[Mono-osx] MonoMac: patch to add coredata

Miguel de Icaza miguel at novell.com
Fri Jul 16 22:00:20 EDT 2010


Hello,

    Another note: in NSExpression you use "IntPtr" to represent objects.
This is not correct.

    When you see "id" you should substitute that for the most specific
derived type from NSObject.   In the worst case scenario you should use
NSObject, but never IntPtr.

    There is one exception, we use "IntPtr" as the return value for the
special name "Constructor" used for the various "init" methods.

Miguel.

On Wed, Jul 14, 2010 at 8:48 AM, Duane Wandless <duane at wandless.net> wrote:

> Here is the update to coredata.cs.
>
> I've also included a patch to generator.cs to handle binding (coredata uses
> (NSError**) heavily):
> //+ (NSData *)sendSynchronousRequest:(NSURLRequest *)request
> returningResponse:(NSURLResponse **)response error:(NSError **)error
> [Static, Export ("sendSynchronousRequest:returningResponse:error:")]
> NSData SynchronousRequest (NSUrlRequest request, out NSUrlResponse
> response, out NSError error);
>
> In genetator.cs:
> * added the ability to bind out parameters
>
> In foundation.cs:
> * added binding for sendSynchronousRequest on NSUrlConnection
> * added binding for NSForegroundColorAttributeName
> * exposed binding for setValue:forHTTPHeaderField: on NSMutableURLRequest
> * added binding for NSExpression
>
> In coredata.cs:
> * added strongly typed arrays
> * added CoreDataEnum.cs
>
> I have a couple more changes/issues that I'll follow up with later.
> Thanks,
> Duane
>
>
> On Sat, Jul 10, 2010 at 12:03 PM, Miguel de Icaza <miguel at novell.com>wrote:
>
>> Hello,
>>
>>     The CoreData binding is not suitable for inclusion in MonoMac since it
>> uses NSArrays instead of strongly-typed arrays.   Although some auditing
>> work was done on the result from the header parser, I do not want to add any
>> APIs that expose NSArrays (there are special cases for this, for low-level
>> interfaces, but they are special cases, not the rule).
>>
>>     When you resubmit, also remove all of the original selector comments
>> that the parser left in place.
>>
>> Miguel.
>>
>> On Fri, Jul 9, 2010 at 12:23 PM, Duane Wandless <duane at wandless.net>wrote:
>>
>>> Not all of the coredata bindings have been tested.  The basic
>>> functionality of a CoreData application does work.
>>>
>>> Duane
>>>
>>> _______________________________________________
>>> Mono-osx mailing list
>>> Mono-osx at lists.ximian.com
>>> http://lists.ximian.com/mailman/listinfo/mono-osx
>>>
>>>
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.ximian.com/pipermail/mono-osx/attachments/20100716/dc37776c/attachment.html 


More information about the Mono-osx mailing list