Le 18/12/2017 à 18:47, KatolaZ a écrit :
> On Mon, Dec 18, 2017 at 06:34:01PM +0100, Didier Kryn wrote:
>> Le 16/12/2017 à 11:45, KatolaZ a écrit :
>>> the Devuan ASCII sprint will continue through the week-end. Just to
>>> let you know that apulse (pulseaudio->alsa compat layer) and surf2
>>> (the suckless browser) are now in ASCII. Feel free to give them a try
>>> and report any issue.
>>     apluse is installed in its own place in the hierarchy. It is not
>> installed as a replacement for pulseaudio - otherwies it should conflict
>> with it.
>>
>>     Therefore its use is limited to software which is compiled locally, and
>> I think it cannot act as a compatibility layer for ready-made packages like
>> Skypeforlinux. Skypeforlinux has no sound when pulseaudio isn't install,
>> even when apulse is.
>>
>>     However I'm not sure Skypeforlinux is still usefull since it seems to
>> have exactly the same UI as Skypeweb.
>>
> Hi Didier,
>
> I have no pulseaudio running and skypeforlinux still works (yes, I
> need to use it, unfortunately). I don't understand what you mean
> "installed in its own place in the hierarchy". apulse is not a
> replacement to pulseaudio, just a wrapper to foul apps around to
> believe that pulseaudio is running. I don't think we can really
> replace deps to pulseaudio with deps to apulse, tbh.
>
    Dear Enzo,

    Sorry if I misunderstood, I thought to foul applications, one
needed to replace a shared library - like eudev or vdev are doing for
libudev. If (eg) Skypeforlinux tries to link to
libpulseaudio-foo-bar.so, how could it be fouled by  a file with another
name, libapulse.so? Skype is a good example because I've read that
apulse was precisely meant to foul it.

    I will further investigate because I now have two laptops running
Ascii and with apulse installed. Maybe they behave differently but I
don't understand how they could interface to apulse.

    Didier