[PATCH] doc: Improve doxygen conversion

[Bill Spitzak]

On 01/02/2015 04:31 AM, Pekka Paalanen wrote:
> On Mon, 22 Dec 2014 14:05:08 -0800
> Bill Spitzak <spitzak at gmail.com> wrote:
>
>> PS: How does an intermediate compositor, which has to both be a server
>> and a client, use these two libraries at the same time? There appear to
>> be some name conflicts, such as two different struct wl_display objects.
>
> Those are type name conflicts only. All the duplicate-but-not-the-same
> types are opaque, and accessed only inside the correct library where it
> came from. Naturally that means that the program writer needs to
> always know if a pointer to a wl_display is for the server or client
> side and use the corresponding functions.
>
> We shouldn't have any same named public functions in both server and
> client libraries with different code, nor any public same named
> transparent types with different definitions.

Thanks, that is what I thought was happening. I am still trying to 
figure out an introduction text for the two libraries and I think this 
information should be added to the libserver one.

> The deprecated transparent server side type struct wl_buffer is the
> only conflicting one that I know of, and conflicts with an opaque client
> side type which is actually a struct wl_proxy. I don't recall having
> any others.
>
> Further conflicts may be found in the library internals, like a third
> definition of struct wl_buffer local to src/connection.c.