Please do not reply to this email- if you want to comment on the bug, go to the URL shown below and enter your comments there.
Changed by [EMAIL PROTECTED] http://bugzilla.ximian.com/show_bug.cgi?id=77530 --- shadow/77530 2006-02-16 08:58:53.000000000 -0500 +++ shadow/77530.tmp.17440 2006-02-18 08:54:33.000000000 -0500 @@ -124,6 +124,52 @@ + WriteLineInd ("if (Reader.NodeType != XmlNodeType.None)"); + WriteLineUni ("Reader.ReadEndElement ();"); WriteLine ("return res;"); + +------- Additional Comments From [EMAIL PROTECTED] 2006-02-18 08:54 ------- +Lluis, + +Thanks for taking the time to review this patch. May I ask one small +favor: would it be possible to number your remarks next time, that +makes it a little easier (for me, at least ;-)). + +In EnumMap.GetXmlName: +1) * I think the call to XmlCustomFormatter.FromEnum should be done +only when the enum is a flags enum. If it is not, shouldn't it just +return the long value? + +2) The cration of the names and values arrays should only be done +once and they should be stored in the enum map. + +=> Ok, I modified my patch accordingly. I wasn't sure on whether we +should do lazy init of these arrays or not. Initializing them in the +ctor allows us to mark these array read-only, perhaps this allows +the JIT to optimize it more ? + +In SerializationCodeGenerator.GenerateGetXmlEnumValue: + +3) Same as before, the names and values arrays should not be created +for each call to the generated conversion method. + +=> Took care of that, see previous remark. + +4) Also, looks like the old line "if (val.ToString () == \"0\") +return string.Empty;" is lost? + +=> FromEnum takes care of that check, but that may waste a few +cycles so I moved it back in. + +5) This is wrong. It's going to fail if types.Count > 0. ....looks +like a better solution. + +=> You're right. I modified my patch. + +6) Which error does this fix: + +=> Deserialization of an empty element to (flags) enum (eg. +<EnumDefaultValue />). There are a few tests for this in +DeserializeTests.cs + + _______________________________________________ mono-bugs maillist - [email protected] http://lists.ximian.com/mailman/listinfo/mono-bugs
