Skip to content

Instantly share code, notes, and snippets.

@darktable
Created November 30, 2011 23:08
Show Gist options
  • Save darktable/1411710 to your computer and use it in GitHub Desktop.
Save darktable/1411710 to your computer and use it in GitHub Desktop.
Unity3D: MiniJSON Decodes and encodes simple JSON strings. Not intended for use with massive JSON strings, probably < 32k preferred. Handy for parsing JSON from inside Unity3d.
/*
* Copyright (c) 2013 Calvin Rien
*
* Based on the JSON parser by Patrick van Bergen
* http://techblog.procurios.nl/k/618/news/view/14605/14863/How-do-I-write-my-own-parser-for-JSON.html
*
* Simplified it so that it doesn't throw exceptions
* and can be used in Unity iPhone with maximum code stripping.
*
* Permission is hereby granted, free of charge, to any person obtaining
* a copy of this software and associated documentation files (the
* "Software"), to deal in the Software without restriction, including
* without limitation the rights to use, copy, modify, merge, publish,
* distribute, sublicense, and/or sell copies of the Software, and to
* permit persons to whom the Software is furnished to do so, subject to
* the following conditions:
*
* The above copyright notice and this permission notice shall be
* included in all copies or substantial portions of the Software.
*
* THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
* EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
* MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.
* IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY
* CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT,
* TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE
* SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
*/
using System;
using System.Collections;
using System.Collections.Generic;
using System.IO;
using System.Text;
namespace MiniJSON {
// Example usage:
//
// using UnityEngine;
// using System.Collections;
// using System.Collections.Generic;
// using MiniJSON;
//
// public class MiniJSONTest : MonoBehaviour {
// void Start () {
// var jsonString = "{ \"array\": [1.44,2,3], " +
// "\"object\": {\"key1\":\"value1\", \"key2\":256}, " +
// "\"string\": \"The quick brown fox \\\"jumps\\\" over the lazy dog \", " +
// "\"unicode\": \"\\u3041 Men\u00fa sesi\u00f3n\", " +
// "\"int\": 65536, " +
// "\"float\": 3.1415926, " +
// "\"bool\": true, " +
// "\"null\": null }";
//
// var dict = Json.Deserialize(jsonString) as Dictionary<string,object>;
//
// Debug.Log("deserialized: " + dict.GetType());
// Debug.Log("dict['array'][0]: " + ((List<object>) dict["array"])[0]);
// Debug.Log("dict['string']: " + (string) dict["string"]);
// Debug.Log("dict['float']: " + (double) dict["float"]); // floats come out as doubles
// Debug.Log("dict['int']: " + (long) dict["int"]); // ints come out as longs
// Debug.Log("dict['unicode']: " + (string) dict["unicode"]);
//
// var str = Json.Serialize(dict);
//
// Debug.Log("serialized: " + str);
// }
// }
/// <summary>
/// This class encodes and decodes JSON strings.
/// Spec. details, see http://www.json.org/
///
/// JSON uses Arrays and Objects. These correspond here to the datatypes IList and IDictionary.
/// All numbers are parsed to doubles.
/// </summary>
public static class Json {
/// <summary>
/// Parses the string json into a value
/// </summary>
/// <param name="json">A JSON string.</param>
/// <returns>An List&lt;object&gt;, a Dictionary&lt;string, object&gt;, a double, an integer,a string, null, true, or false</returns>
public static object Deserialize(string json) {
// save the string for debug information
if (json == null) {
return null;
}
return Parser.Parse(json);
}
sealed class Parser : IDisposable {
const string WORD_BREAK = "{}[],:\"";
public static bool IsWordBreak(char c) {
return Char.IsWhiteSpace(c) || WORD_BREAK.IndexOf(c) != -1;
}
enum TOKEN {
NONE,
CURLY_OPEN,
CURLY_CLOSE,
SQUARED_OPEN,
SQUARED_CLOSE,
COLON,
COMMA,
STRING,
NUMBER,
TRUE,
FALSE,
NULL
};
StringReader json;
Parser(string jsonString) {
json = new StringReader(jsonString);
}
public static object Parse(string jsonString) {
using (var instance = new Parser(jsonString)) {
return instance.ParseValue();
}
}
public void Dispose() {
json.Dispose();
json = null;
}
Dictionary<string, object> ParseObject() {
Dictionary<string, object> table = new Dictionary<string, object>();
// ditch opening brace
json.Read();
// {
while (true) {
switch (NextToken) {
case TOKEN.NONE:
return null;
case TOKEN.COMMA:
continue;
case TOKEN.CURLY_CLOSE:
return table;
default:
// name
string name = ParseString();
if (name == null) {
return null;
}
// :
if (NextToken != TOKEN.COLON) {
return null;
}
// ditch the colon
json.Read();
// value
table[name] = ParseValue();
break;
}
}
}
List<object> ParseArray() {
List<object> array = new List<object>();
// ditch opening bracket
json.Read();
// [
var parsing = true;
while (parsing) {
TOKEN nextToken = NextToken;
switch (nextToken) {
case TOKEN.NONE:
return null;
case TOKEN.COMMA:
continue;
case TOKEN.SQUARED_CLOSE:
parsing = false;
break;
default:
object value = ParseByToken(nextToken);
array.Add(value);
break;
}
}
return array;
}
object ParseValue() {
TOKEN nextToken = NextToken;
return ParseByToken(nextToken);
}
object ParseByToken(TOKEN token) {
switch (token) {
case TOKEN.STRING:
return ParseString();
case TOKEN.NUMBER:
return ParseNumber();
case TOKEN.CURLY_OPEN:
return ParseObject();
case TOKEN.SQUARED_OPEN:
return ParseArray();
case TOKEN.TRUE:
return true;
case TOKEN.FALSE:
return false;
case TOKEN.NULL:
return null;
default:
return null;
}
}
string ParseString() {
StringBuilder s = new StringBuilder();
char c;
// ditch opening quote
json.Read();
bool parsing = true;
while (parsing) {
if (json.Peek() == -1) {
parsing = false;
break;
}
c = NextChar;
switch (c) {
case '"':
parsing = false;
break;
case '\\':
if (json.Peek() == -1) {
parsing = false;
break;
}
c = NextChar;
switch (c) {
case '"':
case '\\':
case '/':
s.Append(c);
break;
case 'b':
s.Append('\b');
break;
case 'f':
s.Append('\f');
break;
case 'n':
s.Append('\n');
break;
case 'r':
s.Append('\r');
break;
case 't':
s.Append('\t');
break;
case 'u':
var hex = new char[4];
for (int i=0; i< 4; i++) {
hex[i] = NextChar;
}
s.Append((char) Convert.ToInt32(new string(hex), 16));
break;
}
break;
default:
s.Append(c);
break;
}
}
return s.ToString();
}
object ParseNumber() {
string number = NextWord;
if (number.IndexOf('.') == -1) {
long parsedInt;
Int64.TryParse(number, out parsedInt);
return parsedInt;
}
double parsedDouble;
Double.TryParse(number, out parsedDouble);
return parsedDouble;
}
void EatWhitespace() {
while (Char.IsWhiteSpace(PeekChar)) {
json.Read();
if (json.Peek() == -1) {
break;
}
}
}
char PeekChar {
get {
return Convert.ToChar(json.Peek());
}
}
char NextChar {
get {
return Convert.ToChar(json.Read());
}
}
string NextWord {
get {
StringBuilder word = new StringBuilder();
while (!IsWordBreak(PeekChar)) {
word.Append(NextChar);
if (json.Peek() == -1) {
break;
}
}
return word.ToString();
}
}
TOKEN NextToken {
get {
EatWhitespace();
if (json.Peek() == -1) {
return TOKEN.NONE;
}
switch (PeekChar) {
case '{':
return TOKEN.CURLY_OPEN;
case '}':
json.Read();
return TOKEN.CURLY_CLOSE;
case '[':
return TOKEN.SQUARED_OPEN;
case ']':
json.Read();
return TOKEN.SQUARED_CLOSE;
case ',':
json.Read();
return TOKEN.COMMA;
case '"':
return TOKEN.STRING;
case ':':
return TOKEN.COLON;
case '0':
case '1':
case '2':
case '3':
case '4':
case '5':
case '6':
case '7':
case '8':
case '9':
case '-':
return TOKEN.NUMBER;
}
switch (NextWord) {
case "false":
return TOKEN.FALSE;
case "true":
return TOKEN.TRUE;
case "null":
return TOKEN.NULL;
}
return TOKEN.NONE;
}
}
}
/// <summary>
/// Converts a IDictionary / IList object or a simple type (string, int, etc.) into a JSON string
/// </summary>
/// <param name="json">A Dictionary&lt;string, object&gt; / List&lt;object&gt;</param>
/// <returns>A JSON encoded string, or null if object 'json' is not serializable</returns>
public static string Serialize(object obj) {
return Serializer.Serialize(obj);
}
sealed class Serializer {
StringBuilder builder;
Serializer() {
builder = new StringBuilder();
}
public static string Serialize(object obj) {
var instance = new Serializer();
instance.SerializeValue(obj);
return instance.builder.ToString();
}
void SerializeValue(object value) {
IList asList;
IDictionary asDict;
string asStr;
if (value == null) {
builder.Append("null");
} else if ((asStr = value as string) != null) {
SerializeString(asStr);
} else if (value is bool) {
builder.Append((bool) value ? "true" : "false");
} else if ((asList = value as IList) != null) {
SerializeArray(asList);
} else if ((asDict = value as IDictionary) != null) {
SerializeObject(asDict);
} else if (value is char) {
SerializeString(new string((char) value, 1));
} else {
SerializeOther(value);
}
}
void SerializeObject(IDictionary obj) {
bool first = true;
builder.Append('{');
foreach (object e in obj.Keys) {
if (!first) {
builder.Append(',');
}
SerializeString(e.ToString());
builder.Append(':');
SerializeValue(obj[e]);
first = false;
}
builder.Append('}');
}
void SerializeArray(IList anArray) {
builder.Append('[');
bool first = true;
foreach (object obj in anArray) {
if (!first) {
builder.Append(',');
}
SerializeValue(obj);
first = false;
}
builder.Append(']');
}
void SerializeString(string str) {
builder.Append('\"');
char[] charArray = str.ToCharArray();
foreach (var c in charArray) {
switch (c) {
case '"':
builder.Append("\\\"");
break;
case '\\':
builder.Append("\\\\");
break;
case '\b':
builder.Append("\\b");
break;
case '\f':
builder.Append("\\f");
break;
case '\n':
builder.Append("\\n");
break;
case '\r':
builder.Append("\\r");
break;
case '\t':
builder.Append("\\t");
break;
default:
int codepoint = Convert.ToInt32(c);
if ((codepoint >= 32) && (codepoint <= 126)) {
builder.Append(c);
} else {
builder.Append("\\u");
builder.Append(codepoint.ToString("x4"));
}
break;
}
}
builder.Append('\"');
}
void SerializeOther(object value) {
// NOTE: decimals lose precision during serialization.
// They always have, I'm just letting you know.
// Previously floats and doubles lost precision too.
if (value is float) {
builder.Append(((float) value).ToString("R"));
} else if (value is int
|| value is uint
|| value is long
|| value is sbyte
|| value is byte
|| value is short
|| value is ushort
|| value is ulong) {
builder.Append(value);
} else if (value is double
|| value is decimal) {
builder.Append(Convert.ToDouble(value).ToString("R"));
} else {
SerializeString(value.ToString());
}
}
}
}
}
@shiwano
Copy link

shiwano commented Apr 13, 2016

MiniJSON fails parsing floating point notation (like 1e-005). I fixed it with the following change.

@@ -290,7 +290,7 @@ namespace MiniJSON {
             object ParseNumber() {
                 string number = NextWord;

-                if (number.IndexOf('.') == -1) {
+                if (number.IndexOf('.') == -1 && number.IndexOf('E') == -1 && number.IndexOf('e') == -1) {
                     long parsedInt;
                     Int64.TryParse(number, out parsedInt);
                     return parsedInt;

@Jackyjjc
Copy link

Jackyjjc commented May 8, 2016

Thanks for this great piece of code! I started using it for my Unity project. However, there has been many comments about issues and fixes, I think it would be a good idea if they are managed somewhere so that someone who want to use the library doesn't have to copy paste all those fixes. Therefore, I've create a repo: https://github.com/Jackyjjc/MiniJSON.cs I've included quite a few fixes mentioned above but not all of them since I didn't understand the full context of the fix. Feel free to submit a pull request for your fix though.

I'm happy to hand it to @darktable if he is willing to manage it :)

@kornman00
Copy link

w.r.t. ParseArray throwing an OutOfMemoryException, I had to change the 'array' variable in ParseArray to be lazily allocated in addition to @prestonmediaspike's changes

List<object> ParseArray() {
    // OOM fix: start out as null
    List<object> array = null;// = new List<object>();

    // ditch opening bracket
    json.Read();

    // [
    var parsing = true;
    while (parsing) {
        TOKEN nextToken = NextToken;

        switch (nextToken) {
        case TOKEN.NONE:
            return null;
        case TOKEN.COMMA:
            continue;
        case TOKEN.SQUARED_CLOSE:
            parsing = false;
            break;
        // OOM fix
        case TOKEN.COLON:
            json.Read(); //invalid array: consume colon to prevent infinite loop
            break;
        default:
            object value = ParseByToken(nextToken);

            // OOM fix
            if (array == null)
                array = new List<object>();

            array.Add(value);
            break;
        }
    }

    // OOM fix
    if (array == null)
        array = new List<object>();

    return array;
}

@sotirosn
Copy link

When serializing arbitrary objects it would be much better to use Unity's JsonUtility.ToJson method which uses reflection instead of serializing the string of the object's ToString() method.

In other words change line 542:
- SerializeString(value.ToString());
+ builder.Append(UnityEngine.JsonUtility.ToJson(value));

The reason is that presently you can serialize objects with Unity but you cannot serialize arbitrary arrays, lists or dictionaries. And with MiniJSON you can serialize arbitrary arrays, lists and dictionaries, but turning an object into its ToString() string and then serializing that string is pretty useless. Now you can do something like this:

[System.Serializable]
public struct Token {
   public long timestamp;
   public string signature;
   public string id;
}

Json.Serialize(new object[] {new Token { timestamp = Now(), signature = "sha256 hash", id="A1234" }, "Hello World!"});

And you get: [{"timestamp":143321321321,"signature":"sha256 hash","id":"A1234"},"Hello World!"]
instead of: ["Token","Hello World!"]

@Jackyjjc
Copy link

Jackyjjc commented Dec 7, 2017

@kornman00 and @sotirosn mind putting your changes to https://github.com/Jackyjjc/MiniJSON.cs? That would be a great help!

@Baraphor
Copy link

At the top it says MiniJSON is best for strings less than 32K characters, so I should not use this for a file ranging around 300,000 characters, are there downsides to using this on larger files besides runtime?

@SunnyChowTheMadOne
Copy link

i load a json from server. but MiniJSON return null, and i feel it get problem in "Convert.ToChar(json.Peek())". What should i do if the string is not in Unicode?

@BestStream
Copy link

There are bug with double. Steps:

var dict = new Dictionary<string, object> { { "d", 0.000000001 } };
var test = Json.Serialize(dict);
dict = Json.Deserialize(test) as Dictionary<string, object>;

Fixed like this:

private object ParseNumber()
{
	string number = this.NextWord;

	if (number.IndexOf('.') >= 0 || number.IndexOf("E-") >= 0)
		return double.Parse(number, numberFormat);
	else
		return long.Parse(number, numberFormat);
}

@AnomalousUnderdog
Copy link

I forked @Jackyjjc's repo and incorporated even more of the mentioned fixes from here. I also added unit tests for them (https://github.com/AnomalousUnderdog/MiniJSON.cs)

@darktable
Copy link
Author

I should say I've stopped using MiniJSON in my projects in favor of a modified version of SimpleJSON which can be imported via package manager and includes unit tests: https://github.com/darktable/SimpleJSON

@Draco18s
Copy link

Draco18s commented Jun 9, 2021

Serializing an object turns the object into a string, which is then serialized.
The deserializer then deserializes a string and says "I'm done" and never turns it back into an object.

Test case:

struct Test {
	string name;
	int value;
	
	public override string ToString() {
		return "{\"name\":\"" + name + "\",\"value\":" + value + "}";
	}
}

Console.write(Json.Deserialize(Json.Serialize(new Test{ name = "Test", value = 10 })).GetType());

Expected: Dictionary<string,object>
Actual: string

@darktable
Copy link
Author

Serializing an object turns the object into a string, which is then serialized.
The deserializer then deserializes a string and says "I'm done" and never turns it back into an object.

Test case:

struct Test {
	string name;
	int value;
	
	public override string ToString() {
		return "{\"name\":" + name + ",\"value\":" + value + "}";
	}
}

Console.write(Json.Deserialize(Json.Serialize(new Test{ name = "Test", value = 10 })).GetType());

Expected: Dictionary<string,object>
Actual: string

Your test isn't outputting valid JSON:
{"name":Test,"value":10}
MiniJSON doesn't throw exceptions with invalid input, it is "garbage in, garbage out". This is because it's over 10 years old, from an era when uncaught exceptions would crash Unity iOS. Your life will be better with a more modern parser.

@Draco18s
Copy link

Draco18s commented Jun 9, 2021

Your test isn't outputting valid JSON:

Oops, I was typing up a simple example (rather than actual code I'm using) and missed the extra pair of ". My actual code doesn't have that typo. Updated.

Your life will be better with a more modern parser.

The life of a modder. miniJson ships with my target, so miniJson it is.

@CodeMasterYi
Copy link

CodeMasterYi commented Nov 1, 2023

There is a bug in some Culture like German.
In German, number 1.2345 is written like 1,2345, but the latter is not an valid JSON number.
in the MiniJSON.cs

void SerializeOther(object value) {
    // NOTE: decimals lose precision during serialization.
    // They always have, I'm just letting you know.
    // Previously floats and doubles lost precision too.
    if (value is float) {
        builder.Append(((float) value).ToString("R"));
    } else if (value is int
        || value is uint
        || value is long
        || value is sbyte
        || value is byte
        || value is short
        || value is ushort
        || value is ulong) {
        builder.Append(value);
    } else if (value is double
        || value is decimal) {
        builder.Append(Convert.ToDouble(value).ToString("R"));
    } else {
        SerializeString(value.ToString());
    }
}

It must be like this(use System.Globalization.CultureInfo.InvariantCulture):

void SerializeOther(object value) {
    // NOTE: decimals lose precision during serialization.
    // They always have, I'm just letting you know.
    // Previously floats and doubles lost precision too.
    if (value is float) {
        builder.Append(((float) value).ToString("R", System.Globalization.CultureInfo.InvariantCulture));
    } else if (value is int
        || value is uint
        || value is long
        || value is sbyte
        || value is byte
        || value is short
        || value is ushort
        || value is ulong) {
        builder.Append(value);
    } else if (value is double
        || value is decimal) {
        builder.Append(Convert.ToDouble(value).ToString("R", System.Globalization.CultureInfo.InvariantCulture));
    } else {
        SerializeString(value.ToString());
    }
}

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment