JSON parsing with Newtonsoft.JSON

Question

I have a file in JSON format with record for individual users. Some of the users have a comment field stuck in the middle of their records. I just want to parse top-level items ( fullName contributorName email)

using the Newtonsoft.JSON parser, but I can't seem to get it to recognize individual objects. When I parse the whole string into one big object, I don't know how to iterate the individual ones.

This is how I was trying to do it (property by property), but it didn't work if they were out of order or had subproperties. I need to put it into an object instead:

StreamReader re = File.OpenText("C:\dropbox\my dropbox\clients\towson\english 317\Ning Archive\ning-members.json"); JsonTextReader reader = new JsonTextReader(re); string ct = "";

        try
        {
            ct += "<table style='border:1px solid black'>";
            while (reader.Read())
            {

                if (reader.TokenType == JsonToken.PropertyName)
                {
                    if (reader.Value.ToString() == "fullName")
                    {

                        reader.Read();
                        ct += "\r\n\r\n<tr><td>" + reader.Value + "</td>";
                    }
                    if (reader.Value.ToString() == "contributorName")
                    {

                        reader.Read();
                        ct += "<td>" + reader.Value + "</td></tr>";
                    }
                    if (reader.Value.ToString() == "email")
                    {

                        reader.Read();
                        ct += "<td>" + reader.Value + "</td>";
                    }

                }





            }

        }
        catch { }
        ct+="</table>";
        namesText.Text = ct;

}

Note the first record has a comments field that I don't care about, but gets in the way of the order when I try to parse as a stream.

[
{
    "createdDate": "2010-09-10T14:16:08.271Z",
    "fullName": "Lisa Meloncon",
    "gender": "f",
    "country": "US",
    "birthdate": "1969-05-14",
    "comments": [
        {
            "id": "6292914:Comment:272",
            "contributorName": "0upfj0fd33932",
            "description": "Thanks for joining! I'm working up a schedule for the students a bit late so I can assess some of their early writing (including the first assignment, a general evaluation of business writing skills) and determine a course that will address their needs. I plan to make liberal use of technology this semester, with a Screencasting assignment, some intermediate Word formatting drills, and various other activities.",
            "createdDate": "2010-09-10T18:07:38.272Z"
        }
    ],
    "email": "[email protected]",
    "profilePhoto": "http://api.ning.com:80/files/251IcCtIBC3dGALHpG3ruYfg0Ip*EFJApPyMVGkiVArSUEvF*dK8A5grvPvl8eC7i7H0grhRH4pakLc9jSOww2GpU2OTq2nq/626617250.png?crop=1%3A1",
    "level": "member",
    "state": "active",
    "contributorName": "2z42css3dgvoi"
},
{
    "createdDate": "2010-09-08T02:57:00.225Z",
    "fullName": "Robert Calabrese",
    "gender": "m",
    "location": "Baltimore, MD",
    "country": "US",
    "zip": "21284",
    "birthdate": "1989-09-29",
    "email": "[email protected]",
    "profilePhoto": "http://api.ning.com:80/files/251IcCtIBC3dGALHpG3ruYfg0Ip*EFJApPyMVGkiVArSUEvF*dK8A5grvPvl8eC7i7H0grhRH4pakLc9jSOww2GpU2OTq2nq/626617250.png?crop=1%3A1",
    "level": "member",
    "state": "active",
    "contributorName": "199ru4hzwc4n4"
},
{
    "createdDate": "2010-09-04T22:36:51.158Z",
    "fullName": "Regis Bamba",
    "gender": "m",
    "location": "Baltimore, MD",
    "country": "US",
    "zip": "21210",
    "birthdate": "1986-09-29",
    "email": "rbamba2xxx.edu",
    "profilePhoto": "http://api.ning.com:80/files/251IcCtIBC3dGALHpG3ruYfg0Ip*EFJApPyMVGkiVArSUEvF*dK8A5grvPvl8eC7i7H0grhRH4pakLc9jSOww2GpU2OTq2nq/626617250.png?crop=1%3A1",
    "level": "member",
    "state": "active",
    "contributorName": "2seadgzt89n6x"
},

Answer 1

Something like:

JArray root = JArray.Load(reader);
foreach(JObject o in root)
{
    ct += "\r\n\r\n<tr><td>" + (string)o["fullName"] + "</td>";
    ct += "<td>" + (string)o["contributorName"] + "</td>";
    ct += "<td>" + (string)o["email"] + "</td>";
}

We use an explicit conversion to get a string value from the JToken returned by JObject.Item.

However, you should consider using StringBuilder, rather than concatenation, for performance.