Hello Nikolay,

thanks for your comments.

> If we really end up doing this (see test comments), it'd be better to

> use a single field to mask change notifications and label one, see

> bDoChangeNotify.

Sure, that's reasonable.

> The question is if the focus is still on edit at this point. I remember

> we already had some problems with it, because applications can easily

> depend on exact sequence of events here, e.g. LVN_ENDLABELEDIT handler

> can interact with edit box directly, or EN_* messages could be intercepted.

Yes, during LVN_ENDLABELEDIT the edit is still focused, that's why taking focus away (example: by shwoing a mesagebox) triggers LVN_ENDLABELEDIT again.

> This should use a message sequence test to avoid another global variable.

Sure, this will take care of the "exact sequence" issue too.

> Does this fix some application or you just noticed that? We should fix

> that regardless, I'm just curious.

It affects wine regedit. Just open it, got to "HKCU/Environment" and try renaming TEMP to TEMp. You'll get the error message twice, due to this bug. It's not a big issue really, but I'd like it fixed.

Regards,

Fabian Maurer