;
}
private static boolean isGreaterThanLong(final BigInteger bigInteger) {
return bigInteger.compareTo(MAX_LONG) > 0 || bigInteger.compareTo(MIN_LONG) < 0;
}
private static final BigInteger MAX_LONG = BigInteger.valueOf(Long.MAX_VALUE);
private static final BigInteger MIN_LONG = BigInteger.valueOf(Long.MIN_VALUE);
// ---------------------------------------------------------------- array
/**
* Parses arrays, once when open bracket has been consumed.
*/
protected Object parseArrayContent(Class targetType, Class componentType) {
// detect special case
if (targetType == Object.class) {
targetType = List.class;
}
// continue
targetType = replaceWithMappedTypeForPath(targetType);
if (componentType == null && targetType != null && targetType.isArray()) {
componentType = targetType.getComponentType();
}
path.push(VALUES);
componentType = replaceWithMappedTypeForPath(componentType);
Collection