[PATCH wayland v2 1/4] Revert "wayland-egl-symbols-check: pass the DSO name via the build system"

[Emil Velikov]

On 9 March 2018 at 09:47, Daniel Stone <daniel at fooishbar.org> wrote:
> Hi Emil,
>
> On 8 March 2018 at 18:32, Emil Velikov <emil.l.velikov at gmail.com> wrote:
>> On 28 February 2018 at 16:38, Emil Velikov <emil.l.velikov at gmail.com> wrote:
>>> This reverts commit 85cb5ed64aa8246f4da93fc5b76dfc34096bf803.
>>>
>>> It seems like we've misread the existing code - the DSO name can be
>>> propagated via the build-system. The one available in the script was a
>>> simple fall-back.
>>
>> Humble ping on the series?
>> It should be dead trivial although do ask for details if anything is unclear.
>
> Patches 2-4 look fine and I'm happy to merge them with my review, but
> could you please explain some more about this patch? I very much like
> keeping details of the build system (specifically its internal build
> paths) in the build system itself and not in the script. I was
> assuming something in 2-4 needed this revert to be applied, but
> couldn't see anything. Is there something I'm missing?
>
There is one word to describe it - compromise:

 - above all, the internal path is a 'dummy' fallback. anyone can
provide the binary name as an argument
$ .../wayland-egl-symbols-check .../libwayland-egl.so
 - since we have a fallback - a plain .../wayland-egl-symbols-check
will work most of the time
 - handling env. variables (as opposed to arguments) is a pain with meson
 - handling arguments (as opposed to env. variable) is a pain with
current testing scheme
 - the original code is shorter

Hope you find at least some of those reasonable.

-Emil