开发者

Getting KeyNotFoundException when using key previously retrieved from key collection?

开发者 https://www.devze.com 2022-12-17 00:46 出处:网络
I\'ve got the following code where for some reason 开发者_JAVA技巧I\'m getting a KeyNotFoundException even though I\'m using a key that I had retrived a couple of lines above.Does anyone know of a sit

I've got the following code where for some reason 开发者_JAVA技巧I'm getting a KeyNotFoundException even though I'm using a key that I had retrived a couple of lines above. Does anyone know of a situation where this wouldn't work? I'm stumped. BTW 'SchemaElementType is an enum.

public class DefaultValue
{
 private Dictionary<Parameter, string> _params;

 public DefaultValue(Dictionary<Parameter, string> parameters)
 {
        _params = parameters;
 }

  public string GetParameterValue(string name)
  {
      foreach(Parameter param in _params.Keys)
      {
           if(param.ParamName.Equals(name))
           {
               // **** Issue here  ****
               return _params[param];
           }
      }
      return string.Empty;
  }
}

[DataContract]
public class Parameter
    {
        #region Members
        private Guid _guid;
        private Guid _formulaGuid;
        private string _name;

        #endregion

        #region Constructor
        public Parameter(Guid guid, Guid formulaGuid, string name, SchemaElementType type)
        {
            ParamGuid = guid;
            FormulaGuid = formulaGuid;
            ParamName = name;
            ParamType = type;
        }

        public Parameter()
        {}

        #endregion

        #region Properties

        [DataMember]
        public Guid ParamGuid
        {
            get { return _guid; }
            set { _guid = value; }
        }

        [DataMember]
        public Guid FormulaGuid
        {
            get { return _formulaGuid; }
            set { _formulaGuid = value; }
        }

        [DataMember]
        public string ParamName
        {
            get { return _name; }
            set { _name = value; }
        }

        [DataMember]
        public SchemaElementType ParamType { get; set; }

        #endregion

        #region Overrides

        public bool Equals(Parameter other)
        {
            if (ReferenceEquals(null, other)) return false;
            if (ReferenceEquals(this, other)) return true;
            bool result =other._guid.Equals(_guid);
            result = result && other._formulaGuid.Equals(_formulaGuid);
            result = result && Equals(other._name, _name);
            result = result && Equals(other.ParamType, ParamType);

            return result;
        }

        public override int GetHashCode()
        {
            unchecked
            {
                int result = _guid.GetHashCode();
                result = (result*397) ^ _formulaGuid.GetHashCode();
                result = (result*397) ^ (_name != null ? _name.GetHashCode() : 0);
                result = (result*397) ^ ParamType.GetHashCode();
                return result;
            }
        }

        public override bool Equals(object obj)
        {
            if (ReferenceEquals(null, obj)) return false;
            if (ReferenceEquals(this, obj)) return true;
            if (obj.GetType() != typeof (Parameter)) return false;
            return Equals((Parameter) obj);
        }

        #endregion
    }


I worry about the fact that Parameter is mutable. If (after adding it to the dictionary) you have changed any of the values that are used when generating GetHashCode() (i.e. all of them) then all bets are off and you are not guaranteed to see your item again. I would not make these public setters, i.e.

    [DataMember]
    public string ParamName // applies to all the properties, not just this one
    {
        get { return _name; }
        private set { _name = value; }
    }

Actaully, I'd probably drop the explicit fields and use C# 3.0 automatically implemented properties:

    [DataMember]
    public string ParamName { get; private set; }

As an example that breaks by changing the parameter:

    var data = new Dictionary<Parameter, string>();
    Parameter p;
    data.Add((p = new Parameter(Guid.NewGuid(), Guid.NewGuid(), "abc",
        SchemaElementType.A)), "def");
    var dv = new DefaultValue(data);
    string val1 = dv.GetParameterValue("abc"); // returns "def"
    p.ParamGuid = Guid.NewGuid();
    string val2 = dv.GetParameterValue("abc"); // BOOM

As a final thought; if the typical usage is to lookup by string, then why not use the name as the key for the internal dictionary? At the moment you aren't using the dictionary properly.


Well strictly speaking you're not retrieving the (lookup key) a few lines above you are retreiving an object that at some point was used for calculating the hash key.

When you insert into a Dictionary the GetHashKey method of the key object will be called. If that changes from the time you insert the key-value pair to the time your code is executed you will get the described behaviour. (unless of cause if the GetHashKey no returns a value matching a key to a different key-value pair, in that case you get really weird behaviour not an exception)

I'd look for the value of the hash key when inserting and when retreiving and see if there's a mismatch between them


You can take advantage of the KeyValuePair<> class:

foreach(var item in _params)
{
   if(item.Key.ParamName.Equals(name))
   {
      return item.Value;
   }
}
0

精彩评论

暂无评论...
验证码 换一张
取 消