Isn't it a good solution to exclude the method from being mapped in the metadata.xml and hand-write the code instead? It certainly doesn't look all THAT difficult in this case.
Kind regards, Willem Meints Lead Developer Professional Development Centre mobile: +31 (0)6 47 506 037 -----Oorspronkelijk bericht----- Van: monodroid-boun...@lists.ximian.com [mailto:monodroid-boun...@lists.ximian.com] Namens Jonathan Pryor Verzonden: maandag 7 januari 2013 04:34 Aan: Discussions related to Mono for Android Onderwerp: Re: [mono-android] Copying data between Dalvik and Mono... major slowdown. On Jan 3, 2013, at 4:32 PM, Jeremy A. Kolb - ARA/NED <jk...@ara.com> wrote: > Furthermore, In: > > [Register("bulkInsert", > "(Landroid/net/Uri;[Landroid/content/ContentValues;)I", "")] > public int BulkInsert(Android.Net.Uri url, ContentValues[] values) > { > IntPtr num1 = JNIEnv.NewArray<ContentValues>(values); > ... > if (values != null) > { > JNIEnv.CopyArray<ContentValues>(num1, values); > JNIEnv.DeleteLocalRef(num1); > } > return num2; > } > > Why is the array copied back in after the function call? values shouldn't > have changed. _You_ know that. The generator does not; it can change on a method-by-method basis, and the generator needs to generate code to support it. > url certainly doesn't get the same treatment. `url` isn't an array, and thus has "pass by value" semantics. For example, consider this potential Java method: public int bulkInsert(Uri url, ContentValues[] values) { url = new Url (...); values [0] = new ContentValues (...); } The caller of `bulkInsert()` will _not_ be able to see that bulkInsert() modified `url`. That is _not_ true for `values[0]`; the caller _will_ be able to see the change performed to `values[0]`, were it to be changed. - Jon _______________________________________________ Monodroid mailing list Monodroid@lists.ximian.com UNSUBSCRIBE INFORMATION: http://lists.ximian.com/mailman/listinfo/monodroid Op dit e-mailbericht is de disclaimer van Info Support van toepassing, zie http://www.infosupport.com/disclaimer _______________________________________________ Monodroid mailing list Monodroid@lists.ximian.com UNSUBSCRIBE INFORMATION: http://lists.ximian.com/mailman/listinfo/monodroid
