Develop Reference

Transforming Microsoft Graph ListItem Output to a corresponding C# type

The work of transforming JSON data into a typed data model through seems to be made much more complex by the "help" the combination of SharePoint and MS Graph offer. :-)
I have a SharePoint List in Microsoft 365 that I'm accessing through the Graph API in C#, where the query destination is a typed class with properties identical to the SharePoint List Column Properties.
The ListItem class Graph API returns the results in the a Fields.AdditionalData of type Dictionary<string,object{System.Text.Json.JsonElement}> It needs to become an IEnumerable<DataItem>, which I can do by taking the List from the query result through a Serialize/Deserialize round trip, as below:
var backToJSON = ListItems.Select(o => System.Text.Json.JsonSerializer.Serialize(o.Fields.AdditionalData));
var stronglyTypedItems = backToJSON.Select(jsonO => System.Text.Json.JsonSerializer.Deserialize<DataItem>(jsonO));
Is there a way to do this, either with smarter OData or something in Graph API I haven't seen, without taking what used to be JSON and sending it back through JSON Serializers twice?
More details below:
Sample output JSON from Graph Explorer, where value contains an array of :
"value" : [
{ "id": "1001,
"fields": {
"Column" : "true",
"Column2" : "value2",
"Column3" : "65"
}
},
{ "id": "1002,
"fields": {
<and so forth until the array terminates>
]
}
Corresponding C# Class (literally built using "Paste JSON as class"):
Public class DataItem {
public bool Column {get; set;}
public string Column2 {get; set;}
public int Column3 {get; set;}
}
The "Helper" classes in the C# Graph API deliver mostly transformed into the array of fields I actually need:
private static GraphServiceClient graphClient;
public static IListItemsCollectionRequest LicenseExpirationsList => graphClient
.Sites["<guid>"]
.Lists["<nameOfList>"].Items
.Request()
.Header("Accept", "application/json;odata.metadata=none")
.Select("fields,id")
.Expand("fields");
var ListItems = (await GraphHelper.LicenseExpirationsList.GetAsync()).CurrentPage;
// JSON round tripping through JSONSerializer to get the strong type...
// But why? ListItems.Fields.AdditionalData is a Dictionary of JSON elements in the first place!
var backToJSON = ListItems.Select(o => System.Text.Json.JsonSerializer.Serialize(o.Fields.AdditionalData));
var stronglyTypedItems = backToJSON.Select(jsonO => System.Text.Json.JsonSerializer.Deserialize<DataItem>(jsonO));
return stronglyTypedItems;

You could customize the client's JSON serialization to return a derived type of default FieldValueSet.
First, define your own extended FieldValueSet:
public class FieldValueSetWithDataItem : FieldValueSet
{
public bool Column { get; set; }
public string Column2 { get; set; }
public int Column3 { get; set; }
}
Second, implement your own JSON converter:
class CustomFieldValueSetJsonConverter : JsonConverter<FieldValueSet>
{
private static readonly JsonEncodedText ODataTypeProperty
= JsonEncodedText.Encode("#odata.type");
private static readonly JsonEncodedText IdProperty
= JsonEncodedText.Encode("id");
private static readonly JsonEncodedText ColumnProperty
= JsonEncodedText.Encode("Column");
private static readonly JsonEncodedText Column2Property
= JsonEncodedText.Encode("Column2");
private static readonly JsonEncodedText Column3Property
= JsonEncodedText.Encode("Column3");
public override FieldValueSet Read(ref Utf8JsonReader reader,
Type typeToConvert, JsonSerializerOptions options)
{
var result = new FieldValueSetWithDataItem();
using var doc = JsonDocument.ParseValue(ref reader);
var root = doc.RootElement;
foreach (var element in root.EnumerateObject())
{
if (element.NameEquals(ODataTypeProperty.EncodedUtf8Bytes))
{
result.ODataType = element.Value.GetString();
}
else if (element.NameEquals(IdProperty.EncodedUtf8Bytes))
{
result.Id = element.Value.GetString();
}
else if (element.NameEquals(ColumnProperty.EncodedUtf8Bytes))
{
result.Column = element.Value.GetBoolean();
}
else if (element.NameEquals(Column2Property.EncodedUtf8Bytes))
{
result.Column2 = element.Value.GetString();
}
else if (element.NameEquals(Column3Property.EncodedUtf8Bytes))
{
result.Column3 = element.Value.GetInt32();
}
else
{
// Capture unknown property in AdditionalData
if (result.AdditionalData is null)
{
result.AdditionalData = new Dictionary<string, object>();
}
result.AdditionalData.Add(element.Name, element.Value.Clone());
}
}
return result;
}
public override void Write(Utf8JsonWriter writer,
FieldValueSet value, JsonSerializerOptions options)
{
// To support roundtrip serialization:
writer.WriteStartObject();
writer.WriteString(ODataTypeProperty, value.ODataType);
writer.WriteString(IdProperty, value.Id);
if (value is FieldValueSetWithDataItem dataItem)
{
writer.WriteBoolean(ColumnProperty, dataItem.Column);
writer.WriteString(Column2Property, dataItem.Column2);
writer.WriteNumber(Column3Property, dataItem.Column3);
}
if (value.AdditionalData is not null)
{
foreach (var kvp in value.AdditionalData)
{
writer.WritePropertyName(kvp.Key);
((JsonElement)kvp.Value).WriteTo(writer);
}
}
writer.WriteEndObject();
}
}
Last, use the JSON converter when making your request:
// Use custom JSON converter when deserializing response
var serializerOptions = new JsonSerializerOptions();
serializerOptions.Converters.Add(new CustomFieldValueSetJsonConverter());
var responseSerializer = new Serializer(serializerOptions);
var responseHandler = new ResponseHandler(responseSerializer);
var request = (ListItemsCollectionRequest)client.Sites[""].Lists[""].Items.Request();
var listItems = await request
.WithResponseHandler(responseHandler)
.GetAsync();
To access your column values:
var col3 = ((FieldValueSetWithDataItem)listItem.Fields).Column3;

You may find the HttpProvider of the GraphServiceClient helpful in this scenario:
var listItemsCollectionRequest = graphServiceClient
.Sites["<guid>"]
.Lists["<nameOfList>"]
.Items
.Request()
.Header("Accept", "application/json;odata.metadata=none")
.Select("fields,id")
.Expand("fields");
using (var requestMessage = listItemsCollectionRequest.GetHttpRequestMessage())
{
using var responseMessage = await graphServiceClient.HttpProvider.SendAsync(requestMessage);
//deserialize the response body into DataItem
}
By using the HttpProvider you can directly work with the response from the Graph API and deserialize the response body into your custom class.

Manipulating a Json string from within a custom Json Converter

In order to ask my question, I will be referring to #Brian Rogers 's answer here.
In this answer ReadJson is doing something relatively simple. How could I however add a layer in there in order to manipulate the incoming Json string before deserialising it into an object and then returning it?
Here is the type of things I would like to do (modified version of Brian's WrappedObjectConvert class):
class WrappedObjectConverter : JsonConverter
{
private string CustomParsing(string jsonString)
{
string modifiedJsonString;
// Some renaming
modifiedJsonString= Regex.Replace(modifiedJsonString, $#"(?<="")CarName(?="":\s)", "Myname", RegexOptions.IgnoreCase);
modifiedJsonString= Regex.Replace(modifiedJsonString, $#"(?<="")CustName(?="":\s)", "Myname", RegexOptions.IgnoreCase);
modifiedJsonString= Regex.Replace(modifiedJsonString, $#"(?<="")MyName(?="":\s)", "Myname", RegexOptions.IgnoreCase);
modifiedJsonString= Regex.Replace(modifiedJsonString, $#"(?<="")SomeAddr(?="":\s)", "AddressLine1 ", RegexOptions.IgnoreCase);
// Renaming IsPublic true/false to IsPrivate false/true
modifiedJsonString= Regex.Replace(modifiedJsonString, "\"IsPublic\": true,", "\"IsPrivate\": false,", RegexOptions.IgnoreCase);
modifiedJsonString = Regex.Replace(modifiedJsonString, "\"IsPublic\": false,", "\"IsPrivate\": true,", RegexOptions.IgnoreCase);
}
public override bool CanConvert(Type objectType)
{
return true;
}
public override object ReadJson(JsonReader reader, Type objectType, object existingValue, JsonSerializer serializer)
{
JToken token = JToken.Load(reader);
string modifiedJsonString = CustomParsing(token.ToString());
return ????; // How to return the object
// I could do something of the sort, but not sure it's got its place here:
// return JsonConvert.DeserializeObject<RootObject>(modifiedJsonString );
}
public override bool CanWrite
{
get { return false; }
}
public override void WriteJson(JsonWriter writer, object value, JsonSerializer serializer)
{
throw new NotImplementedException();
}
}
The client class has also been slightly modified by adding the field "IsPrivate":
public class Client
{
[JsonConverter(typeof(WrappedObjectConverter))]
public List<Product> ProductList { get; set; }
[JsonConverter(typeof(WrappedObjectConverter))]
public string Name { get; set; }
[JsonConverter(typeof(WrappedObjectConverter))]
public bool IsPrivate { get; set; }
[JsonConverter(typeof(WrappedObjectConverter))]
public string AddressLine1 { get; set; }
}
And the demo with a modified Json (some labels have been changed from Brian's example, which need to be parsed and modified):
class Program
{
static void Main(string[] args)
{
string json = #"
{
""Result"": {
""Client"": {
""ProductList"": {
""Product"": [
{
""MyName"": {
""CarName"": ""Car polish"",
""IsPublic"": ""True""
}
}
]
},
""MyName"": {
""CustName"": ""Mr. Clouseau""
},
""AddressLine1"": {
""SomeAddr"": ""Hightstreet 13""
}
}
}
}";
RootObject obj = JsonConvert.DeserializeObject<RootObject>(json);
Client client = obj.Result.Client;
foreach (Product product in client.ProductList)
{
Console.WriteLine(product.Name);
}
Console.WriteLine(client.Name);
Console.WriteLine(client.AddressLine1);
}
}
As you can see, the way the parsing is being done is a bit hacky, so my questions are:
Can I incorporate this parsing to the classes themselves without making a mess of my classes?
If my approach is the way to go, how do I recreate the object so that ReadJson() can return it (see
question marks in code above)
Taking it to the next level: If the client class had a constructor taking in arguments (passed to a base class), how would you do 2. (as the extra level of nesting would complicate things I believe)
If this is the wrong way to go, I am open to suggestions

From your question and comments it sounds like you have some complex JSON and your goal is to flatten it down into a simpler class structure. But you have some additional constraints:
The target classes for deserialization don't have default (parameterless) constructors.
You want to map the keys in the JSON to differently-named properties in the classes.
For some properties you want to translate the values as well (e.g. convert true to false and vice versa).
You want to keep the mapping logic all in one place, separate from the classes.
You can do all this with a custom JsonConverter. The key is to load the JSON data into a JObject inside the converter. From there you can use SelectToken to specify paths to retrieve specific pieces of data from the JObject. You can then use these pieces to construct your objects via their non-default constructors. At the same time you can translate any values that require it.
For example, let's say you are starting with the JSON in your question, and the classes you really want to deserialize to look like this:
public class Client
{
public Client(string name, string addressLine1, List<Product> productList)
{
Name = name;
AddressLine1 = addressLine1;
ProductList = productList;
}
public List<Product> ProductList { get; set; }
public string Name { get; set; }
public string AddressLine1 { get; set; }
}
public class Product
{
public Product(string name, bool isPrivate)
{
Name = name;
IsPrivate = isPrivate;
}
public string Name { get; set; }
public bool IsPrivate { get; set; }
}
Here is a custom converter that will handle the deserialization:
class CustomConverter : JsonConverter
{
public override bool CanConvert(Type objectType)
{
return objectType == typeof(Client);
}
public override object ReadJson(JsonReader reader, Type objectType, object existingValue, JsonSerializer serializer)
{
JObject obj = JObject.Load(reader);
string name = (string)obj.SelectToken("Result.Client.MyName.CustName");
string addressLine1 = (string)obj.SelectToken("Result.Client.AddressLine1.SomeAddr");
List<Product> productList = obj.SelectToken("Result.Client.ProductList.Product")
.Select(jt =>
{
string prodName = (string)jt.SelectToken("MyName.CarName");
bool isPublic = string.Equals((string)jt.SelectToken("MyName.IsPublic"), "True", StringComparison.OrdinalIgnoreCase);
return new Product(prodName, !isPublic);
})
.ToList();
Client client = new Client(name, addressLine1, productList);
return client;
}
public override bool CanWrite => false;
public override void WriteJson(JsonWriter writer, object value, JsonSerializer serializer)
{
throw new NotImplementedException();
}
}
To use the converter, you can either add a [JsonConverter] attribute to the Client class like this:
[JsonConverter(typeof(CustomConverter))]
public class Client
{
...
}
Or you can pass the converter as a parameter to JsonConvert.DeserializeObject() like this:
Client client = JsonConvert.DeserializeObject<Client>(json, new CustomConverter());
Here is a working demo: https://dotnetfiddle.net/EwtQHh

How can I parse json to a C# object (Class)?

I have JSON data that I need to parse from C# object.
this is JSON Example.
{
"types":
[
[
"tour_type",
[
["groups",1],
["individual",2]
]
]
]
}
Here are my C# classes that are meant to contain that data:
using System;
using Newtonsoft.Json;
namespace JsonDeserializationTest
{
[JsonProperty("types")]
public class Types
{
[JsonProperty]
public List<Type> Values {get;set;}
}
public class Type
{
[JsonProperty]
public string Key {get;set;}
[JsonProperty]
public List<Dictionary<string, int>> Values { get; set; }
}
}
It's not working now.
How can I fix it?

Use the JsonSerializer (System.Text.Json) object.
Code:
YourClass obj = JsonSerializer.Deserialize<YourClass>(jsonString);

Your json has a list of list of the object... but you are declaring only List of the object.
public class Types
{
[JsonProperty("types")]
public List<List<object>> Values { get; set; }
// ------ UPDATE: This can only be list of list of 'object' ------- \\
}
Also, you are using the JsonProperty on the class, which is not where that normally goes. You want to use that on the property of the class.
UPDATE:
You cannot use List<List<Type>> for the json you are getting, it can only be List<List<object>>. You have to use object because it can either be a string or a List<List<string>>. After you update your Types class, you can successfully deserialize the json above.
var obj = JsonConvert.DeserializeObject<Types>(json);
and based on your json definition, you can access tour_type by using the following code
types.Values.First()[0].ToString()
// output: tour_type
List<List<string>> data = JsonConvert.DeserializeObject<List<List<string>>>(types.Values.First()[1].ToString())
// data[0]
[0]: "groups"
[1]: "1"
// data[1]
[0]: "individual"
[1]: "2"
Since both of the items in the types are objects, you will either have to convert them to string or a list of list of strings or whatever object they actually are.

The JSON payload in the provided example is formatted quite strangely, especially since it contains seemingly unnecessary array nesting. A payload like this usually includes more nested objects (rather than a bunch of nested arrays). Additionally, it has a list of (string, int) pairs, which is semantically very similar to a Dictionary<string, int>, but the payload doesn't lend itself to that. It would be helpful to know where it is coming from (what context) to understand how it might change.
The example JSON brings up a few questions (that you may want to ask yourself):
Can the "types" array contain multiple entries (at its immediate nesting)?
Can the "tour_type" key name appear after the array of string, int pairs? Is it possible for an entry where no such name exists?
What other elements can exist in the arrays within "tour_type"?
Is it guaranteed that the most nested array will contain just a single (string, int) pair?
Similarly, it is hard to understand what the example C# class is trying to encapsulate. Is List<Dictionary<string, int>> necessary?
All that said, here's a solution using the built-in System.Text.Json library, that could work for you. You could write something similar using Newtonsoft.Json, if necessary. The solution assumes:
We can't change the JSON payload (and that the third party API response will always returns something that is structurally similar to the example)
We can only make minimal changes to the C# class object provided in the example
The solution creates and a JsonConverter<T> that uses the low-level Utf8JsonReader to manually parse and create the custom object. This is required since nested "[" are being used to delineate what should be objects rather than "{". The converter is then registered by annotating the class with the attribute. Now, simply call JsonSerializer.Deserialize, passing in the JSON payload.
public class Tours
{
[JsonPropertyName("types")]
public List<UserType> Types { get; set; }
}
// Annotate the type to register the converter to use
[JsonConverter(typeof(CustomUserTypeConverter))]
public class UserType
{
public string Key { get; set; }
public Dictionary<string, int> Values { get; set; }
}
// This will use the low-level reader to build up the UserType
public class CustomUserTypeConverter : JsonConverter<UserType>
{
// Extra structural validation was done for invalid/incomplete JSON
// which might be too strict or incorrect and hence might require adjustments.
public override UserType Read(ref Utf8JsonReader reader, Type typeToConvert, JsonSerializerOptions options)
{
var result = new UserType();
if (!reader.Read())
{
throw new JsonException("Incomplete JSON.");
}
if (reader.TokenType != JsonTokenType.EndArray)
{
result.Key = reader.GetString();
ReadAndValidate(ref reader, JsonTokenType.StartArray);
int depthSnapshot = reader.CurrentDepth;
var values = new Dictionary<string, int>();
do
{
reader.Read();
if (reader.TokenType != JsonTokenType.StartArray && reader.TokenType != JsonTokenType.EndArray)
{
throw new JsonException($"Invalid JSON payload. Expected Start or End Array. TokenType: {reader.TokenType}, Depth: {reader.CurrentDepth}.");
}
if (reader.CurrentDepth <= depthSnapshot)
{
break;
}
reader.Read();
if (reader.TokenType != JsonTokenType.EndArray)
{
string key = reader.GetString();
reader.Read();
int value = reader.GetInt32();
values.Add(key, value);
ReadAndValidate(ref reader, JsonTokenType.EndArray);
}
} while (true);
ReadAndValidate(ref reader, JsonTokenType.EndArray);
result.Values = values;
}
return result;
}
private void ReadAndValidate(ref Utf8JsonReader reader, JsonTokenType expectedTokenType)
{
bool readNext = reader.Read();
if (!readNext || reader.TokenType != expectedTokenType)
{
string message = readNext ?
$"Invalid JSON payload. TokenType: {reader.TokenType}, Depth: {reader.CurrentDepth}, Expected: {expectedTokenType}" :
$"Incomplete JSON. Expected: {expectedTokenType}";
throw new JsonException(message);
}
}
// Implement this method if you need to Serialize (i.e. write) the object
// back to JSON
public override void Write(Utf8JsonWriter writer, UserType value, JsonSerializerOptions options)
{
throw new NotImplementedException();
}
}
Here's how you would use the above converter to serialize the JSON string provided in the example, along with how to access the values.
public static Tours ParseJson(string json)
{
Tours tours = JsonSerializer.Deserialize<Tours>(json);
return tours;
}
public static void AccessValues(Tours tours)
{
foreach (UserType data in tours.Types)
{
string typeName = data.Key; // "tour_type"
foreach (KeyValuePair<string, int> pairs in data.Values)
{
string key = pairs.Key; // "groups" or "individual
int value = pairs.Value; // 1 or 2
}
}
}
For what it's worth, Visual Studio suggests the following C# class structure for the example JSON (which is similar to what #Jawad suggested):
public class Rootobject
{
public object[][] types { get; set; }
}
Hope that helps.

I couldn't figure out your JSON so I created an example with verified JSON.
Try this:
JSON:
{
"Items": [
{
"Name": "tour",
"Attributes": [
{
"Name": "groups",
"Value": 1
},
{
"Name": "individual",
"Value": 2
}
]
},
{
"Name": "demo",
"Attributes": [
{
"Name": "this is demo",
"Value": 3
},
{
"Name": "design pattern",
"Value": 99
}
]
}
]
}
Types foo = JsonSerializer.Deserialize<Types>(jsonString);
public class TypeAttribute
{
public string Name { get; set; }
public int Value { get; set; }
}
public class Type
{
private readonly ICollection<TypeAttribute> _attributes;
public Type()
{
_attributes = new Collection<TypeAttribute>();
}
public void AddAttributes(IEnumerable<TypeAttribute> attrs)
{
foreach(TypeAttribute ta in attrs)
{
_attributes.Add(ta);
}
}
public string Name { get; set; }
public IEnumerable<TypeAttribute> Attributes
{
get { return _attributes; }
set
{
foreach(TypeAttribute ta in value)
{
_attributes.Add(ta);
}
}
}
}
public class Types
{
ICollection<Type> _items;
public Types()
{
_items = new Collection<Type>();
}
public void AddItems(IEnumerable<Type> tps)
{
foreach (Type t in tps)
{
_items.Add(t);
}
}
public IEnumerable<Type> Items
{
get { return _items; }
set
{
foreach (Type t in value)
{
_items.Add(t);
}
}
}
}

JSON .NET Custom Name Resolver for Sub-Properties

I have an API that returns a JSON object from a MongoDB in which one of the properties is an "open-ended" document, meaning it can be any valid JSON for that property. I don't know what the names of the properties are ahead of time, they can be any string. I only know that this particular property needs to be serialized exactly how it is stored in the database. So if the property name that was originally stored was "Someproperty", the serialized response in JSON needs to be "Someproperty", NOT "someProperty".
We have this configuration:
ContractResolver = new CamelCasePropertyNamesContractResolver();
in our CustomJsonSerializer, but it is messing with the formatting of the response when returning the "open ended" JSON. It is camel-casing all of these properties when in fact we want the response to be exactly how they are stored in MongoDB (BSON). I know the values are maintaining their case when storing/retrieving via the database, so that is not the issue.
How can I tell JSON.net to essentially bypass the CamelCasePropertyNameResolver for all of the child properties of a particular data point?
EDIT:
Just to give a bit more info, and share what I have already tried:
I thought about overriding the PropertyNameResolver like so:
protected override string ResolvePropertyName(string propertyName)
{
if (propertyName.ToLower().Equals("somedocument"))
{
return propertyName;
}
else return base.ResolvePropertyName(propertyName);
}
However, if I have a JSON structure like this:
{
"Name" : "MyObject",
"DateCreated" : "11/14/2016",
"SomeDocument" :
{
"MyFirstProperty" : "foo",
"mysecondPROPERTY" : "bar",
"another_random_subdoc" :
{
"evenmoredata" : "morestuff"
}
}
}
then I would need all of the properties any child properties' names to remain exactly as is. The above override I posted would (I believe) only ignore on an exact match to "somedocument", and would still camelcase all of the child properties.

What you can do is, for the property in question, create a custom JsonConverter that serializes the property value in question using a different JsonSerializer created with a different contract resolver, like so:
public class AlternateContractResolverConverter : JsonConverter
{
[ThreadStatic]
static Stack<Type> contractResolverTypeStack;
static Stack<Type> ContractResolverTypeStack { get { return contractResolverTypeStack = (contractResolverTypeStack ?? new Stack<Type>()); } }
readonly IContractResolver resolver;
JsonSerializerSettings ExtractAndOverrideSettings(JsonSerializer serializer)
{
var settings = serializer.ExtractSettings();
settings.ContractResolver = resolver;
settings.CheckAdditionalContent = false;
if (settings.PreserveReferencesHandling != PreserveReferencesHandling.None)
{
// Log an error throw an exception?
Debug.WriteLine(string.Format("PreserveReferencesHandling.{0} not supported", serializer.PreserveReferencesHandling));
}
return settings;
}
public AlternateContractResolverConverter(Type resolverType)
{
if (resolverType == null)
throw new ArgumentNullException("resolverType");
resolver = (IContractResolver)Activator.CreateInstance(resolverType);
if (resolver == null)
throw new ArgumentNullException(string.Format("Resolver type {0} not found", resolverType));
}
public override bool CanRead { get { return ContractResolverTypeStack.Count == 0 || ContractResolverTypeStack.Peek() != resolver.GetType(); } }
public override bool CanWrite { get { return ContractResolverTypeStack.Count == 0 || ContractResolverTypeStack.Peek() != resolver.GetType(); } }
public override bool CanConvert(Type objectType)
{
throw new NotImplementedException("This contract resolver is intended to be applied directly with [JsonConverter(typeof(AlternateContractResolverConverter), typeof(SomeContractResolver))] or [JsonProperty(ItemConverterType = typeof(AlternateContractResolverConverter), ItemConverterParameters = ...)]");
}
public override object ReadJson(JsonReader reader, Type objectType, object existingValue, JsonSerializer serializer)
{
using (ContractResolverTypeStack.PushUsing(resolver.GetType()))
return JsonSerializer.CreateDefault(ExtractAndOverrideSettings(serializer)).Deserialize(reader, objectType);
}
public override void WriteJson(JsonWriter writer, object value, JsonSerializer serializer)
{
using (ContractResolverTypeStack.PushUsing(resolver.GetType()))
JsonSerializer.CreateDefault(ExtractAndOverrideSettings(serializer)).Serialize(writer, value);
}
}
internal static class JsonSerializerExtensions
{
public static JsonSerializerSettings ExtractSettings(this JsonSerializer serializer)
{
// There is no built-in API to extract the settings from a JsonSerializer back into JsonSerializerSettings,
// so we have to fake it here.
if (serializer == null)
throw new ArgumentNullException("serializer");
var settings = new JsonSerializerSettings
{
CheckAdditionalContent = serializer.CheckAdditionalContent,
ConstructorHandling = serializer.ConstructorHandling,
ContractResolver = serializer.ContractResolver,
Converters = serializer.Converters,
Context = serializer.Context,
Culture = serializer.Culture,
DateFormatHandling = serializer.DateFormatHandling,
DateFormatString = serializer.DateFormatString,
DateParseHandling = serializer.DateParseHandling,
DateTimeZoneHandling = serializer.DateTimeZoneHandling,
DefaultValueHandling = serializer.DefaultValueHandling,
EqualityComparer = serializer.EqualityComparer,
// No Get access to the error event, so it cannot be copied.
// Error = += serializer.Error
FloatFormatHandling = serializer.FloatFormatHandling,
FloatParseHandling = serializer.FloatParseHandling,
Formatting = serializer.Formatting,
MaxDepth = serializer.MaxDepth,
MetadataPropertyHandling = serializer.MetadataPropertyHandling,
MissingMemberHandling = serializer.MissingMemberHandling,
NullValueHandling = serializer.NullValueHandling,
ObjectCreationHandling = serializer.ObjectCreationHandling,
ReferenceLoopHandling = serializer.ReferenceLoopHandling,
// Copying the reference resolver doesn't work in the default case, since the
// actual BidirectionalDictionary<string, object> mappings are held in the
// JsonSerializerInternalBase.
// See https://github.com/JamesNK/Newtonsoft.Json/blob/master/Src/Newtonsoft.Json/Serialization/DefaultReferenceResolver.cs
ReferenceResolverProvider = () => serializer.ReferenceResolver,
PreserveReferencesHandling = serializer.PreserveReferencesHandling,
StringEscapeHandling = serializer.StringEscapeHandling,
TraceWriter = serializer.TraceWriter,
TypeNameHandling = serializer.TypeNameHandling,
// Changes in Json.NET 10.0.1
//TypeNameAssemblyFormat was obsoleted and replaced with TypeNameAssemblyFormatHandling in Json.NET 10.0.1
//TypeNameAssemblyFormat = serializer.TypeNameAssemblyFormat,
TypeNameAssemblyFormatHandling = serializer.TypeNameAssemblyFormatHandling,
//Binder was obsoleted and replaced with SerializationBinder in Json.NET 10.0.1
//Binder = serializer.Binder,
SerializationBinder = serializer.SerializationBinder,
};
return settings;
}
}
public static class StackExtensions
{
public struct PushValue<T> : IDisposable
{
readonly Stack<T> stack;
public PushValue(T value, Stack<T> stack)
{
this.stack = stack;
stack.Push(value);
}
// By using a disposable struct we avoid the overhead of allocating and freeing an instance of a finalizable class.
public void Dispose()
{
if (stack != null)
stack.Pop();
}
}
public static PushValue<T> PushUsing<T>(this Stack<T> stack, T value)
{
if (stack == null)
throw new ArgumentNullException();
return new PushValue<T>(value, stack);
}
}
Then use it like so:
public class RootObject
{
public string Name { get; set; }
public DateTime DateCreated { get; set; }
[JsonProperty(NamingStrategyType = typeof(DefaultNamingStrategy))]
[JsonConverter(typeof(AlternateContractResolverConverter), typeof(DefaultContractResolver))]
public SomeDocument SomeDocument { get; set; }
}
public class SomeDocument
{
public string MyFirstProperty { get; set; }
public string mysecondPROPERTY { get; set; }
public AnotherRandomSubdoc another_random_subdoc { get; set; }
}
public class AnotherRandomSubdoc
{
public string evenmoredata { get; set; }
public DateTime DateCreated { get; set; }
}
(Here I am assuming you want the "SomeDocument" property name to be serialized verbatim, even though it wasn't entirely clear from your question. To do that, I'm using JsonPropertyAttribute.NamingStrategyType from Json.NET 9.0.1. If you're using an earlier version, you'll need to set the property name explicitly.)
Then the resulting JSON will be:
{
"name": "Question 40597532",
"dateCreated": "2016-11-14T05:00:00Z",
"SomeDocument": {
"MyFirstProperty": "my first property",
"mysecondPROPERTY": "my second property",
"another_random_subdoc": {
"evenmoredata": "even more data",
"DateCreated": "2016-11-14T05:00:00Z"
}
}
}
Note that this solution does NOT work well with preserving object references. If you need them to work together, you may need to consider a stack-based approach similar to the one from Json.NET serialize by depth and attribute
Demo fiddle here.
Incidentally, have you considered storing this JSON as a raw string literal, as in the answer to this question?

I think you should look at this backwards.
Instead of trying to NOT touch the properties you don't know, let that be the default behavior and touch the ones you DO know.
In other words, don't use the CamelCasePropertyNamesContractResolver. Deal with the properties you know appropriately and let the other ones pass through transparently.

Custom JSON Derivative Format

I would like to have a serialization format that is nearly identical to JSON, except that key-values are represented as <key>="<value>" instead of "<key>":"<value>".
With Newtonsoft I made a custom JsonConverter called TsonConverter that works fairly well, except that it can't "see" an embedded dictionary. Given the following type:
public class TraceyData
{
[Safe]
public string Application { get; set; }
[Safe]
public string SessionID { get; set; }
[Safe]
public string TraceID { get; set; }
[Safe]
public string Workflow { get; set; }
[Safe]
public Dictionary<string, string> Tags {get; set; }
[Safe]
public string[] Stuff {get; set;}
}
And the following code:
TsonConverter weird = new TsonConverter();
JsonSerializerSettings settings = new JsonSerializerSettings();
settings.NullValueHandling = NullValueHandling.Ignore;
settings.Converters.Add(weird);
var tracey = new TraceyData();
tracey.TraceID = Guid.NewGuid().ToString();
tracey.SessionID = "5";
tracey.Tags["Referrer"] = "http://www.sky.net/deals";
tracey.Stuff = new string[] { "Alpha", "Bravo", "Charlie" };
tracey.Application = "Responsive";
string stuff = JsonConvert.SerializeObject(tracey, settings);
I get this:
[Application="Responsive" SessionID="5" TraceID="082ef853-92f8-4ce8-9f32-8e4f792fb022" Tags={"Referrer":"http://www.sky.net/deals"} Stuff=["Alpha","Bravo","Charlie"]]
Obviously I have also overridden the StartObject/EndObject notation, replacing { } with [ ]. Otherwise the results are not bad.
However, there is still the problem of the internal dictionary. In order
to convert the dictionary as well to use my <key>="<value>" format, it looks like I must make a deep dictionary converter.
I'm wondering if there is an easier way to do this.
Perhaps the Newtonsoft tool has a "property generator" and "key-value" generator property that I can set that globally handles this for me?
Any suggestions?
And while we're here, I wonder if there is a StartObject/EndObject formatter property override I can set, which would handle the other customization I've shown above. It would be nice to "skip" making JsonConverter tools for these kinds of simple alterations.
Incidentally:
My custom JsonConverter is choosing properties to serialize based on the [Safe] attribute shown in my sample. This is another nice-to-have. It would be wonderful if the JSon settings could expose an "attribute handler" property that lets me override the usual JSon attributes in favor of my own.
I have no need to de-serialize this format. It is intended as a one-way operation. If someone wishes also to explain how to de-serialize my custom format as well that is an interesting bonus, but definitely not necessary to answer this question.
Appendix
Below is the TraceConverter I had made. It references a FieldMetaData class that simply holds property info.
public class TsonConverter : JsonConverter
{
public override bool CanRead
{
get
{
return false;
}
}
public override bool CanConvert(Type ObjectType)
{
return DataClassifier.TestForUserType(ObjectType);
}
public override void WriteJson(
JsonWriter writer, object value, JsonSerializer serializer)
{
Type objType = value.GetType();
var props = objType.GetProperties(BindingFlags.Instance | BindingFlags.Public | BindingFlags.NonPublic);
var propMap = from p in props
from a in p.GetCustomAttributes(typeof(ProfileAttribute), false)
select new FieldMetaData(p, (ProfileAttribute)a);
//writer.WriteStartObject();
writer.WriteStartArray();
bool loopStarted = true;
foreach(var prop in propMap){
object rawValue = prop.GetValue(value);
if (rawValue != null || serializer.NullValueHandling == NullValueHandling.Include)
{
string jsonValue = JsonConvert.SerializeObject(prop.GetValue(value), this);
if (loopStarted)
{
loopStarted = false;
writer.WriteRaw(String.Format("{0}={1}", prop.Name, jsonValue));
}
else
{
writer.WriteRaw(String.Format(" {0}={1}", prop.Name, jsonValue));
}
}
//writer.WriteRaw(String.Format("{0}={1}", prop.Name, prop.GetValue(value)));
//writer.WritePropertyName(prop.Name, false);
//writer.WriteValue(prop.GetValue(value));
}
writer.WriteEndArray();
}
public override object ReadJson(
JsonReader reader, Type objectType,
object existingValue, JsonSerializer serializer)
{
throw new NotImplementedException();
}
}

Rather than creating your own converter, you're going to need to create your own subclass of JsonWriter that writes to your custom file format. (This is how Json.NET implements its BsonWriter.) In your case, your file format is close enough to JSON that you can inherit from JsonTextWriter:
public class TsonTextWriter : JsonTextWriter
{
TextWriter _writer;
public TsonTextWriter(TextWriter textWriter)
: base(textWriter)
{
if (textWriter == null)
throw new ArgumentNullException("textWriter");
QuoteName = false;
_writer = textWriter;
}
public override void WriteStartObject()
{
SetWriteState(JsonToken.StartObject, null);
_writer.Write('[');
}
protected override void WriteEnd(JsonToken token)
{
switch (token)
{
case JsonToken.EndObject:
_writer.Write(']');
break;
default:
base.WriteEnd(token);
break;
}
}
public override void WritePropertyName(string name)
{
WritePropertyName(name, true);
}
public override void WritePropertyName(string name, bool escape)
{
SetWriteState(JsonToken.PropertyName, name);
var escaped = name;
if (escape)
{
escaped = JsonConvert.ToString(name, '"', StringEscapeHandling);
escaped = escaped.Substring(1, escaped.Length - 2);
}
// Maybe also escape the space character if it appears in a name?
_writer.Write(escaped.Replace("=", #"\u003d"));// Replace "=" with unicode escape sequence.
_writer.Write('=');
}
/// <summary>
/// Writes the JSON value delimiter. (Remove this override if you want to retain the comma separator.)
/// </summary>
protected override void WriteValueDelimiter()
{
_writer.Write(' ');
}
/// <summary>
/// Writes an indent space.
/// </summary>
protected override void WriteIndentSpace()
{
// Do nothing.
}
}
Having done this, now all classes will be serialized to your custom format when you use this writer, for instance:
var tracey = new TraceyData();
tracey.TraceID = Guid.NewGuid().ToString();
tracey.SessionID = "5";
tracey.Tags["Referrer"] = "http://www.sky.net/deals";
tracey.Stuff = new string[] { "Alpha", "Bravo", "Charlie" };
tracey.Application = "Responsive";
JsonSerializerSettings settings = new JsonSerializerSettings();
settings.NullValueHandling = NullValueHandling.Ignore;
using (var sw = new StringWriter())
{
using (var jsonWriter = new TsonTextWriter(sw))
{
JsonSerializer.CreateDefault(settings).Serialize(jsonWriter, tracey);
}
Debug.WriteLine(sw.ToString());
}
Produces the output
[Application="Responsive" SessionID="5" TraceID="2437fe67-9788-47ba-91ce-2e5b670c2a34" Tags=[Referrer="http://www.sky.net/deals"] Stuff=["Alpha" "Bravo" "Charlie"]]
As far as deciding whether to serialize properties based on the presence of a [Safe] attribute, that's sort of a second question. You will need to create your own ContractResolver and override CreateProperty, for instance as shown here: Using JSON.net, how do I prevent serializing properties of a derived class, when used in a base class context?
Update
If you want to retain the comma separator for arrays but not objects, modify WriteValueDelimiter as follows:
/// <summary>
/// Writes the JSON value delimiter. (Remove this override if you want to retain the comma separator.)
/// </summary>
protected override void WriteValueDelimiter()
{
if (WriteState == WriteState.Array)
_writer.Write(',');
else
_writer.Write(' ');
}