This attribute gives us a horrible debugging experience when the
'Value' of the KeyValuePair is null. By using this attribute, the
user is told that the KeyValuePair itself is null when they inspect
something like 'new KeyValuePair<string, string> (null, null)'.
This is very confusing.
We should just let the ToString method be invoked, which returns a
nicely formatted string. This is also the behaviour observed when
debugging using Visual Studio.
2009-12-06 Chris Toshok <[email protected]>
* KeyValuePair.cs: we need setters for Key and Value so they can
be set via reflection from
System.ServiceModel.System.Runtime.Serialization.Json.TypeMap.cs.
in class/System.ServiceModel.Web/System.Runtime.Serialization.Json/:
2009-12-06 Chris Toshok <[email protected]>
* JsonSerializationReader.cs (DeserializeGenericCollection): this
is jb's (iirc) patch, actually. Types subclassing from
ICollection<T> need to be supported, and since it's an interface
the implementation might be explicit. So instead of groveling
around in the actual type, we need to dispatch through the
interface's method.
* TypeMap.cs (CreateDefaultTypeMap): we need to include non-public
properties, since KeyValuePair items must be able to write to Key
and Value properties.
* JsonReader.cs (ReadContent): for builtin values (null, true,
false), use TryReadString instead of individual ReadChar's, since
they aren't reversible if the string didn't match. In the default
case, if we're LameSilverlightLiseralParser, we need to push back
the current character onto the stream -- this is because for
netflix's AppConfig.json we were parsing "frameRatePolicy" as
"rameRatePolicy" (the 'f' was triggering the "false" parsing).
(class PushBackReader): add this class to unify all the pushing
back of characters we need to do. 1 character (JsonReader's old
"saved_char") isn't enough given that you could have a property
named, e.g., "falsifiable", and we'd need 4 characters worth of
pushback to recognize that correctly.
svn path=/trunk/mcs/; revision=147766