ToJson()无法对特殊字符进行正确的转译

[mnonmcn]

            JsonData jd = new JsonData();
            jd["name"] = "\u0003";
            string json = jd.ToJson();
            File.WriteAllText("c:\\test.json", json);

ToJson () can't translate special characters correctly, which inverts the string of ToJson () data in browsers such as Chorme.
JSON. parse (json) cannot be parsed correctly
Expected results：{"name":"\u0003"}
Actual results：{"name": "�"}

Sorry, it's automatically translated with a translator.

ToJson()无法对特殊字符进行正确的转译,这倒置ToJson() 数据的字符串在Chorme等浏览器中 JSON.parse(json)不能正确解析
正确的结果:{"name":"\u0003"}
实测的结果:{"name": "�"}

[devlead]

Which version of LitJson har you using?

Did a quick .NET Fiddle ( https://dotnetfiddle.net/3cGvQN ) and it outputs {"name":"\u0003"}, which you write as exeected result? Isn't this the expected result?

Sending that json to json lint says it's valid json ( https://jsonlint.com/?json={%22name%22:%22\u0003%22} )

Created an jsfiddle ( https://jsfiddle.net/nmecws5x/ ) and it parses the outputed json fine

So unsure what's expected and unexpected.

[mnonmcn]

            JsonData jd = new JsonData();
            jd["name"] = "\u0003";
            jd["key"] = "中文";
            string json = jd.ToJson();
            Console.WriteLine(json);

The results are as follows: {"name":"\u0003","key":"\u4E2D\u6587"}

Sorry, I used an old version. I upgraded it to the latest version, but because of the particularity of Chinese, it converted all Chinese into unicode, which caused a serious decline in readability. In fact, Chinese does not need to be re-coded.

---

结果为：{"name":"\u0003","key":"\u4E2D\u6587"}
很抱歉，我使用了一个很古老的版本，我将它升级到了最新版本后可以了，但是因为中文的特殊性，它将中文全部也转换成了 unicode ,这造成了可读性严重下降，实际上，中文是不需要重新编码的

[devlead]

@mnonmcn ok, but there's no breaking bug then? But a feature request to not encode certain characters?

[mnonmcn]

                //对中文的支持
                //https://www.qqxiuzi.cn/zh/hanzi-unicode-bianma.php
                //暂时只对常用的基本汉字进行支持	20902字
                if ((int)str[i] >= 0X4E00 && (int)str[i] <= 0X9FA5)
                {
                    writer.Write(str[i]);
                    continue;
                }

I support the code in Chinese after line 269 of the JsonWriter.cs file. enhance readability,This solves another problem at the same time.https://github.com/LitJSON/litjson/issues/78

我对代码进行了对中文的支持，放在JsonWriter.cs文件的第 269 行后，增强可读性,这同时解决了另外一个问题https://github.com/LitJSON/litjson/issues/78

[devlead]

I see, would it be possible for you to send a pull request with this change?

[nchetan-zz]

Is this still open? I found this from https://up-for-grabs.net/#/filters?labels=88&tags=c%23 and was about to consider this as an option to contribute to (assuming it's a good first issue on this).

[salyu9]

Currently the escaping strategy of LitJson that escapes all non-ascii characters is actually fine. The problem is that people may want those graphic characters to be kept, not only chinese characters, so imho excluding the range [0x4e00, 0x9fa5] (more precisely, CJK Unified Ideographs) is not the solution.
According to RFC 4627: All Unicode characters may be placed within the quotation marks except for the characters that must be escaped: quotation mark, reverse solidus, and the control characters (U+0000 through U+001F). Maybe a switch can be added for just escape these characters, like the StringEscapeHandling.Default/EscapeNonAscii in Newtonsoft.Json. And I thought this may better be a feature request.