Dictionary serialization with DataContractSerializer generate below data. How to replace d2p1:KeyValueOfintint, d2p1:Key and d2p1:Value with our own attributes / tags / identifier.
Serializing Dictionary in [CashCounter],
Output Generate after Serialization given below
<CashCounter xmlns:i="http://www.w3.org/2001/XMLSchema-instance" xmlns="http://schemas.datacontract.org/2004/07/DictionarySerlization">
<BankNote xmlns:d2p1="http://schemas.microsoft.com/2003/10/Serialization/Arrays">
<d2p1:KeyValueOfintint>
<d2p1:Key>10</d2p1:Key>
<d2p1:Value>6</d2p1:Value>
</d2p1:KeyValueOfintint>
<d2p1:KeyValueOfintint>
<d2p1:Key>5</d2p1:Key>
<d2p1:Value>10</d2p1:Value>
</d2p1:KeyValueOfintint>
</BankNote>
<TotalCount>16</TotalCount>
<TotalSum>110</TotalSum>
</CashCounter>
static void Main(string[] args) {
CashCounter cashCounter = addCashCounter();
string serilizedData = GetXml(cashCounter);
}
private static CashCounter addCashCounter() {
CashCounter cashCounter = CreateCounter();
for(var i = 0; i < 6; i++) { cashCounter = incrementCountAmount(cashCounter, 10); }
for(var i = 0; i < 10; i++) { cashCounter = incrementCountAmount(cashCounter, 5); }
return cashCounter;
}
private static CashCounter CreateCounter()
{
var cashCounter = new CashCounter
{
BanknotesCount = new Dictionary<int, int>(),
TotalSum = 0,
TotalCount = 0
};
return cashCounter;
}
private static CashCounter incrementCountAmount(CashCounter cashCounter, int amount){
const int count = 1;
cashCounter.TotalCount += count;
cashCounter.TotalSum += amount * count;
if (cashCounter.BanknotesCount.ContainsKey(amount))
{
cashCounter.BanknotesCount[amount] += count;
}
else
{
cashCounter.BanknotesCount.Add(amount, count);
}
return cashCounter;
}
public static string GetXml<T>(T obj, DataContractSerializer serializer)
{
using (var textWriter = new StringWriter())
{
var settings = new XmlWriterSettings {OmitXmlDeclaration = true,Indent = true, IndentChars = " " };
using (var xmlWriter = XmlWriter.Create(textWriter, settings))
{
serializer.WriteObject(xmlWriter, obj);
}
return textWriter.ToString();
}
}
public static string GetXml<T>(T obj)
{
var serializer = new DataContractSerializer(typeof(T));
return GetXml(obj, serializer);
}
You can control the item, key and value element names of a dictionary when serialized to XML by subclassing Dictionary<TKey, TValue>, applying CollectionDataContractAttribute, and setting the following attribute values:
ItemName: gets or sets a custom name for a dictionary key/value pair element.
KeyName: gets or sets a custom name for a dictionary key name element.
ValueName: gets or sets a custom name for a dictionary value element.
Namespace: if needed, gets or sets a namespace for the data contract.
Name: if needed, gets or sets the data contract name for the dictionary type. This becomes the XML root element name when the dictionary is serialized as the root object.
(Since the dictionary is not the root object in your data model, setting this particular property is not needed in this case.)
Thus, if you define your CashCounter data model as follows (simplified to remove irrelevant members):
[DataContract(Namespace = "http://schemas.datacontract.org/2004/07/DictionarySerlization")]
public class CashCounter
{
[DataMember]
public BankNoteDictionary BankNote { get; set; }
}
[CollectionDataContract(ItemName = "MyItemName", KeyName = "MyKeyName", ValueName = "MyValueName",
Namespace = "http://schemas.datacontract.org/2004/07/DictionarySerlization")]
public class BankNoteDictionary : Dictionary<int, int>
{
}
The resulting XML will look like:
<CashCounter xmlns:i="http://www.w3.org/2001/XMLSchema-instance" xmlns="http://schemas.datacontract.org/2004/07/DictionarySerlization">
<BankNote>
<MyItemName>
<MyKeyName>10</MyKeyName>
<MyValueName>6</MyValueName>
</MyItemName>
<MyItemName>
<MyKeyName>5</MyKeyName>
<MyValueName>10</MyValueName>
</MyItemName>
</BankNote>
</CashCounter>
Notes:
You should consider replacing the namespace with something permanent.
The namespace is part of the data contract name. Currently it has some default value related to your c# namespace, so if you refactor your code and put classes into different c# namespaces, the data contract namespace might change. Since the data contract namespace is actually published in your WSDL that could be a nuisance for your customers. You may also want to use the namespace for branding purposes by including your organization's URL at the beginning.
For further reading see What does adding Name and Namespace to DataContract do? and What are XML namespaces for?.
For documentation see Collection Types in Data Contracts: Customizing Dictionary Collections.
Related
TLDR version
I am serializing objects into XML to match a schema provided by a third party. Their validator requires one of the child objects to have a namespace explicitly declared which matches it's ancestor's namespace . The data is complex enough that I don't want to roll my own serializer for this purpose. How can I force the XMLSerializer class to explicitly render a namespace even though it is technically redundant?
Full version
I am running into an issue where the CoreItemsMkt namespace is not rendered by the XMLSerializer. I believe that this is because both the attribute and the namespaces exactly match the ancestor's namespace that it is inheriting from, therefore the serializer omits it - however, the site validator that this file gets submitted to requires it.
For example:
<?xml version="1.0" encoding="utf-8"?>
<FSAMarketsFeed xmlns="http://www.fsa.gov.uk/XMLSchema/FSAMarketsFeed-v1-2">
<FSAFeedHeader xmlns="http://www.fsa.gov.uk/XMLSchema/FSAFeedCommon-v1-2">
[...contents omitted, this item appears once...]
</FSAFeedHeader>
<FSAMarketsFeedMsg>
<CoreItemsMkt xmlns="http://www.fsa.gov.uk/XMLSchema/FSAMarketsFeed-v1-2"> <!--//This namespace is the issue//-->
[...contents omitted, this item appears multiple times...]
</CoreItemsMkt?
</FSAMarketsFeedMsg>
<FSAMarketsFeedMsg>
<CoreItemsMkt xmlns="http://www.fsa.gov.uk/XMLSchema/FSAMarketsFeed-v1-2"> <!--//This namespace is the issue//-->
[...contents omitted, this item appears multiple times...]
</CoreItemsMkt?
</FSAMarketsFeedMsg>
I'm serializing with a method like this:
var path = GetFilePath();
var ns = new XmlSerializerNamespaces();
ns.Add("", "http://www.fsa.gov.uk/XMLSchema/FSAMarketsFeed-v1-2");
var ser = new XmlSerializer(typeof(FSAMarketsFeed));
var settings = new XmlWriterSettings
{ Encoding = Encoding.UTF8, Indent = true, IndentChars = "\t", NamespaceHandling = NamespaceHandling.Default };
using (var writer = XmlWriter.Create(path, settings))
{
ser.Serialize(writer, GetDataToSerialize(), ns);
}
My root class is defined as:
[XmlType(AnonymousType = true)]
[XmlRoot(Namespace = "http://www.fsa.gov.uk/XMLSchema/FSAMarketsFeed-v1-2", IsNullable = false)]
public class FSAMarketsFeed
{
public FSAMarketsFeed()
{
FSAMarketsFeedMsg = new FSAMarketsFeedMsg[0];
}
[XmlElement("FSAFeedHeader", IsNullable = true, Namespace = "http://www.fsa.gov.uk/XMLSchema/FSAFeedCommon-v1-2")]
public FSAFeedHeader FeedHeader { get; set; }
[XmlElement("FSAMarketsFeedMsg")]
public FSAMarketsFeedMsg[] FSAMarketsFeedMsg { get; set; }
}
The working feed header class:
[XmlType(AnonymousType = true)]
public class FSAFeedHeader
{
[XmlElement("FeedTargetSchemaVersion", IsNullable = true)]
public string FeedTargetSchemaVersion { get; set; }
[XmlElement("Submitter", IsNullable = true)]
public Submitter Submit { get; set; }
[XmlElement("ReportDetails", IsNullable = true)]
public ReportDetails ReportDetail { get; set; }
}
The parent Feed Message Class:
[XmlType(AnonymousType = true)]
public class FSAMarketsFeedMsg
{
[XmlElement("CoreItemsMkt", IsNullable = true, Namespace = "http://www.fsa.gov.uk/XMLSchema/FSAMarketsFeed-v1-2")]
public CoreItemsMkt CoreMarket { get; set; }
[XmlElement("Transaction", IsNullable = true)]
public Transaction Trans { get; set; }
}
Finally, the CoreItemsMkt class which is failing to render its namespace:
[XmlType(Namespace = "http://www.fsa.gov.uk/XMLSchema/FSAMarketsFeed-v1-2", AnonymousType = true)]
public class CoreItemsMkt
{
//[... Children omitted ...]]
}
Tried so far:
Using XMmlType(AnonymousType = true) to try to break the inheritance chain
Explicitly setting xmlns as an XmlAttributeAttribute w/ a hard coded value.
Setting and removing XmlType(Namespace = "http://www.fsa.gov.uk/XMLSchema/FSAMarketsFeed-v1-2") on CoreItemsMkt
Adding and removing XmlElement(Namespace = "the value") on the FSAMArketsFeedMsg's property.
Implementing ISerializable on CoreItmsMkt (Couldn't quite figure out how to get that to work though.)
Stack overflow searches - I've found 1 similar question that was answered with "This is unsupported, change your output namespace." Unfortunately, that answer doesn't work for me.
So, without hand rendering this, is there any way to force the XmlSerializer class to render those namespace attributes on CoreItmsMkt?
Try to use custom XML writer.
public class CustomWriter : XmlTextWriter
{
public CustomWriter(TextWriter writer) : base(writer) { }
public CustomWriter(Stream stream, Encoding encoding) : base(stream, encoding) { }
public CustomWriter(string filename, Encoding encoding) : base(filename, encoding) { }
public override void WriteStartElement(string prefix, string localName, string ns)
{
base.WriteStartElement(prefix, localName, ns);
if (localName == "CoreItemsMkt")
{
base.WriteAttributeString("xmlns",
"http://www.fsa.gov.uk/XMLSchema/FSAMarketsFeed-v1-2");
//base.WriteAttributeString("xmlns", ns);
}
}
}
The custom writer forcibly adds the required attribute to every element with the CoreItemsMkt name.
Usage
using (var customWriter = new CustomWriter(path, Encoding.UTF8))
{
customWriter.Formatting = Formatting.Indented;
customWriter.Indentation = 1;
customWriter.IndentChar = '\t';
ser.Serialize(customWriter, GetDataToSerialize(), ns);
}
You would like to be able to force XmlSerializer to emit redundant xmlns= attributes when serializing specified nested elements. Unfortunately, I don't know of any API to make this happen automatically. You also wrote The data is complex enough that I don't want to roll my own serializer for this purpose so you don't want to have to implement IXmlSerializable on FSAMarketsFeedMsg. (ISerializable is not used by XmlSerializer so implementing it will not help.) Thus you're going to want to do something "semi-manual". There are at least a couple of options for this.
Option 1: Serialize to a temporary XDocument then fix the attributes.
With this solution, you serialize to a temporary XDocument in memory, then add an XAttribute for each desired redundant xmlns=, as follows:
// Generate the temporary XDocument
var ns = Namespaces.GetFSAMarketsFeedNamespace();
var doc = data.SerializeToXDocument(null, ns);
var root = doc.Root;
// Add redundate xmlns= attributes
var name = XName.Get("CoreItemsMkt", Namespaces.FSAMarketsFeed);
var query = doc.Descendants(name); // Could be a more complex query, possibly even an XPath query.
foreach (var element in query)
{
if (!element.Attributes().Any(a => a.IsNamespaceDeclaration))
{
var prefix = element.GetPrefixOfNamespace(element.Name.Namespace);
if (string.IsNullOrEmpty(prefix))
element.Add(new XAttribute("xmlns", element.Name.NamespaceName));
else
element.Add(new XAttribute(XNamespace.Xmlns + prefix, element.Name.NamespaceName));
}
}
// Write the XDocument to disk.
Using the static extension classes:
public static class Namespaces
{
public const string FSAMarketsFeed = #"http://www.fsa.gov.uk/XMLSchema/FSAMarketsFeed-v1-2";
public const string FSAFeedCommon = #"http://www.fsa.gov.uk/XMLSchema/FSAFeedCommon-v1-2";
public static XmlSerializerNamespaces GetFSAMarketsFeedNamespace()
{
var ns = new XmlSerializerNamespaces();
ns.Add("", Namespaces.FSAMarketsFeed);
return ns;
}
}
public static class XObjectExtensions
{
public static T Deserialize<T>(this XContainer element, XmlSerializer serializer)
{
using (var reader = element.CreateReader())
{
serializer = serializer ?? new XmlSerializer(typeof(T));
object result = serializer.Deserialize(reader);
if (result is T)
return (T)result;
}
return default(T);
}
public static XDocument SerializeToXDocument<T>(this T obj, XmlSerializer serializer, XmlSerializerNamespaces ns)
{
var doc = new XDocument();
using (var writer = doc.CreateWriter())
{
serializer = serializer ?? new XmlSerializer(obj.GetType());
serializer.Serialize(writer, obj, ns);
}
return doc;
}
public static XElement SerializeToXElement<T>(this T obj, XmlSerializer serializer, XmlSerializerNamespaces ns)
{
var doc = obj.SerializeToXDocument(serializer, ns);
var element = doc.Root;
if (element != null)
element.Remove();
return element;
}
}
Which produces the XML:
<FSAMarketsFeed xmlns="http://www.fsa.gov.uk/XMLSchema/FSAMarketsFeed-v1-2">
<FSAFeedHeader xmlns="http://www.fsa.gov.uk/XMLSchema/FSAFeedCommon-v1-2">
<FeedTargetSchemaVersion>value of FeedTargetSchemaVersion</FeedTargetSchemaVersion>
</FSAFeedHeader>
<FSAMarketsFeedMsg>
<CoreItemsMkt xmlns="http://www.fsa.gov.uk/XMLSchema/FSAMarketsFeed-v1-2" />
</FSAMarketsFeedMsg>
</FSAMarketsFeed>
Option 2: Do a nested serialization of CoreMarket using [XmlAnyElement] on its containing type.
Using an [XmlAnyElement] property, a type can serialize and deserialize any arbitrary child element. You can use this functionality to do a nested serialization of CoreMarket with the necessary namespace declarations included.
To do this, modify FSAMarketsFeedMsg as follows:
[XmlType(AnonymousType = true)]
public class FSAMarketsFeedMsg
{
[XmlIgnore]
public CoreItemsMkt CoreMarket { get; set; }
[XmlAnyElement(Name = "CoreItemsMkt", Namespace = Namespaces.FSAMarketsFeed)]
[Browsable(false), EditorBrowsable(EditorBrowsableState.Never), DebuggerBrowsable(DebuggerBrowsableState.Never)]
public XElement CoreMarketXml
{
get
{
return (CoreMarket == null ? null : XObjectExtensions.SerializeToXElement(CoreMarket,
XmlSerializerFactory.Create(typeof(CoreItemsMkt), "CoreItemsMkt", Namespaces.FSAMarketsFeed),
Namespaces.GetFSAMarketsFeedNamespace()));
}
set
{
CoreMarket = (value == null ? null : XObjectExtensions.Deserialize<CoreItemsMkt>(value,
XmlSerializerFactory.Create(typeof(CoreItemsMkt), "CoreItemsMkt", Namespaces.FSAMarketsFeed)));
}
}
// Remainder of properties are left unchanged.
}
In addition to the static extension classes from Option 1, you will need the following to avoid a substantial memory leak:
public static class XmlSerializerFactory
{
static readonly Dictionary<Tuple<Type, string, string>, XmlSerializer> table;
static readonly object padlock;
static XmlSerializerFactory()
{
table = new Dictionary<Tuple<Type, string, string>, XmlSerializer>();
padlock = new object();
}
public static XmlSerializer Create(Type serializedType, string rootName, string rootNamespace)
{
if (serializedType == null)
throw new ArgumentNullException();
if (rootName == null && rootNamespace == null)
return new XmlSerializer(serializedType);
lock (padlock)
{
var key = Tuple.Create(serializedType, rootName, rootNamespace);
XmlSerializer serializer;
if (!table.TryGetValue(key, out serializer))
{
var attr = (string.IsNullOrEmpty(rootName) ? new XmlRootAttribute() { Namespace = rootNamespace } : new XmlRootAttribute(rootName) { Namespace = rootNamespace });
serializer = table[key] = new XmlSerializer(serializedType, attr);
}
return serializer;
}
}
}
Note that the [XmlAnyElement] property will be called for all unknown elements, so if your XML for some reason has unexpected elements, you may get an exception thrown from XObjectExtensions.Deserialize because the root element name is wrong. You may want to catch and ignore exceptions from this method if that is a possibility.
Serialize to disk as you are currently doing. The redundant xmlns= attributes will be present as in Option 1.
I have already tried various possibilities but maybe I am just too tired of seeing the solution -.-
I have an xml structure like this:
<diagnosisList>
<diagnosis>
<surgery1>
<date>1957-08-13</date>
<description>a</description>
<ops301>0-000</ops301>
</surgery1>
<surgery2>
<date>1957-08-13</date>
<description>a</description>
<ops301>0-000</ops301>
</surgery2>
<surgery...>
</surgery...>
</diagnosis>
</diagnosisList>
As you see there is a variable number of surgeries. I have a class "surgery" containing the XML elements.
class Surgery
{
[XmlElement("date")]
public string date { get; set; }
[XmlElement("description")]
public string description { get; set; }
[XmlElement("ops301")]
public string ops301 { get; set; }
public Surgery()
{
}
}
and a class diagnosis creating the structure by adding the surgery class to the constructor.
diagnosis.cs
class Diagnosis
{
[XmlElement("surgery")]
public Surgery surgery
{
get;
set;
}
public Diagnosis(Surgery Surgery)
{
surgery = Surgery;
}
}
I need to be able to serialize the class name of the surgery dynamically by adding a number before serialization happens.
does anybody know a way to achieve that?
any help is really appreciated :)
Kind regards
Sandro
-- EDIT
I create the whole structure starting from my root class "Import". this class then will be passed to the serializer. So I cannot use XMLWriter in the middle of creation of the structure. I Need to create the whole structure first and finally it will be serialized:
private static void XmlFileSerialization(Import import)
{
string filename = #"c:\dump\trauma.xml";
// default file serialization
XmlSerializer<Import>.SerializeToFile(import, filename);
XmlSerializerNamespaces namespaces = new XmlSerializerNamespaces();
namespaces.Add("", "");
XmlWriterSettings settings = new XmlWriterSettings();
settings.Encoding = Encoding.UTF8;
settings.Indent = true;
settings.IndentChars = "\t";
XmlSerializer<Import>.SerializeToFile(import, filename, namespaces, settings);
}
and then in the Method "SerializeToFile"
public static void SerializeToFile(T source, string filename, XmlSerializerNamespaces namespaces, XmlWriterSettings settings)
{
if (source == null)
throw new ArgumentNullException("source", "Object to serialize cannot be null");
XmlSerializer serializer = new XmlSerializer(source.GetType());
using (XmlWriter xmlWriter = XmlWriter.Create(filename, settings))
{
System.Xml.Serialization.XmlSerializer x = new System.Xml.Serialization.XmlSerializer(typeof(T));
x.Serialize(xmlWriter, source, namespaces);
}
}
}
What I Need is to be able to instantiate a variable number of classes based on the main class "Surgery". The class must have a variable Name, i.e.
surgery1, surgery2, surgery3, etc.
This cannot be changed because this is given by the Institution defining the XML structure.
the class must be accessible by its dynamic Name because the property in the class must be set.
so:
surgery1.Property = "blabla";
surgery2. Property = "babla";
etc.
I am even thinking about using T4 methods to create this part of code, but there must be another way to achieve dynamic class names.
I also thought of creating instances with variable names of the class by using reflection:
System.Reflection.Assembly.GetExecutingAssembly().CreateInstance(string className)
But this doesn't work actually -.-
Does anybody have a hint and could put me in the right direction?
I think, you could try implement methods from IXmlSerializable in object contains diagnosisList.
Try to use custom xml writer and reader.
public class SurgeryWriter : XmlTextWriter
{
public SurgeryWriter(string url) : base(url, Encoding.UTF8) { }
private int counter = 1;
public override void WriteStartElement(string prefix, string localName, string ns)
{
if (localName == "surgery")
{
base.WriteStartElement(prefix, "surgery" + counter, ns);
counter++;
}
else
base.WriteStartElement(prefix, localName, ns);
}
}
public class SurgeryReader : XmlTextReader
{
public SurgeryReader(string url) : base(url) { }
public override string LocalName
{
get
{
if (base.LocalName.StartsWith("surgery"))
return "surgery";
return base.LocalName;
}
}
}
Classes:
[XmlRoot("diagnosisList")]
public class DiagnosisList
{
[XmlArray("diagnosis")]
[XmlArrayItem("surgery")]
public Surgery[] Diagnosis { get; set; }
}
[XmlRoot("surgery")]
public class Surgery
{
[XmlElement("date", DataType = "date")]
public DateTime Date { get; set; }
[XmlElement("description")]
public string Description { get; set; }
[XmlElement("ops301")]
public string Ops301 { get; set; }
}
Use:
var xs = new XmlSerializer(typeof(DiagnosisList));
DiagnosisList diagnosisList;
using (var reader = new SurgeryReader("test.xml"))
diagnosisList = (DiagnosisList)xs.Deserialize(reader);
using (var writer = new SurgeryWriter("test2.xml"))
xs.Serialize(writer, diagnosisList);
Don't mix XML and C#.
You don't need dynamic names in the C# code!
If you need an arbitrary number of instances of a class, create them in a loop and place it in any collection.
var surgeries = new List<Surgery>();
for (int i = 0; i < 10; i++)
{
var surgery = new Surgery();
surgeries.Add(surgery);
}
Later you can access them by index or by enumerating.
surgeries[5]
foreach (var surgery in surgeries)
{
// use surgery
}
As you can see no need dynamic names!
Alternatively, use the dictionary with arbitrary names as keys.
var surgeryDict = new Dictionary<string, Surgery>();
for (int i = 0; i < 10; i++)
{
var surgery = new Surgery();
surgeryDict["surgery" + i] = surgery;
}
Access by name:
surgeryDict["surgery5"]
I want to write an xml document to disk in a compact format. To this end, I use the net framework method XmlDictionaryWriter.CreateBinaryWriter(Stream stream,IXmlDictionary dictionary)
This method writes a custom compact binary xml representation, that can later be read by XmlDictionaryWriter.CreateBinaryReader. The method accepts an XmlDictionary that can contain common strings, so that those strings do not have to be printed in the output each time. Instead of the string, the dictionary index will be printed in the file. CreateBinaryReader can later use the same dictionary to reverse the process.
However the dictionary I pass is apparently not used. Consider this code:
using System.IO;
using System.Xml;
using System.Xml.Linq;
class Program
{
public static void Main()
{
XmlDictionary dict = new XmlDictionary();
dict.Add("myLongRoot");
dict.Add("myLongAttribute");
dict.Add("myLongValue");
dict.Add("myLongChild");
dict.Add("myLongText");
XDocument xdoc = new XDocument();
xdoc.Add(new XElement("myLongRoot",
new XAttribute("myLongAttribute", "myLongValue"),
new XElement("myLongChild", "myLongText"),
new XElement("myLongChild", "myLongText"),
new XElement("myLongChild", "myLongText")
));
using (Stream stream = File.Create("binaryXml.txt"))
using (var writer = XmlDictionaryWriter.CreateBinaryWriter(stream, dict))
{
xdoc.WriteTo(writer);
}
}
}
The produced output is this (binary control characters not shown)
#
myLongRootmyLongAttribute˜myLongValue#myLongChild™
myLongText#myLongChild™
myLongText#myLongChild™
myLongText
So apparently the XmlDictionary has not been used. All strings appear in their entirety in the output, even multiple times.
This is not a problem limited to XDocument. In the above minimal example I used a XDocument to demonstrate the problem, but originally I stumbled upon this while using XmlDictionaryWriter in conjunction with a DataContractSerializer, as it is commonly used. The results were the same:
[Serializable]
public class myLongChild
{
public double myLongText = 0;
}
...
using (Stream stream = File.Create("binaryXml.txt"))
using (var writer = XmlDictionaryWriter.CreateBinaryWriter(stream, dict))
{
var dcs = new DataContractSerializer(typeof(myLongChild));
dcs.WriteObject(writer, new myLongChild());
}
The resulting output did not use my XmlDictionary.
How can I get XmlDictionaryWriter to use the suplied XmlDictionary?
Or have I misunderstood how this works?
with the DataContractSerializer approach, I tried debugging the net framework code (visual studio/options/debugging/enable net. framework source stepping). Apparently the Writer does attempt to lookup each of the above strings in the dictionary, as expected. However the lookup fails in line 356 of XmlbinaryWriter.cs, for reasons that are not clear to me.
Alternatives I have considered:
There is an overload for XmlDictionaryWriter.CreatebinaryWriter, that also accepts a XmlBinaryWriterSession. The writer then adds any new strings it encounters into the session dictionary. However, I want to only use a static dictionary for reading and writing, which is known beforehand.
I could wrap the whole thing into a GzipStream and let the compression take care of the multiple instances of strings. However, this would not compress the first instance of each string, and seems like a clumsy workaround overall.
Yes there is a misunderstanding. XmlDictionaryWriter is primarily used for serialization of objects and it is child class of XmlWriter. XDocument.WriteTo(XmlWriter something) takes XmlWriter as argument. The call XmlDictionaryWriter.CreateBinaryWriter will create an instance of System.Xml.XmlBinaryNodeWriter internally. This class has both methods for "regular" writing:
// override of XmlWriter
public override void WriteStartElement(string prefix, string localName)
{
// plain old "xml" for me please
}
and for dictionary based approach:
// override of XmlDictionaryWriter
public override void WriteStartElement(string prefix, XmlDictionaryString localName)
{
// I will use dictionary to hash element names to get shorter output
}
The later is mostly used if you serialize object via DataContractSerializer (notice its method WriteObject takes argument of both XmlDictionaryWriter and XmlWriter type), while XDocument takes just XmlWriter.
As for your problem - if I were you I'd make my own XmlWriter:
class CustomXmlWriter : XmlWriter
{
private readonly XmlDictionaryWriter _writer;
public CustomXmlWriter(XmlDictionaryWriter writer)
{
_writer = writer;
}
// override XmlWriter methods to use the dictionary-based approach instead
}
UPDATE (based on your comment)
If you indeed use DataContractSerializer you have few mistakes in your code.
1) POC classes have to be decorated with [DataContract] and [DataMember] attribute, the serialized value should be property and not field; also set namespace to empty value or you'll have to deal with namespaces in your dictionary as well. Like:
namespace XmlStuff {
[DataContract(Namespace = "")]
public class myLongChild
{
[DataMember]
public double myLongText { get; set; }
}
[DataContract(Namespace = "")]
public class myLongRoot
{
[DataMember]
public IList<myLongChild> Items { get; set; }
}
}
2) Provide instance of session as well; for null session the dictionary writer uses default (XmlWriter-like) implementation:
// order matters - add new items only at the bottom
static readonly string[] s_Terms = new string[]
{
"myLongRoot", "myLongChild", "myLongText",
"http://www.w3.org/2001/XMLSchema-instance", "Items"
};
public class CustomXmlBinaryWriterSession : XmlBinaryWriterSession
{
private bool m_Lock;
public void Lock() { m_Lock = true; }
public override bool TryAdd(XmlDictionaryString value, out int key)
{
if (m_Lock)
{
key = -1;
return false;
}
return base.TryAdd(value, out key);
}
}
static void InitializeWriter(out XmlDictionary dict, out XmlBinaryWriterSession session)
{
dict = new XmlDictionary();
var result = new CustomXmlBinaryWriterSession();
var key = 0;
foreach(var term in s_Terms)
{
result.TryAdd(dict.Add(term), out key);
}
result.Lock();
session = result;
}
static void InitializeReader(out XmlDictionary dict, out XmlBinaryReaderSession session)
{
dict = new XmlDictionary();
var result = new XmlBinaryReaderSession();
for (var i = 0; i < s_Terms.Length; i++)
{
result.Add(i, s_Terms[i]);
}
session = result;
}
static void Main(string[] args)
{
XmlDictionary dict;
XmlBinaryWriterSession session;
InitializeWriter(out dict, out session);
var root = new myLongRoot { Items = new List<myLongChild>() };
root.Items.Add(new myLongChild { myLongText = 24 });
root.Items.Add(new myLongChild { myLongText = 25 });
root.Items.Add(new myLongChild { myLongText = 27 });
byte[] buffer;
using (var stream = new MemoryStream())
{
using (var writer = XmlDictionaryWriter.CreateBinaryWriter(stream, dict, session))
{
var dcs = new DataContractSerializer(typeof(myLongRoot));
dcs.WriteObject(writer, root);
}
buffer = stream.ToArray();
}
XmlBinaryReaderSession readerSession;
InitializeReader(out dict, out readerSession);
using (var stream = new MemoryStream(buffer, false))
{
using (var reader = XmlDictionaryReader.CreateBinaryReader(stream, dict, new XmlDictionaryReaderQuotas(), readerSession))
{
var dcs = new DataContractSerializer(typeof(myLongRoot));
var rootCopy = dcs.ReadObject(reader);
}
}
}
I am trying to write a save routine for my application where several parts of the application add items to a Dictionary and then the save function writes them to a XML file. The open routine needs to read those files and re-populate the Dictionary and I can then place those objects back into my application. I am struggling with the de-serialization of the routine I have now. My save routine is as follows
XmlDocument xmlDoc = new XmlDocument();
// Write down the XML declaration
XmlDeclaration xmlDeclaration = xmlDoc.CreateXmlDeclaration("1.0", "utf-8", null);
// Create the root element
XmlElement rootNode = xmlDoc.CreateElement("TireStudy");
xmlDoc.InsertBefore(xmlDeclaration, xmlDoc.DocumentElement);
xmlDoc.AppendChild(rootNode);
foreach (var saveItem in _SaveItems)
{
XPathNavigator nav = rootNode.CreateNavigator();
using (var writer = nav.AppendChild())
{
var serializer = new XmlSerializer(saveItem.Value.GetType());
writer.WriteWhitespace("");
serializer.Serialize(writer, saveItem.Value);
writer.Close();
}
}
xmlDoc.Save(fileName);
This routine works to create a file, but I would like the key value of the dictionary to be saved in the file as well and I am not sure how to de-serialize the file this creates because I do not know the types of the objects before I read them.
Part 2 (I hate adding new parts to a question, but I don't see a better way to address the problems going forward)
I now have the following code,
var knownTypes = new List<Type>
{
typeof(ObservableCollection<string>),
typeof(ObservableCollection<Segments>),
typeof(Segments),
typeof(List<string>)
};
var serialized = _SaveItems.Serialize(knownTypes);
but I get the following exception
Type 'System.Collections.Generic.List`1[System.String]' cannot be added to list of known types since another type 'System.Collections.ObjectModel.ObservableCollection`1[System.String]' with the same data contract name 'http://schemas.microsoft.com/2003/10/Serialization/Arrays:ArrayOfstring' is already present. If there are different collections of a particular type - for example, List<Test> and Test[], they cannot both be added as known types. Consider specifying only one of these types for addition to the known types list.
If I delete either the typeof(ObservableCollection) or the typeof(List) it exceptions complaining it needs the one I deleted.
You could use DataContractSerializer as explained in this post but you may have to pass the known types as a parameter to the serializer to support nested object typed classes:
using System;
using System.Collections.Generic;
using System.Runtime.Serialization;
using System.IO;
using System.Xml;
public static class SerializationExtensions
{
public static string Serialize<T>(this T obj, IEnumerable<Type> knownTypes)
{
var serializer = new DataContractSerializer(obj.GetType(), knownTypes);
using (var writer = new StringWriter())
using (var stm = new XmlTextWriter(writer))
{
serializer.WriteObject(stm, obj);
return writer.ToString();
}
}
public static T Deserialize<T>(this string serialized, IEnumerable<Type> knownTypes)
{
var serializer = new DataContractSerializer(typeof(T), knownTypes);
using (var reader = new StringReader(serialized))
using (var stm = new XmlTextReader(reader))
{
return (T)serializer.ReadObject(stm);
}
}
}
public class Address
{
public string Country { get; set; }
public string City { get; set; }
}
public class CodedAddress
{
public int CountryCode { get; set; }
public int CityCode { get; set; }
}
class Program
{
static void Main(string[] args)
{
var persons1 = new Dictionary<string, Address>();
persons1.Add("John Smith", new Address { Country = "US", City = "New York" });
persons1.Add("Jean Martin", new Address { Country = "France", City = "Paris" });
// no need to provide known types to the serializer
var serializedPersons1 = persons1.Serialize(null);
var deserializedPersons1 = serializedPersons1.Deserialize<Dictionary<string, Address>>(null);
var persons2 = new Dictionary<string, object>();
persons2.Add("John Smith", new Address { Country = "US", City = "New York" });
persons2.Add("Jean Martin", new CodedAddress { CountryCode = 33, CityCode = 75 });
// must provide known types to the serializer
var knownTypes = new List<Type> { typeof(Address), typeof(CodedAddress) };
var serializedPersons2 = persons2.Serialize(knownTypes);
var deserializedPersons2 = serializedPersons2.Deserialize<Dictionary<string, object>>(knownTypes);
}
}
I want to be able to convert a List<T> into a specific JSON table-like format. In my case, the T will always be a simple object (no nested properties). Here are two examples to illustrate what I want.
Example #1: List<Person> to JSON
// C# list of Persons
var list = new List<Person>() {
new Person() { First = "Jesse", Last = "Gavin", Twitter = "jessegavin" },
new Person() { First = "John", Last = "Sheehan", Twitter = "johnsheehan" }
};
// I want to transform the list above into a JSON object like so
{
columns : ["First", "Last", "Twitter"],
rows: [
["Jesse", "Gavin", "jessegavin"],
["John", "Sheehan", "johnsheehan"]
]
}
Example #2: List<Address> to JSON
// C# list of Locations
var list = new List<Location>() {
new Location() { City = "Los Angeles", State = "CA", Zip = "90210" },
new Location() { City = "Saint Paul", State = "MN", Zip = "55101" },
};
// I want to transform the list above into a JSON object like so
{
columns : ["City", "State", "Zip"],
rows: [
["Los Angeles", "CA", "90210"],
["Saint Paul", "MN", "55101"]
]
}
Is there a way to tell JSON.net to serialize an object in this manner? If not, how could I accomplish this? Thanks.
UPDATE:
Thanks to #Hightechrider's answer, I was able to write some code that solves the problem.
You can view a working example here https://gist.github.com/1153155
Using reflection you can get a list of properties for the type:
var props = typeof(Person).GetProperties();
Given an instance of a Person p you can get an enumeration of the property values thus:
props.Select(prop => prop.GetValue(p, null))
Wrap those up in a generic method, add your favorite Json serialization and you have the format you want.
Assuming your using .Net 4 this should do everything you want. The class actually lets you convert to either XML or JSON. The Enum for CommunicationType is at the bottom. The serializer works best if the class your passing it has been decorated with DataContract & DataMember attributes. I've included a sample at the bottom. It will also take an anonymous type so long as it's all simple types.
Reflection would work as well but then you have to understand all the JSON nuances to output complex data types, etc. This used the built-in JSON serializer in .Net 4. One more note, because JSON does not define a date type .Net puts dates in a funky ASP.Net custom format. So long as your deserializing using the built-in deserializer it works just fine. I can dig up the documentation on that if you need.
using System;
using System.Xml.Serialization;
using System.Text;
using System.Runtime.Serialization.Json;
using System.IO;
using System.Xml.Linq;
internal class Converter
{
public static string Convert<T>(T obj, CommunicationType format, bool indent = false, bool includetype = false)
{
if (format == CommunicationType.XML)
{
return ToXML<T>(obj, includetype, indent);
}
else if (format == CommunicationType.JSON)
{
return ToJSON<T>(obj);
}
else
{
return string.Empty;
}
}
private static string ToXML<T>(T obj, bool includetype, bool indent = false)
{
if (includetype)
{
XElement xml = XMLConverter.ToXml(obj, null, includetype);
if(indent) {
return xml.ToString();
}
else
{
return xml.ToString(SaveOptions.DisableFormatting);
}
}
else
{
System.Xml.Serialization.XmlSerializerNamespaces ns = new System.Xml.Serialization.XmlSerializerNamespaces();
XmlSerializer xs = new XmlSerializer(typeof(T));
StringBuilder sbuilder = new StringBuilder();
var xmlws = new System.Xml.XmlWriterSettings() { OmitXmlDeclaration = true, Indent = indent };
ns.Add(string.Empty, string.Empty);
using (var writer = System.Xml.XmlWriter.Create(sbuilder, xmlws))
{
xs.Serialize(writer, obj, ns);
}
string result = sbuilder.ToString();
ns = null;
xs = null;
sbuilder = null;
xmlws = null;
return result;
}
}
private static string ToJSON<T>(T obj)
{
DataContractJsonSerializer ser = new DataContractJsonSerializer(typeof(T));
using (MemoryStream ms = new MemoryStream())
{
string result = string.Empty;
System.Text.UTF8Encoding encoding = new System.Text.UTF8Encoding();
ser.WriteObject(ms, obj);
result = encoding.GetString(ms.ToArray());
ms.Close();
encoding = null;
ser = null;
return result;
}
}
}
[DataContract()]
public enum CommunicationType : int
{
[XmlEnum("0"), EnumMember(Value = "0")]
XML = 0,
[XmlEnum("1"), EnumMember(Value = "1")]
JSON = 1
}
[DataContract(Namespace = "")]
public partial class AppData
{
[DataMember(Name = "ID")]
public string ID { get; set; }
[DataMember(Name = "Key")]
public string Key { get; set; }
[DataMember(Name = "Value")]
public string Value { get; set; }
[DataMember(Name = "ObjectType")]
public string ObjectType { get; set; }
}
Any specific reason why you don't need the standard format?
To actually answer the question:
Since this is something that is outside of JSON syntax I can't think of a way to implement this within the default framework.
One solution would be to leverage attributes decorate the properties you want transported over the wired with a custom attribute and using Reflection cycle through the properties and output their property names as the column headers and then cycle throw the objects and write the values. Generic enough so it could be applied across other objects as well.
public class Location
{
[JsonFooAttribute("City")]
public string city {get;set;}
[JsonFooAttribute("State")]
public string state {get;set;}
}