gdiplus: Implement GdipCreateRegionRgnData. Take 2.

[Alexandre Julliard]

Dmitry Timoshkov <dmitry at baikal.ru> writes:

> Alexandre Julliard <julliard at winehq.org> wrote:
>
>> > I hoped that this is some kind of a joke or a test, and I simply didn't
>> > get it, but looks like my hope was futile. It appears that just moving
>> > existing structure definition (yes, they all exist in current code) to
>> > the beginning of the file (so that it could be used in more places) is
>> > forbidden without any real explanation. That's too much even for a person
>> > like me with 14 years history of working on Wine, I can imaging what
>> > new-comer feels about such a reject, and there should be not wonders
>> > why he/she would go away.
>> 
>> The structs were not used at all previously, only the variables. Now
>> that they are used as structs they should have decent names.
>
> I don't recall any rule for structure naming that everyone working on Wine
> should follow. Simple grep in dlls/*.c source shows quite a bit of structures
> with underscores in their names. How is my patch different from existing ones?
> What makes an underscore in the private structure name unacceptable? Is that
> just a personal preference or something else?

A leading underscore is traditionally used to denote symbols that are
somehow private or outside of the normal namespace. Basically it's for
places where you'd want an anonymous object, but the syntax requires a
name. For instance a struct that's only used for a typedef, or a
function that's not meant to be called, or a variable that's a
placeholder but never referenced, things like that. It's a standard C
convention, and adhering to such conventions makes code easier to
understand for everybody.

>> > Alexandre, if you would silently remove those underscores if you really
>> > don't like to see them I'd just probably decided not bother to comment
>> > once I saw it in the commit, but plain rejection of the patch just because
>> > of that looks at least strange and unexplainable.
>> 
>> You also received comments from Vincent that you need to address.
>
> I think that I've answered to Vincent's concerns, besides they have nothing
> to do with the reasons this patch has been rejected, Vincent sent his comments
> much later after your e-mail. If you think that I didn't address some of
> Vincent's questions please ask for clarifications in appropriate thread.

Vincent asked you to support more cases, you refused to do that. Then he
asked you to at least add FIXMEs for the cases you don't want to
support, and you refused to do that too. I wouldn't call that addressing
his concerns.

-- 
Alexandre Julliard
julliard at winehq.org